From lionel.orry at gmail.com  Wed Mar  1 02:28:04 2006
From: lionel.orry at gmail.com (Lionel ORRY)
Date: Wed, 1 Mar 2006 08:28:04 +0100
Subject: [Nitro] Is nitrohq.com/rdoc updating possible ?
Message-ID: <c36174660602282328o3d5b3569y@mail.gmail.com>

  Hello everybody,

I am quite new here, so please be patient and accept my stupid questions for
a while ;-)

I already know that there are some issues with the current wiki at
nitrohq.com (Page modifications not taken into account, etc.) and I would
just like to know whether there is a possibility to make the main rdoc work.
The rdoc hosted in Finland works fine, thanks a lot for this, but Nitro
development is quite active and the API structure evolves each day. so an
rdoc against glycerin would be a great feature on the website.

Enough complaining, I also wanted to congratulate all the hard work done
here. Keep it up, it is very very good.

Lionel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060301/6c0c98a4/attachment.html 

From george.moschovitis at gmail.com  Wed Mar  1 04:10:14 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 11:10:14 +0200
Subject: [Nitro] Is nitrohq.com/rdoc updating possible ?
In-Reply-To: <c36174660602282328o3d5b3569y@mail.gmail.com>
References: <c36174660602282328o3d5b3569y@mail.gmail.com>
Message-ID: <afd80b540603010110p61b3e627gd217f2ebc3654df7@mail.gmail.com>

Welcome Lionel,

he suggestion about RDOC against glycerin is very good. Hopefully a
member of our community can step up and make this happen.

-g.


On 3/1/06, Lionel ORRY <lionel.orry at gmail.com> wrote:
>   Hello everybody,
>
>  I am quite new here, so please be patient and accept my stupid questions
> for a while ;-)
>
>  I already know that there are some issues with the current wiki at
> nitrohq.com (Page modifications not taken into account, etc.) and I would
> just like to know whether there is a possibility to make the main rdoc work.
> The rdoc hosted in Finland works fine, thanks a lot for this, but Nitro
> development is quite active and the API structure evolves each day. so an
> rdoc against glycerin would be a great feature on the website.
>
>  Enough complaining, I also wanted to congratulate all the hard work done
> here. Keep it up, it is very very good.
>
>  Lionel
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar  1 04:15:05 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 11:15:05 +0200
Subject: [Nitro] [PATCH] hierarchical.rb bugfix
In-Reply-To: <bd93db0d0602281211q7c93ab17q60896aac0cbb903@mail.gmail.com>
References: <op.s5ouo4y2g3glhz@jo.local>
	<bd93db0d0602281211q7c93ab17q60896aac0cbb903@mail.gmail.com>
Message-ID: <afd80b540603010115r6250c475j791ab58a6bbaca5f@mail.gmail.com>

I will *try* to get someone here at navel to setup ssh access to the darcs repo.
stay tunned.

-g.


On 2/28/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 2/28/06, Kashia Buch <kashia at vfemail.net> wrote:
> > Hi,
> >
> > 2-line fix for hierarchical.rb
>
>  Ah! That one was a tough one for me. Good work!
>
>  Attached quiets the output from the test..
>
> > And a Question:
> >
> > Is there a way, to set up a darcs repository, public or semi-public where
> one can send patches to.
> >
> > My idea:
> >   * make the darcs available to a group of people, with upload access for
> them
> >   * make the darcs run unittest, if they fail, reject patch
> >   * apply correct patches
> >   * make a website with list of applied/rejected patches (Trac ?)
> >
> > It would take:
> >   * A public accessable computer
> >   * Time to secure the computer
> >
> > which is pretty hard to come by I think...
> >
> > And it would require that all unittests are working on that computer...
> >
> > Of course this is only a sketchy idea, but one which would really help
> with such one-line fixes like this one.  (If my idea is technically possible
> :P)
>
>  For what it's worth, I think these are all good ideas. As it is,
> development tends to stall when the repo caretaker (George, zimba) is
> unavailable. Darcs seems to support emailing patches into a repo. Perhaps
> that could be set up and a few people given the email address? If that is
> difficult, perhaps ssh and sftp access could be allowed? More committers
> would mean less pressure on any one individual.
>
>  I like the review process though of emailing patches to the list. Things
> like your patch, because they fix a failing test, should get applied
> quickly. But bigger patches are able to get a better review.
>  Hopefully more people will join in this thread. It's good to talk about
> things like this, especially if it improves things. :)
>
>  Bryan
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar  1 04:17:17 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 11:17:17 +0200
Subject: [Nitro] PATCH: adds a test case for Rob's double slashes in URL
	patch.
In-Reply-To: <bd93db0d0602281318m4e69107fhd493d582c57771a4@mail.gmail.com>
References: <bd93db0d0602281318m4e69107fhd493d582c57771a4@mail.gmail.com>
Message-ID: <afd80b540603010117g4dbfa8e4i94a9e9cdbf26abf7@mail.gmail.com>

Bryan,

can you please  explain the new testing enhancements?

thanks in advance!

-g.

On 2/28/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Also a slight example for Rob's new testing enhancements.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar  1 05:25:55 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 12:25:55 +0200
Subject: [Nitro] Zimba's Control patch
Message-ID: <afd80b540603010225k55dfd1dap75b85a593f6f23ff@mail.gmail.com>

Dear devs,

I am wondering if someone could provide me some info about which
particular problems zimba's patch solves and how people are using this
in practice. Some example source code of usage patterns would help
here.

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From zimba.tm at gmail.com  Wed Mar  1 07:08:46 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Wed, 1 Mar 2006 13:08:46 +0100
Subject: [Nitro] I'm back too (+ Trac installed)
Message-ID: <200603011308.46623.zimba.tm@gmail.com>

Hi list,

I'm back from Amsterdam. It was a nice trip :-)

For my return, I have a little present for you :

http://devlab.oree.ch/

Features :
 * trac+darcs
 * ssh access (for core devs only)

There is still a lot of things to do like make a better html presentation, 
change the trac template to integrate a nitro look, ...

I will take in charge every ticket that is listed in the "trac" component.

The old repo at http://oree.ch/nitro is now located at 
http://devlab.oree.ch/darcs/nitrohq

Oh and as usual, let me know what you think :-)

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From aidan at yoyo.org  Wed Mar  1 07:52:13 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Wed, 1 Mar 2006 23:52:13 +1100
Subject: [Nitro] I'm back too (+ Trac installed)
In-Reply-To: <200603011308.46623.zimba.tm@gmail.com>
References: <200603011308.46623.zimba.tm@gmail.com>
Message-ID: <9BF7279B-EBF6-45E0-8232-F1347FE4AFA0@yoyo.org>

> For my return, I have a little present for you :
>
> http://devlab.oree.ch/
>
> Features :
>  * trac+darcs
>  * ssh access (for core devs only)

In which case, I won't bother publicising the one I created,  
presuming that devlab.oree.ch can stay around?

Aidan



From george.moschovitis at gmail.com  Wed Mar  1 08:35:57 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 15:35:57 +0200
Subject: [Nitro] I'm back too (+ Trac installed)
In-Reply-To: <9BF7279B-EBF6-45E0-8232-F1347FE4AFA0@yoyo.org>
References: <200603011308.46623.zimba.tm@gmail.com>
	<9BF7279B-EBF6-45E0-8232-F1347FE4AFA0@yoyo.org>
Message-ID: <afd80b540603010535v790895a6rbbeaa654a072d3e1@mail.gmail.com>

> In which case, I won't bother publicising the one I created,
> presuming that devlab.oree.ch can stay around?

Aidan, I am very sorry for this :( Many thanks for your efforts. I am 
sure you can help zimba with his new server.

best regards,
George.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From m.fellinger at gmail.com  Wed Mar  1 09:30:13 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Wed, 1 Mar 2006 23:30:13 +0900
Subject: [Nitro] Zimba's Control patch
In-Reply-To: <afd80b540603010225k55dfd1dap75b85a593f6f23ff@mail.gmail.com>
References: <afd80b540603010225k55dfd1dap75b85a593f6f23ff@mail.gmail.com>
Message-ID: <9c00d3e00603010630r2ef30941n29574b45b053f53a@mail.gmail.com>

most of the features were outlined by a request from Riku R?is?nen.http://rubyforge.org/pipermail/nitro-general/2006-February/002719.htmlwich has some example-usages as well and this is basically how i tryto use it, Bryan Soto made the actual implementation, so he should beable to tell us a bit more about it :)
not sure if that helps and if it was possible with the Element, but iuse Control as follows:
### skin or wherever ###class Foo < Nitro::Control  def render    %~      \#{@bar}    ~  endend
### in the template ###<Foo bar="#{something}" />
(hope the code was not too messed up)~~~~manveru
On 3/1/06, George Moschovitis <george.moschovitis at gmail.com> wrote:> Dear devs,>> I am wondering if someone could provide me some info about which> particular problems zimba's patch solves and how people are using this> in practice. Some example source code of usage patterns would help> here.>> regards,> George.>>> --> http://www.gmosx.com> http://www.navel.gr> http://www.nitrohq.com>> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From mneumann at ntecs.de  Wed Mar  1 10:31:17 2006
From: mneumann at ntecs.de (Michael Neumann)
Date: Wed, 1 Mar 2006 16:31:17 +0100
Subject: [Nitro] Controls
Message-ID: <EE68A7E6-76C8-4C02-A5A8-EEC7783A89DD@ntecs.de>

Hi,

I'm curious about how Nitro::Controls are used and how useful they  
are. Can anyone give a good example?
Thanks!

Regards,

   Michael


From george.moschovitis at gmail.com  Wed Mar  1 10:36:02 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 17:36:02 +0200
Subject: [Nitro] Controls
In-Reply-To: <EE68A7E6-76C8-4C02-A5A8-EEC7783A89DD@ntecs.de>
References: <EE68A7E6-76C8-4C02-A5A8-EEC7783A89DD@ntecs.de>
Message-ID: <afd80b540603010736u1280f3efvfa2929c7f881a08e@mail.gmail.com>

Michael,

see the answer to my own question about controls.

-g.

On 3/1/06, Michael Neumann <mneumann at ntecs.de> wrote:
> Hi,
>
> I'm curious about how Nitro::Controls are used and how useful they
> are. Can anyone give a good example?
> Thanks!
>
> Regards,
>
>    Michael
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From guillaume.pierronnet at gmail.com  Wed Mar  1 11:16:53 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Wed, 1 Mar 2006 17:16:53 +0100
Subject: [Nitro] [PATCH] support for memcached and other fixes
Message-ID: <6a7d49ca0603010816q168a2fb4y@mail.gmail.com>

hi Nitro devs,

here is a bundle with following patches:

  * cache improvements

  * added --address and --port options to force server to run on those settings

  * little Glue::Cache improvements

  * Glue::MemCached implemented (and test case updated)

  * Nitro session memcached support (and session test case improved)

  * little fixes on session handlers

bye!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.gz
Type: application/x-gzip
Size: 7003 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060301/a0f0c0d2/attachment.gz 

From george.moschovitis at gmail.com  Wed Mar  1 11:26:41 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 18:26:41 +0200
Subject: [Nitro] [PATCH] support for memcached and other fixes
In-Reply-To: <6a7d49ca0603010816q168a2fb4y@mail.gmail.com>
References: <6a7d49ca0603010816q168a2fb4y@mail.gmail.com>
Message-ID: <afd80b540603010826j660edac8j2a1937f15cc3b705@mail.gmail.com>

thanks a lot ;-)

On 3/1/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> hi Nitro devs,
>
> here is a bundle with following patches:
>
>   * cache improvements
>
>   * added --address and --port options to force server to run on those settings
>
>   * little Glue::Cache improvements
>
>   * Glue::MemCached implemented (and test case updated)
>
>   * Nitro session memcached support (and session test case improved)
>
>   * little fixes on session handlers
>
> bye!
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar  1 11:50:14 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 18:50:14 +0200
Subject: [Nitro] [PATCH] support for memcached and other fixes
In-Reply-To: <6a7d49ca0603010816q168a2fb4y@mail.gmail.com>
References: <6a7d49ca0603010816q168a2fb4y@mail.gmail.com>
Message-ID: <afd80b540603010850u7981d1d2v20d4d31b07984016@mail.gmail.com>

are you sure this is the correct bundle file?

-g.

On 3/1/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> hi Nitro devs,
>
> here is a bundle with following patches:
>
>   * cache improvements
>
>   * added --address and --port options to force server to run on those settings
>
>   * little Glue::Cache improvements
>
>   * Glue::MemCached implemented (and test case updated)
>
>   * Nitro session memcached support (and session test case improved)
>
>   * little fixes on session handlers
>
> bye!
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar  1 11:52:00 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 1 Mar 2006 18:52:00 +0200
Subject: [Nitro] [PATCH] support for memcached and other fixes
In-Reply-To: <afd80b540603010850u7981d1d2v20d4d31b07984016@mail.gmail.com>
References: <6a7d49ca0603010816q168a2fb4y@mail.gmail.com>
	<afd80b540603010850u7981d1d2v20d4d31b07984016@mail.gmail.com>
Message-ID: <afd80b540603010852q2b11169dld491ae394aca86cd@mail.gmail.com>

this seems like an older patch bundle...

-g.

On 3/1/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> are you sure this is the correct bundle file?
>
> -g.
>
> On 3/1/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > hi Nitro devs,
> >
> > here is a bundle with following patches:
> >
> >   * cache improvements
> >
> >   * added --address and --port options to force server to run on those settings
> >
> >   * little Glue::Cache improvements
> >
> >   * Glue::MemCached implemented (and test case updated)
> >
> >   * Nitro session memcached support (and session test case improved)
> >
> >   * little fixes on session handlers
> >
> > bye!
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
> >
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From guillaume.pierronnet at gmail.com  Wed Mar  1 11:58:20 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Wed, 1 Mar 2006 17:58:20 +0100
Subject: [Nitro] [PATCH] support for memcached and other fixes
In-Reply-To: <afd80b540603010852q2b11169dld491ae394aca86cd@mail.gmail.com>
References: <6a7d49ca0603010816q168a2fb4y@mail.gmail.com>
	<afd80b540603010850u7981d1d2v20d4d31b07984016@mail.gmail.com>
	<afd80b540603010852q2b11169dld491ae394aca86cd@mail.gmail.com>
Message-ID: <6a7d49ca0603010858j763b2442h@mail.gmail.com>

yep, i'm drunk maybe... :-)


2006/3/1, George Moschovitis <george.moschovitis at gmail.com>:
> this seems like an older patch bundle...
>
> -g.
>
> On 3/1/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > are you sure this is the correct bundle file?
> >
> > -g.
> >
> > On 3/1/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > > hi Nitro devs,
> > >
> > > here is a bundle with following patches:
> > >
> > >   * cache improvements
> > >
> > >   * added --address and --port options to force server to run on those settings
> > >
> > >   * little Glue::Cache improvements
> > >
> > >   * Glue::MemCached implemented (and test case updated)
> > >
> > >   * Nitro session memcached support (and session test case improved)
> > >
> > >   * little fixes on session handlers
> > >
> > > bye!
> > >
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> > >
> > >
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.gz
Type: application/x-gzip
Size: 10386 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060301/93d3fc62/attachment.gz 

From reid.thompson at ateb.com  Wed Mar  1 13:04:29 2006
From: reid.thompson at ateb.com (Reid Thompson)
Date: Wed, 01 Mar 2006 13:04:29 -0500
Subject: [Nitro] I'm back too (+ Trac installed)
In-Reply-To: <200603011308.46623.zimba.tm@gmail.com>
References: <200603011308.46623.zimba.tm@gmail.com>
Message-ID: <4405E22D.5020703@ateb.com>

zimba.tm wrote:
> Hi list,
>
> I'm back from Amsterdam. It was a nice trip :-)
>
> For my return, I have a little present for you :
>
> http://devlab.oree.ch/
>
> Features :
>  * trac+darcs
>  * ssh access (for core devs only)
>
> There is still a lot of things to do like make a better html presentation, 
> change the trac template to integrate a nitro look, ...
>
> I will take in charge every ticket that is listed in the "trac" component.
>
> The old repo at http://oree.ch/nitro is now located at 
> http://devlab.oree.ch/darcs/nitrohq
>
> Oh and as usual, let me know what you think :-)
>
>   
>From http://devlab.oree.ch/trac/nitrohq/wiki/Repositories ....

http://devlab.oree.ch/darcs/nitrohq/ => 403 Forbidden


http://devlab.oree.ch/darcs/nitrohq =>


    Index of /

Name 	Last Modified 	Size 	Type
Parent Directory <http://repo.nitrohq.com/>/ 	  	-   	Directory
_darcs <http://repo.nitrohq.com/_darcs/>/ 	2006-Mar-01 19:10:28 	-   
Directory
examples <http://repo.nitrohq.com/examples/>/ 	2006-Feb-05 11:54:16 	- 
  	Directory
flare <http://repo.nitrohq.com/flare/>/ 	2006-Feb-05 11:54:16 	-   
Directory
gen <http://repo.nitrohq.com/gen/>/ 	2006-Feb-05 11:54:16 	-   	Directory
glue <http://repo.nitrohq.com/glue/>/ 	2006-Feb-27 16:27:47 	-   	Directory
nitro <http://repo.nitrohq.com/nitro/>/ 	2006-Feb-06 18:58:23 	-   
Directory
og <http://repo.nitrohq.com/og/>/ 	2006-Feb-06 11:03:49 	-   	Directory
spark <http://repo.nitrohq.com/spark/>/ 	2006-Feb-27 16:27:47 	-   
Directory
apply.rb <http://repo.nitrohq.com/apply.rb> 	2005-Nov-17 18:32:06 
0.1K 	text/html
build.rb <http://repo.nitrohq.com/build.rb> 	2006-Feb-05 11:54:16 
0.2K 	text/html
bundle <http://repo.nitrohq.com/bundle> 	2006-Mar-01 19:09:51 	35.2K 
text/html
cleanup.rb <http://repo.nitrohq.com/cleanup.rb> 	2006-Jan-12 
15:47:58 	0.8K 	text/html
ctabs.rb <http://repo.nitrohq.com/ctabs.rb> 	2005-Nov-17 18:32:06 
0.4K 	text/html
glycerin.rb <http://repo.nitrohq.com/glycerin.rb> 	2006-Feb-28 
11:52:42 	0.3K 	text/html
pushdocs.rb <http://repo.nitrohq.com/pushdocs.rb> 	2005-Nov-17 
18:32:06 	0.4K 	text/html
release.rb <http://repo.nitrohq.com/release.rb> 	2005-Nov-17 
18:32:06 	0.7K 	text/html

lighttpd/1.4.6



From zimba.tm at gmail.com  Wed Mar  1 13:28:05 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Wed, 1 Mar 2006 19:28:05 +0100
Subject: [Nitro] I'm back too (+ Trac installed)
In-Reply-To: <4405E22D.5020703@ateb.com>
References: <200603011308.46623.zimba.tm@gmail.com> <4405E22D.5020703@ateb.com>
Message-ID: <200603011928.05548.zimba.tm@gmail.com>

On Wednesday 01 March 2006 19:04, Reid Thompson wrote:
> http://devlab.oree.ch/darcs/nitrohq/ => 403 Forbidden

Hi Reid,

apparently some people got caught by this message. I didn't activated dir 
listing or used a darcs web frontend. Now I have one but it's ugly. But if 
you use "darcs get http://devlab.oree.ch/darcs/nitrohq/" then it works 
perfectly because darcs looks at specific files and don't need dir listing.

If you want to have a nice view of the darcs repo, use the trac's integrated 
repository browser at http://devlab.oree.ch/trac/nitrohq/browser

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From bryan.a.soto at gmail.com  Wed Mar  1 19:09:46 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 1 Mar 2006 16:09:46 -0800
Subject: [Nitro] So what does it all mean.
Message-ID: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>

In the span of 12 hours we have two separate offers to host a repo and
Trac. Many thanks to both Aidan and zimba! It definitely put a smile
on my face. :)

But what does this mean for Nitro/Og? I note that George's
announcement of the core development team* back in January somewhat
covers this situation but not really as George is here.

Will it be the host for a real glycerin (basically a test bed for any
and all new ideas) with the best new features, bugfixes and RDoc being
incorporated into the official repo?

Are we going to be officially moving off of repo.nitrohq.com and
centering all development at devlab.oree.ch to take advantage of the
Trac/Darcs integration?  Who else would get commit privs, if any?

Does the core dev team remain the same? If not, how do we handle
changes? Elections? Community nominations and selections by George?

Any other thoughts on how to take advantage of the situation before us?

I guess, ultimately, I wonder if this changes anything and, if so, how?


Bryan

* http://rubyforge.org/pipermail/nitro-general/2006-January/002630.html



From epiperak at gmail.com  Wed Mar  1 20:56:26 2006
From: epiperak at gmail.com (Emmanouil Piperakis)
Date: Thu, 2 Mar 2006 10:56:26 +0900
Subject: [Nitro] So what does it all mean.
In-Reply-To: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
Message-ID: <dea8525a0603011756u7e0a6906r3657db31134566b9@mail.gmail.com>

I love elections!

On 3/2/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
>
>
> Does the core dev team remain the same? If not, how do we handle
> changes? Elections? Community nominations and selections by George?
>
> Any other thoughts on how to take advantage of the situation before us?
>

I believe that the necessery thing at the moment is organization and
prioritization. George should delegate some "power of decission" to the core
developers like Bryan and the rest in order for things to move on faster.
Even though he is the heart of Nitro/Og he is only one man! (or not? have
you always been here?) The more the better... I place my vote on that...

Also, I would definately like to add / remind everyone who is helping
Nitro/Og advance that we should CONCENTRATE on the DOCS! People...!!! If we
had the docs out there Nitro/Og would be very popular right now...

:-)

Emmanouil (aka EerieShadow)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060301/382d2533/attachment.html 

From porges at porg.es  Wed Mar  1 21:33:06 2006
From: porges at porg.es (George Pollard)
Date: Thu, 02 Mar 2006 15:33:06 +1300
Subject: [Nitro] So what does it all mean.
In-Reply-To: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
Message-ID: <1141266786.8498.2.camel@localhost>

On Wed, 2006-03-01 at 16:09 -0800, Bryan Soto wrote:
> In the span of 12 hours we have two separate offers to host a repo and
> Trac.

Might I also suggest having a look at Launchpad? It?s what Ubuntu are
using for their translation/bugs/etc, and it?s all in one package.

I?ve set up a placeholder at http://launchpad.net/products/nitro/ so you
can have a look and see what you think of their system.

The services that Launchpad provides are:
- Translation system (Rosetta)
- Bug reporting/tracking (Malone)
- Support tickets
- Bounties
- Meeting manager
- Specs/milestones
- Series/branch tracking

:) George
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 191 bytes
Desc: This is a digitally signed message part
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060301/c9c03273/attachment.bin 

From bakki.kudva at gmail.com  Wed Mar  1 21:33:26 2006
From: bakki.kudva at gmail.com (Bakki Kudva)
Date: Wed, 1 Mar 2006 21:33:26 -0500
Subject: [Nitro] So what does it all mean.
In-Reply-To: <dea8525a0603011756u7e0a6906r3657db31134566b9@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<dea8525a0603011756u7e0a6906r3657db31134566b9@mail.gmail.com>
Message-ID: <f7f43c920603011833p2b856bfaqe8441b49f7413bce@mail.gmail.com>

I'd like to throw my name in the hat for assisting with the Docs.
I am trying to grok the source, various READMEs and trying to
construct an image of the overall architecture in my head. I noticed
some names among authors ( citizen428, vikingtux)who seem interested
in Docs. So Doc authors please let me know, off the list if you wish,
how I can help.

I like really tightly written docs like the Dennis Richtie's book on
C. These days there's so much fluff in programming books which I find
very distracting.

-bakki



On 3/1/06, Emmanouil Piperakis <epiperak at gmail.com> wrote:
> Also, I would definately like to add / remind everyone who is helping
> Nitro/Og advance that we should CONCENTRATE on the DOCS! People...!!! If we
> had the docs out there Nitro/Og would be very popular right now...



From aidan at yoyo.org  Wed Mar  1 21:44:36 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Thu, 2 Mar 2006 13:44:36 +1100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
Message-ID: <F5B869C9-F295-4AD2-A19B-6BDA5B3AE2BD@yoyo.org>

My 2 cents worth.

I think different people should be given charge over different  
things.  E.g. Bryan posts a lot of Og related code, so perhaps he  
should be the lead for that? (I know there are other candidates)   
Anyone who wants to should be able to nominate themselves or someone  
else to look after an untended section of code, and anyone who looks  
after that section should then pass on the torch when they no longer  
have the time.

Give these "champions" commit privileges to the whole code base, and  
let them decide what gets included and what doesn't.  In this way, we  
get people who are enthusiastic about particular areas contributing  
and controlling contributions to those areas.  They'll feel a real  
sense of ownership, whilst having the community and the other  
champion peers to ensure they don't do anything too crazy :-)

How do we decide who these people are?  I'm happy to have George be  
arbitrary about it - let's see who raises their hands.  I think  
elections at this point would be meaningless - there's only about 20  
people who would actually have an informed opinion of who to vote  
for, and we'd probably all agree with George's decisions anyway.

Zimba was nice enough to contact me off list and see if I wanted to  
help administer trac etc.  I think it would be useful to have the two  
of us (plus any other volunteers, but I know that both of us have  
actually done work toward this goal) maintain a strong presence for  
Nitro on the web, through ensuring that all issues get looked at,  
maintaining the wiki, creating discussion forums, etc.  (Both of us  
feel that NitroHQ's instability has been part of the hindrance in  
promoting Nitro to the world.)

Let's get started with putting bugs, feature requests (including doc  
requests!) and documentation into trac and then just allow people to  
grab issues and work on them.  The organisation will fall out as we go.

Aidan
---
http://www.infurious.com




From aidan at yoyo.org  Wed Mar  1 21:45:38 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Thu, 2 Mar 2006 13:45:38 +1100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <1141266786.8498.2.camel@localhost>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<1141266786.8498.2.camel@localhost>
Message-ID: <422016BE-E667-43E5-B954-86A70392DFE2@yoyo.org>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi George - that's 3 systems in 24 hours :-)

I think we decided to go with Trac because it allow integration with  
our chosen version control system (darcs) and it is hosted by someone  
in the community, giving us greater control/flexibility in features.   
I'd give Trac +1 over a hosted set-up like Launchpad (or indeed,  
Rubyforge).

Aidan

- ---
http://www.infurious.com


On 02/03/2006, at 1:33 PM, George Pollard wrote:

> On Wed, 2006-03-01 at 16:09 -0800, Bryan Soto wrote:
>> In the span of 12 hours we have two separate offers to host a repo  
>> and
>> Trac.
>
> Might I also suggest having a look at Launchpad? It?s what Ubuntu are
> using for their translation/bugs/etc, and it?s all in one package.
>
> I?ve set up a placeholder at http://launchpad.net/products/nitro/  
> so you
> can have a look and see what you think of their system.
>
> The services that Launchpad provides are:
> - Translation system (Rosetta)
> - Bug reporting/tracking (Malone)
> - Support tickets
> - Bounties
> - Meeting manager
> - Specs/milestones
> - Series/branch tracking
>
> :) George
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (Darwin)

iD8DBQFEBlxSIa4FeDc4plwRArWoAKCInaTA3LEtE/G2ATIwTpZjteXfcACgmysr
hJKU5DVkSDSzvPDScptZiuw=
=U20T
-----END PGP SIGNATURE-----



From rf at bicou.com  Thu Mar  2 03:15:39 2006
From: rf at bicou.com (Bicou)
Date: Thu, 2 Mar 2006 09:15:39 +0100
Subject: [Nitro] Og : mysql store mysqladmin
Message-ID: <1284195695.20060302091539@bicou.com>

Hello,

a  little  bug?  in  lib/og/store/mysql.rb  with mysqladmin, forgot to
include --host param, patch attached.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mysql.rb-mysqladmin-host.diff
Type: application/octet-stream
Size: 661 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060302/3999d687/attachment.obj 

From rf at bicou.com  Thu Mar  2 03:16:17 2006
From: rf at bicou.com (bicou)
Date: Thu, 2 Mar 2006 09:16:17 +0100
Subject: [Nitro] Og : Ez clause with arrays
Message-ID: <372852631.20060302091617@bicou.com>

Found a bug with the ez === operator, something like :

topics = Topic.find { |t|
       t.name === ['NAME1','NAME2','NAME3']
}

would be written as :

SELECT [...] WHERE name IN (---
- NAME1
- NAME2
... (yaml is escaping the array)


I've made a patch with a new prepared statement like operator :

WHERE name IN (?*)

with ?* replaced with the array as a list of quoted values
instead of the yaml dump.

patch attached
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ez-where-in-array.diff
Type: application/octet-stream
Size: 1437 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060302/dc0f21d1/attachment.obj 

From nusgnaf at gmail.com  Thu Mar  2 03:45:07 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 2 Mar 2006 08:45:07 +0000
Subject: [Nitro] Best way to pass parameter to Nitro::Server?
Message-ID: <716700c90603020045u71bd4519v1ce0fb1883ee17c4@mail.gmail.com>

Hi, I am working on imporve SCGI adapters, and am planning to port
goodies like DRb control and scgi cluster from Zed's scgi-rails
package.

Right now we use Nitro::Runner to fireup the scgi server. and hard
code many default setting in SCGI.start method. What's the prefered to
way to pass parameters to adapters in Nitro?



From george.moschovitis at gmail.com  Thu Mar  2 03:53:58 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 2 Mar 2006 10:53:58 +0200
Subject: [Nitro] Og : mysql store mysqladmin
In-Reply-To: <1284195695.20060302091539@bicou.com>
References: <1284195695.20060302091539@bicou.com>
Message-ID: <afd80b540603020053t62c8b10amf29d68835f4c1604@mail.gmail.com>

I have allready included this...
-g.

On 3/2/06, Bicou <rf at bicou.com> wrote:
> Hello,
>
> a  little  bug?  in  lib/og/store/mysql.rb  with mysqladmin, forgot to
> include --host param, patch attached.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar  2 03:58:32 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 2 Mar 2006 10:58:32 +0200
Subject: [Nitro] So what does it all mean.
In-Reply-To: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
Message-ID: <afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>

> Will it be the host for a real glycerin (basically a test bed for any
> and all new ideas) with the best new features, bugfixes and RDoc being
> incorporated into the official repo?

I will like to propose the following. I will prepare a single homepage
for www.nitrohq.com. Then I will redirect trac.nitrohq.com and
repo.nitrohq.com to the trac and darcs on zimbas server respectively.
Zimba will give ssh access to the core development team members. And
we should probably start copying content from nitrohq.com to the trac
wiki.

> Does the core dev team remain the same? If not, how do we handle
> changes? Elections? Community nominations and selections by George?

Here are the core development team members I propose:

Rob Pitt
Bryan Sotto
Zimba
Aidan

What do you think?

regards,
George.



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From zimba.tm at gmail.com  Thu Mar  2 04:22:28 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 10:22:28 +0100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
Message-ID: <200603021022.29021.zimba.tm@gmail.com>

On Thursday 02 March 2006 09:58, George Moschovitis wrote:
> > Will it be the host for a real glycerin (basically a test bed for any
> > and all new ideas) with the best new features, bugfixes and RDoc being
> > incorporated into the official repo?
>
> I will like to propose the following. I will prepare a single homepage
> for www.nitrohq.com. Then I will redirect trac.nitrohq.com and
> repo.nitrohq.com to the trac and darcs on zimbas server respectively.
> Zimba will give ssh access to the core development team members. And
> we should probably start copying content from nitrohq.com to the trac
> wiki.
>
> > Does the core dev team remain the same? If not, how do we handle
> > changes? Elections? Community nominations and selections by George?
>
> Here are the core development team members I propose:
>
> Rob Pitt
> Bryan Sotto
> Zimba
> Aidan
>
> What do you think?

Great ! All proposed core developpers now have ssh access.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From george.moschovitis at gmail.com  Thu Mar  2 04:37:07 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 2 Mar 2006 11:37:07 +0200
Subject: [Nitro] So what does it all mean.
In-Reply-To: <200603021022.29021.zimba.tm@gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<200603021022.29021.zimba.tm@gmail.com>
Message-ID: <afd80b540603020137y3e321cc3p1fb9164589ea66c@mail.gmail.com>

> > Rob Pitt
> > Bryan Sotto
> > Zimba
> > Aidan
> >
> > What do you think?
>
> Great ! All proposed core developpers now have ssh access.

Perhaps I am missing someone who should be included? Any ideas?

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From zimba.tm at gmail.com  Thu Mar  2 04:50:32 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 10:50:32 +0100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <afd80b540603020137y3e321cc3p1fb9164589ea66c@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<200603021022.29021.zimba.tm@gmail.com>
	<afd80b540603020137y3e321cc3p1fb9164589ea66c@mail.gmail.com>
Message-ID: <200603021050.32379.zimba.tm@gmail.com>

On Thursday 02 March 2006 10:37, George Moschovitis wrote:
> > > Rob Pitt
> > > Bryan Sotto
> > > Zimba
> > > Aidan
> > >
> > > What do you think?
> >
> > Great ! All proposed core developpers now have ssh access.
>
> Perhaps I am missing someone who should be included? Any ideas?
>
> -g.

I don't know. Some other people have submites patches, like itsme123 and 
Michael Fellinger but not as much as Bryan and Rob

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From guillaume.pierronnet at gmail.com  Thu Mar  2 04:55:35 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Thu, 2 Mar 2006 10:55:35 +0100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <afd80b540603020137y3e321cc3p1fb9164589ea66c@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<200603021022.29021.zimba.tm@gmail.com>
	<afd80b540603020137y3e321cc3p1fb9164589ea66c@mail.gmail.com>
Message-ID: <6a7d49ca0603020155k7813dec7i@mail.gmail.com>

with such a nice bug tracking system, i'll be happy to join the team!


2006/3/2, George Moschovitis <george.moschovitis at gmail.com>:
> > > Rob Pitt
> > > Bryan Sotto
> > > Zimba
> > > Aidan
> > >
> > > What do you think?
> >
> > Great ! All proposed core developpers now have ssh access.
>
> Perhaps I am missing someone who should be included? Any ideas?
>
> -g.
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From zimba.tm at gmail.com  Thu Mar  2 05:39:00 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 11:39:00 +0100
Subject: [Nitro] [devlab] userdir, rdoc and help
Message-ID: <200603021139.00618.zimba.tm@gmail.com>

Hi list,

here are some today enhancements for devlab.

Userdir
=======

All ssh users now have their own personnal homepage on url like 
http://devlab.oree.ch/~username

Use scp to copy your files over

Please don't use this space to put heavy files. It is more intended to publish 
your own branch of the darcs repo.

I will look in the future to allow you to run your own ruby apps there for 
testing.


RDoc
====

A bundle rdoc containing all the nitro, glue, og libraries is generated every 
night at 00:00 automatically.

In the future I will try to separate the different packages like the stdlib on 
ruby-doc.org

Help
====

== I still need a nice homepage for devlab.oree.ch

I want to build a little site containing nitro and other nitro and ruby 
related projects. Make everything nice and practical. If someone has some 
graphic skills and want to build a homepage, I'm interested.

== Trac needs some customization

Does somebody know how to change the templates of trac ? It would be nice if 
somebody could give me a pointer.

== All in all

I think that we need a graphist to make everything like nice, shiny and 
integrated.


See you soon for more news. All requests and comments are welcome.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From george.moschovitis at gmail.com  Thu Mar  2 06:04:59 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 2 Mar 2006 13:04:59 +0200
Subject: [Nitro] So what does it all mean.
In-Reply-To: <6a7d49ca0603020155k7813dec7i@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<200603021022.29021.zimba.tm@gmail.com>
	<afd80b540603020137y3e321cc3p1fb9164589ea66c@mail.gmail.com>
	<6a7d49ca0603020155k7813dec7i@mail.gmail.com>
Message-ID: <afd80b540603020304n31ee7b1asdd7e02fa53a5e01d@mail.gmail.com>

Oops, Guill should *certainly* have repo access! Zimb, can you please fix this?

regards,
George.

On 3/2/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> with such a nice bug tracking system, i'll be happy to join the team!

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar  2 06:06:30 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 2 Mar 2006 13:06:30 +0200
Subject: [Nitro] [devlab] userdir, rdoc and help
In-Reply-To: <200603021139.00618.zimba.tm@gmail.com>
References: <200603021139.00618.zimba.tm@gmail.com>
Message-ID: <afd80b540603020306kd5d0613o80ee561a7ff46ea9@mail.gmail.com>

>
> A bundle rdoc containing all the nitro, glue, og libraries is generated every
> night at 00:00 automatically.

where can we find this?

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From nusgnaf at gmail.com  Thu Mar  2 06:19:37 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 2 Mar 2006 11:19:37 +0000
Subject: [Nitro] Misterious Nitro bug
Message-ID: <716700c90603020319i3896086ey42d0b0ae89536b1f@mail.gmail.com>

While refactoring SCGI I was hitten by this. I think it's Nitro related:
gen app ./test #create a nitro site directory
cd test
cat > d.rb
require 'nitro'
YAML.load("conf/scgi.yaml")

irb d.rb
d.rb(main):001:0> require 'nitro/adapter/scgi'
=> true
d.rb(main):002:0> config = YAML.load_file "conf/scgi.yaml"
=> {:password=>"6ryDR0ZuDaB6Q", :host=>"127.0.0.1",
:control_url=>"druby://127.0.0.1:8999", :mode=>"live", :maxconns=>100,
:config=>"conf/scgi.yaml", :port=>9999, :logfile=>"log/scgi.log"}

mv d.rb ..
irb ../d.rb
d.rb(main):001:0> require 'nitro'
D, [2006-03-02T18:57:57.727022 #5118] DEBUG -- : Using Memory sessions.
=> true
d.rb(main):002:0> config = YAML.load_file "conf/scgi.yaml"
Errno::ENOENT: No such file or directory - conf/scgi.yaml
        from /usr/lib/ruby/1.8/yaml.rb:143:in `initialize'
        from /usr/lib/ruby/1.8/yaml.rb:143:in `load_file'
        from ../d.rb:2

cat ../a.rb
require 'yaml'
YAML.load("conf/scgi.yaml")

irb ../a.rb
a.rb(main):001:0> require 'yaml'
=> true
a.rb(main):002:0> config = YAML.load_file "conf/scgi.yaml"
=> {:host=>"127.0.0.1", :password=>"6ryDR0ZuDaB6Q", :port=>9999,
:logfile=>"log/scgi.log", :maxconns=>100, :mode=>"live",
:config=>"conf/scgi.yaml", :control_url=>"druby://127.0.0.1:8999"}

it seems that after nitro loaded you cannot use relative file path to
the working directory unless excuted script is in the working
directory. It cause a little trouble to write command line tools. Any
ideas to fix this?



From fabian at oggu.de  Thu Mar  2 06:53:23 2006
From: fabian at oggu.de (Fabian Buch)
Date: Thu, 2 Mar 2006 12:53:23 +0100
Subject: [Nitro] Powered by Nitro button
Message-ID: <3312293ba97892d0b97a76da9170fe86@oggu.de>

There's a nice "Powered by Nitro"-Button on ruby-doc.org. Who made it? 
James? Is it allowed to use it on other nitro-powered websites?

Fabian



From zimba.tm at gmail.com  Thu Mar  2 06:58:39 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 12:58:39 +0100
Subject: [Nitro] About the new Control system
Message-ID: <200603021258.39943.zimba.tm@gmail.com>

Hi list,

like I said yesterday, today I'm introducing the new Nitro::Control system. 
This new system is intended to replace the old Nitro::Form::Control system.

But first I have to talk about Nitro::Action which Control depends on.

== What is Nitro::Action ?

An action is like a Proc. It is a controller independent method. Right now it 
is not used anywhere except than in Control.

== How to use an action ?

class MyAction < Action
  def run
    # code you want to run
  end

  def template
    %~<p>Your template goes there</p>~
  end
end

There is still a lot of things missing.
 * Use the compiler to build the template from a file (with the transformation 
pipeline)
 * Make a method to bind an action to a controller or to the dispatcher :

class MyController < Nitro::Controller
  attach :index, MyAction
end

The dispatcher would use MyAction.path => "my_action" (String)   to determine 
the default binding path.

== What is Nitro::Control ?

A Control is a ruby class that is capable of generating html associated to a 
value. It is also integrated in the transformer pipeline so that it's usage 
gets natural as an html child node.

== How to use Nitro::Control ?

First, build your control :

class Calendar < Nitro::Control
  def run
    # by default, @name, @value and @options are available. You also have 
access to the Controller's varaibles
    "<input type="text" name="#{@name}" value="#{@value}" />"
  end
end

Then use it in your template :

<Calendar name="hello" value="Time.now" />

The data in value gets evaled at rendering so that you can put controller's 
variables without having to convert them to strings (like Og Entities's 
instances)

You can also put controls in modules : <MyModule::MyControl />

== Dynamic Controls

if you use <Control /> in your template, a control will be chosen depending of 
the data type, using the DynamicControl.map property.

== Nitro::Form::Control shortcomings

The old system has various shortcoming that I found annoying.
 * It is not easy to add new controls.
 * No support for external templating.
 * The mapping can't be done on object's class names and is not dynamical. You 
have to specify one.

== Nitro::Control shortcomings

Here are my implementation's shortcomings
 
 * There is no emit system. Especially emis_js and emit_css.


== The end

Ok, my system is far from perfect. In the fact, I wish I hadn't submitted it 
yet because it is not finished. It might look a little bit awkward, but I had 
good reasons to do it. First of all, I wanted to merge the Element and 
Control things together. While I was at it, I wanted to merge the 
Controller's actions too. If you look at them, they are pretty much the same, 
what makes them different is the context where they are used.

A last word on actions : i liked the idea to group an action with it's 
properties. Instead of defining pre-post things in the controller, you can 
have everything there. Options, settings, ...

Let the questions flow

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From zimba.tm at gmail.com  Thu Mar  2 07:06:29 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 13:06:29 +0100
Subject: [Nitro] [devlab] userdir, rdoc and help
In-Reply-To: <afd80b540603020306kd5d0613o80ee561a7ff46ea9@mail.gmail.com>
References: <200603021139.00618.zimba.tm@gmail.com>
	<afd80b540603020306kd5d0613o80ee561a7ff46ea9@mail.gmail.com>
Message-ID: <200603021306.29863.zimba.tm@gmail.com>

On Thursday 02 March 2006 12:06, George Moschovitis wrote:
> > A bundle rdoc containing all the nitro, glue, og libraries is generated
> > every night at 00:00 automatically.
>
> where can we find this?

Hem :-)

http://devlab.oree.ch/rdoc/

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From lionel.orry at gmail.com  Thu Mar  2 07:30:40 2006
From: lionel.orry at gmail.com (Lionel Orry)
Date: Thu, 02 Mar 2006 13:30:40 +0100
Subject: [Nitro] [devlab] userdir, rdoc and help
In-Reply-To: <200603021306.29863.zimba.tm@gmail.com>
References: <200603021139.00618.zimba.tm@gmail.com>	<afd80b540603020306kd5d0613o80ee561a7ff46ea9@mail.gmail.com>
	<200603021306.29863.zimba.tm@gmail.com>
Message-ID: <4406E570.4050904@gmail.com>



zimba.tm wrote:
> On Thursday 02 March 2006 12:06, George Moschovitis wrote:
>>> A bundle rdoc containing all the nitro, glue, og libraries is generated
>>> every night at 00:00 automatically.
>> where can we find this?
> 
> Hem :-)
> 
> http://devlab.oree.ch/rdoc/
> 
Many thanks for this. This is a good base for newbies to have an overview of the projects, and an extensive reference API at the same time.

I may really get into Nitro this time... ;-)

Lionel



From george.moschovitis at gmail.com  Thu Mar  2 07:37:26 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 2 Mar 2006 14:37:26 +0200
Subject: [Nitro] Powered by Nitro button
In-Reply-To: <3312293ba97892d0b97a76da9170fe86@oggu.de>
References: <3312293ba97892d0b97a76da9170fe86@oggu.de>
Message-ID: <afd80b540603020437l7738f36l199b7437b483002d@mail.gmail.com>

This is the original logo I have created for nitro. I would suggest
that you use the new logo though. Some buttons will be available soon.

-g.

On 3/2/06, Fabian Buch <fabian at oggu.de> wrote:
> There's a nice "Powered by Nitro"-Button on ruby-doc.org. Who made it?
> James? Is it allowed to use it on other nitro-powered websites?
>
> Fabian
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From nusgnaf at gmail.com  Thu Mar  2 07:43:13 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 2 Mar 2006 12:43:13 +0000
Subject: [Nitro] Misterious Nitro bug
In-Reply-To: <716700c90603020319i3896086ey42d0b0ae89536b1f@mail.gmail.com>
References: <716700c90603020319i3896086ey42d0b0ae89536b1f@mail.gmail.com>
Message-ID: <716700c90603020443n283a1fa1k934e2f7f993a779b@mail.gmail.com>

I found something interesting.
$pwd  => /dev/shm/test
$cat > ../a.rb
Dir.pwd # => /dev/shm/test
require 'nitro'
Dir.pwd # => /dev/shm

so, the nitro automatically update working directory to the script lives.
Is this a bug or feature?



From guillaume.pierronnet at gmail.com  Thu Mar  2 07:54:21 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Thu, 2 Mar 2006 13:54:21 +0100
Subject: [Nitro] Misterious Nitro bug
In-Reply-To: <716700c90603020443n283a1fa1k934e2f7f993a779b@mail.gmail.com>
References: <716700c90603020319i3896086ey42d0b0ae89536b1f@mail.gmail.com>
	<716700c90603020443n283a1fa1k934e2f7f993a779b@mail.gmail.com>
Message-ID: <6a7d49ca0603020454o29d88490q@mail.gmail.com>

nitro just Dir.chdir to the directory where the main script (run.rb
most of the time) lives

put  $NITRO_NO_ENVIRONMENT = true just before require 'nitro' to skip
this behaviour.


2006/3/2, Fang Sun <nusgnaf at gmail.com>:
> I found something interesting.
> $pwd  => /dev/shm/test
> $cat > ../a.rb
> Dir.pwd # => /dev/shm/test
> require 'nitro'
> Dir.pwd # => /dev/shm
>
> so, the nitro automatically update working directory to the script lives.
> Is this a bug or feature?
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From nusgnaf at gmail.com  Thu Mar  2 08:03:17 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Thu, 2 Mar 2006 13:03:17 +0000
Subject: [Nitro] Misterious Nitro bug
In-Reply-To: <716700c90603020443n283a1fa1k934e2f7f993a779b@mail.gmail.com>
References: <716700c90603020319i3896086ey42d0b0ae89536b1f@mail.gmail.com>
	<716700c90603020443n283a1fa1k934e2f7f993a779b@mail.gmail.com>
Message-ID: <716700c90603020503t22c1e82dqa0d34f50c08dd5be@mail.gmail.com>

I track down to lib/nitro.rb, line 54
Dir.chdir(File.dirname($0)) cause the problem. I suggest to remove
this. It's always feasible to explicitely chdir to fireup the
application rather than let Nitro set up this behind the scene.
Comments?



From aidan at yoyo.org  Thu Mar  2 08:13:04 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Fri, 3 Mar 2006 00:13:04 +1100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
Message-ID: <79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>



> I will like to propose the following. I will prepare a single homepage
> for www.nitrohq.com. Then I will redirect trac.nitrohq.com and
> repo.nitrohq.com to the trac and darcs on zimbas server respectively.
> Zimba will give ssh access to the core development team members. And
> we should probably start copying content from nitrohq.com to the trac
> wiki.


This all sounds good :-)

I tried to run all the tests on my copy of the repo on devlab, which  
failed.  It brought to light a couple of issues:

1) we need to get all these tests passing (I know Bryan is working  
away on this one) - Bryan, I'd like to run rcov once we get there.

2) there's no database or anything on devlab, so we shouldn't really  
be running code there if at all avoidable.

However, if we want to get a regular build/test cycle going, we'll  
need that at some point.  It's been a long day for me today, so no  
more Nitro contributions from me tonight.  However, I'll be adding to  
the rdoc over the coming days (I'm half thinking of writing a quick  
script which tells us how much we haven't documented - kind of a rcov  
for rdoc :-) ).

Aidan
---
http://www.infurious.com


From zimba.tm at gmail.com  Thu Mar  2 08:34:35 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 14:34:35 +0100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>
Message-ID: <200603021434.35669.zimba.tm@gmail.com>

On Thursday 02 March 2006 14:13, Aidan Rogers wrote:
> > I will like to propose the following. I will prepare a single homepage
> > for www.nitrohq.com. Then I will redirect trac.nitrohq.com and
> > repo.nitrohq.com to the trac and darcs on zimbas server respectively.
> > Zimba will give ssh access to the core development team members. And
> > we should probably start copying content from nitrohq.com to the trac
> > wiki.
>
> This all sounds good :-)
>
> I tried to run all the tests on my copy of the repo on devlab, which
> failed.  It brought to light a couple of issues:
>
> 1) we need to get all these tests passing (I know Bryan is working
> away on this one) - Bryan, I'd like to run rcov once we get there.

I could automatize the run of rcov on devlab. what do you think ?

> 2) there's no database or anything on devlab, so we shouldn't really
> be running code there if at all avoidable.

oree.ch is my own vhost so there is no problem to install any database. 
sqlite3 is already available.

> However, if we want to get a regular build/test cycle going, we'll
> need that at some point. 

Me too. How is it called again ? Continuous control ? I could also install a 
bot that monitors the darcs repo.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From zimba.tm at gmail.com  Thu Mar  2 09:02:39 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 15:02:39 +0100
Subject: [Nitro] working on devlab
Message-ID: <200603021502.39651.zimba.tm@gmail.com>

Hi list,

i'm still working on devlab so it might be a little unstable in the following 
hours.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From bakki.kudva at gmail.com  Thu Mar  2 11:22:31 2006
From: bakki.kudva at gmail.com (Bakki Kudva)
Date: Thu, 2 Mar 2006 11:22:31 -0500
Subject: [Nitro] [devlab] userdir, rdoc and help
In-Reply-To: <200603021306.29863.zimba.tm@gmail.com>
References: <200603021139.00618.zimba.tm@gmail.com>
	<afd80b540603020306kd5d0613o80ee561a7ff46ea9@mail.gmail.com>
	<200603021306.29863.zimba.tm@gmail.com>
Message-ID: <f7f43c920603020822p30857158nf2f2e9880633bc8d@mail.gmail.com>

I get a connection refused msg when I try the url.

bakki

On 3/2/06, zimba.tm <zimba.tm at gmail.com> wrote:
> On Thursday 02 March 2006 12:06, George Moschovitis wrote:
> > > A bundle rdoc containing all the nitro, glue, og libraries is generated
> > > every night at 00:00 automatically.
> >
> > where can we find this?
>
> Hem :-)
>
> http://devlab.oree.ch/rdoc/
>
> --
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From zimba.tm at gmail.com  Thu Mar  2 11:57:04 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 2 Mar 2006 17:57:04 +0100
Subject: [Nitro] [devlab] userdir, rdoc and help
In-Reply-To: <f7f43c920603020822p30857158nf2f2e9880633bc8d@mail.gmail.com>
References: <200603021139.00618.zimba.tm@gmail.com>
	<200603021306.29863.zimba.tm@gmail.com>
	<f7f43c920603020822p30857158nf2f2e9880633bc8d@mail.gmail.com>
Message-ID: <200603021757.04430.zimba.tm@gmail.com>

Yes sorry I have little problems. It will be back later today.

On Thursday 02 March 2006 17:22, Bakki Kudva wrote:
> I get a connection refused msg when I try the url.
>
> bakki
>
> On 3/2/06, zimba.tm <zimba.tm at gmail.com> wrote:
> > On Thursday 02 March 2006 12:06, George Moschovitis wrote:
> > > > A bundle rdoc containing all the nitro, glue, og libraries is
> > > > generated every night at 00:00 automatically.
> > >
> > > where can we find this?
> >
> > Hem :-)
> >
> > http://devlab.oree.ch/rdoc/
> >
> > --
> > Cheers,
> >   zimba.tm
> >
> > weblog : http://zimba.oree.ch
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From bryan.a.soto at gmail.com  Thu Mar  2 13:33:31 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 10:33:31 -0800
Subject: [Nitro] So what does it all mean.
In-Reply-To: <afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
Message-ID: <bd93db0d0603021033x1a34414bv1f2fddb8d82a1d94@mail.gmail.com>

So much to respond to. So many changes. And me in the wrong timezone. :(

On 3/2/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Will it be the host for a real glycerin (basically a test bed for any
> > and all new ideas) with the best new features, bugfixes and RDoc being
> > incorporated into the official repo?
>
> I will like to propose the following. I will prepare a single homepage
> for www.nitrohq.com. Then I will redirect trac.nitrohq.com and
> repo.nitrohq.com to the trac and darcs on zimbas server respectively.
> Zimba will give ssh access to the core development team members. And
> we should probably start copying content from nitrohq.com to the trac
> wiki.

Excellent ideas. Has the problem with caching on the wiki been fixed?
If not, perhaps a dump of the database could be provided? Or if it's
sqlite, maybe the db file itself.

> > Does the core dev team remain the same? If not, how do we handle
> > changes? Elections? Community nominations and selections by George?
>
> Here are the core development team members I propose:
>
> Rob Pitt
> Bryan Sotto
> Zimba
> Aidan
>
> What do you think?

Excellent choices, with your later addition of Guill. Those would have
been my nominations as well. Except for myself, of course. ;)



From james_b at neurogami.com  Thu Mar  2 13:40:59 2006
From: james_b at neurogami.com (James Britt)
Date: Thu, 02 Mar 2006 11:40:59 -0700
Subject: [Nitro] Powered by Nitro button
In-Reply-To: <3312293ba97892d0b97a76da9170fe86@oggu.de>
References: <3312293ba97892d0b97a76da9170fe86@oggu.de>
Message-ID: <44073C3B.9040508@neurogami.com>

Fabian Buch wrote:
> There's a nice "Powered by Nitro"-Button on ruby-doc.org. Who made it? 

I did the first version; Dan Ritz, the designer at 30 Second Rule, fixed 
it up a bit.

> James? Is it allowed to use it on other nitro-powered websites?

Oh, yes.  Please grab a local copy for your Nitro-powered sites.

-- 
James Britt

http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://www.jamesbritt.com     - Playing with Better Toys
http://www.30secondrule.com   - Building Better Tools


From james_b at neurogami.com  Thu Mar  2 13:42:08 2006
From: james_b at neurogami.com (James Britt)
Date: Thu, 02 Mar 2006 11:42:08 -0700
Subject: [Nitro] Powered by Nitro button
In-Reply-To: <afd80b540603020437l7738f36l199b7437b483002d@mail.gmail.com>
References: <3312293ba97892d0b97a76da9170fe86@oggu.de>
	<afd80b540603020437l7738f36l199b7437b483002d@mail.gmail.com>
Message-ID: <44073C80.2090108@neurogami.com>

George Moschovitis wrote:
> This is the original logo I have created for nitro. I would suggest
> that you use the new logo though. Some buttons will be available soon.
> 

Which is the new logo?

James


From bryan.a.soto at gmail.com  Thu Mar  2 13:41:32 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 10:41:32 -0800
Subject: [Nitro] So what does it all mean.
In-Reply-To: <79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>
Message-ID: <bd93db0d0603021041h6ea1e4eai6f75e8cadca6d917@mail.gmail.com>

On 3/2/06, Aidan Rogers <aidan at yoyo.org> wrote:
>
>
> > I will like to propose the following. I will prepare a single homepage
> > for www.nitrohq.com. Then I will redirect trac.nitrohq.com and
> > repo.nitrohq.com to the trac and darcs on zimbas server respectively.
> > Zimba will give ssh access to the core development team members. And
> > we should probably start copying content from nitrohq.com to the trac
> > wiki.
>
>
> This all sounds good :-)
>
> I tried to run all the tests on my copy of the repo on devlab, which
> failed.  It brought to light a couple of issues:
>
> 1) we need to get all these tests passing (I know Bryan is working
> away on this one) - Bryan, I'd like to run rcov once we get there.
>

Installed when I saw Mauricio's post on eigenclass.org. ;)

> 2) there's no database or anything on devlab, so we shouldn't really
> be running code there if at all avoidable.
>
> However, if we want to get a regular build/test cycle going, we'll
> need that at some point.  It's been a long day for me today, so no

It'd also be nice to have something to determine that we're not
failing for different databases. Or that the same tests are failing
for everyone.

> more Nitro contributions from me tonight.  However, I'll be adding to
> the rdoc over the coming days (I'm half thinking of writing a quick
> script which tells us how much we haven't documented - kind of a rcov
> for rdoc :-) ).
>
> Aidan
> ---
> http://www.infurious.com
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From bryan.a.soto at gmail.com  Thu Mar  2 13:49:03 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 10:49:03 -0800
Subject: [Nitro] So what does it all mean.
In-Reply-To: <200603021434.35669.zimba.tm@gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>
	<200603021434.35669.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603021049n4f03e25dj3309b93bf722e942@mail.gmail.com>

On 3/2/06, zimba.tm <zimba.tm at gmail.com> wrote:
> On Thursday 02 March 2006 14:13, Aidan Rogers wrote:
> > > I will like to propose the following. I will prepare a single homepage
> > > for www.nitrohq.com. Then I will redirect trac.nitrohq.com and
> > > repo.nitrohq.com to the trac and darcs on zimbas server respectively.
> > > Zimba will give ssh access to the core development team members. And
> > > we should probably start copying content from nitrohq.com to the trac
> > > wiki.
> >
> > This all sounds good :-)
> >
> > I tried to run all the tests on my copy of the repo on devlab, which
> > failed.  It brought to light a couple of issues:
> >
> > 1) we need to get all these tests passing (I know Bryan is working
> > away on this one) - Bryan, I'd like to run rcov once we get there.
>
> I could automatize the run of rcov on devlab. what do you think ?

Hmm... provides detailed code coverage info... Let's see what it gives
us first. If it's really useful then we can think about automating.

>
> > 2) there's no database or anything on devlab, so we shouldn't really
> > be running code there if at all avoidable.
>
> oree.ch is my own vhost so there is no problem to install any database.
> sqlite3 is already available.

How about Mysql and Postgres then? If we do those, perhaps we can hack
something up to make the Og test suite run against all three. This
would definitely be worth automating as it'd give pretty good feedback
on breakage from patches.

>
> > However, if we want to get a regular build/test cycle going, we'll
> > need that at some point.
>
> Me too. How is it called again ? Continuous control ? I could also install a
> bot that monitors the darcs repo.

How about a full test run every day. Nitro, Glue and Og (all three
major databases). Then have it post results errors and failures. Once
a day would provide pretty quick feedback.

Possible?

>
> --
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From aidan at yoyo.org  Thu Mar  2 15:57:14 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Fri, 3 Mar 2006 07:57:14 +1100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <bd93db0d0603021049n4f03e25dj3309b93bf722e942@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>
	<200603021434.35669.zimba.tm@gmail.com>
	<bd93db0d0603021049n4f03e25dj3309b93bf722e942@mail.gmail.com>
Message-ID: <865FC226-4F59-47DF-B6FA-A6FD7D563641@yoyo.org>

Ideally I'd like to install DamageControl (http:// 
dev.buildpatterns.com/trac/wiki/DamageControl) to run continuous  
integration.  i.e. whenever we check in a patch, all tests are run  
and the build breaks if the tests don't all pass.  DamageControl's  
support for darcs is a bit flaky, but it uses rscm (http:// 
rscm.rubyforge.org) for repository access, which shouldn't take one  
of us long to patch and fix.  My buddy Aslak runs both projects, so  
if we provide the code it can get included quickly enough.  Sounds  
like I'm volunteering, so if anyone else would like to help let me  
know :-)

Also, I'd like to make a recommendation that we don't allow any  
further check-ins unless accompanied by rdoc and unit tests.  Comments?

Aidan
---
http://www.infurious.com


On 03/03/2006, at 5:49 AM, Bryan Soto wrote:


>
> How about a full test run every day. Nitro, Glue and Og (all three
> major databases). Then have it post results errors and failures. Once
> a day would provide pretty quick feedback.
>
> Possible?
>




From bryan.a.soto at gmail.com  Thu Mar  2 16:46:40 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 13:46:40 -0800
Subject: [Nitro] So what does it all mean.
In-Reply-To: <865FC226-4F59-47DF-B6FA-A6FD7D563641@yoyo.org>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<afd80b540603020058r18f53b73x87bae82768c532ec@mail.gmail.com>
	<79EFC0C2-6EB6-4B36-8091-5195CDB843C1@yoyo.org>
	<200603021434.35669.zimba.tm@gmail.com>
	<bd93db0d0603021049n4f03e25dj3309b93bf722e942@mail.gmail.com>
	<865FC226-4F59-47DF-B6FA-A6FD7D563641@yoyo.org>
Message-ID: <bd93db0d0603021346r3090121dya429d9496ac13b11@mail.gmail.com>

On 3/2/06, Aidan Rogers <aidan at yoyo.org> wrote:
> Also, I'd like to make a recommendation that we don't allow any
> further check-ins unless accompanied by rdoc and unit tests.  Comments?

I was expecting this one to come up... ;)

I can agree with the RDoc. If you're fixing a bug or have written a
new method, you're definitely in the best position to write some. Unit
tests though, I'm not opposed to requiring them in theory, but there
are some things that, to me anyway, seem very hard to test
automatically. As an example, see:
http://rubyforge.org/pipermail/nitro-general/2006-January/002601.html

I couldn't think of a good way to unit test that personally. Perhaps
it's just a lack of unit testing experience on my part. Or perhaps
it's a problem with the current design that it's not easily testable.
Or even a hint that unit tests are inappropriate for testing
application level problems which this would, I guess, be. But I
wouldn't want to continue shipping with a bug because I couldn't think
of an automatic test.

Actually, you said you were a TDD nut. Is there a good way to unit
test that? Short of massive refactoring? It'd be nice to ensure it
doesn't happen again. I don't claim to be a good tester. It wasn't
until Nitro/Og that I really understood the value of unit testing and
test suites for regression testing.

Anyway, I suggest this. Adding or updating RDoc is expected. Absence
of a unit test is grounds for rejecting a patch. But, we should allow
some leeway for things that are just plain hard to test and use them
as use cases for a better method of testing.

Sound fair?

And since we're discussing practices, I'd like to suggest that core
devs don't apply their own patches. It's seems to me this would take
advantage of human nature. I, for one, would be a bit more inclined to
follow coding standards and rules regarding RDoc and unit tests if I
were going to be faced with a public "Patch rejected due to lack of
RDoc" message... ;)

Obviously, that's not enforceable. But it seems to me a good practice
and ensures we hold ourselves to the same standards we hold everyone
else to.

And lastly, do we want to continue sending patches to the list? I ask
partly out of ignorance at what Trac supports. I personally like
public review, but I'm sure others have opinions that I'd like to
hear.

Bryan



From bryan.a.soto at gmail.com  Thu Mar  2 17:43:56 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 14:43:56 -0800
Subject: [Nitro] [devlab] userdir, rdoc and help
In-Reply-To: <200603021139.00618.zimba.tm@gmail.com>
References: <200603021139.00618.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603021443y7a5c55c9u12e69625c78c991e@mail.gmail.com>

On 3/2/06, zimba.tm <zimba.tm at gmail.com> wrote:
> Hi list,
>
> here are some today enhancements for devlab.
>
> Userdir
> =======
>
> All ssh users now have their own personnal homepage on url like
> http://devlab.oree.ch/~username
>
> Use scp to copy your files over
>
> Please don't use this space to put heavy files. It is more intended to publish
> your own branch of the darcs repo.

I couldn't resist. I had to try. http://devlab.oree.ch/~bryan/bryans-nitro
Very cool! Nothing major. Just two test fixes that I'll submit
formally. But very cool that you're providing this.

>
> I will look in the future to allow you to run your own ruby apps there for
> testing.
>
>
> RDoc
> ====
>
> A bundle rdoc containing all the nitro, glue, og libraries is generated every
> night at 00:00 automatically.
>
> In the future I will try to separate the different packages like the stdlib on
> ruby-doc.org

Excellent!

>
> Help
> ====
>
> == I still need a nice homepage for devlab.oree.ch
> == Trac needs some customization
> == All in all
>
> I think that we need a graphist to make everything like nice, shiny and
> integrated.

Anyone out there interested in a blank canvas? ;)

>
>
> See you soon for more news. All requests and comments are welcome.
>

zimba, thank you for what you're doing. I hope the paypal account
manveru suggested is set up so you can get help with your expenses.

> --
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From bryan.a.soto at gmail.com  Thu Mar  2 19:36:05 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 16:36:05 -0800
Subject: [Nitro] Should we create a nitro-dev mailing list?
Message-ID: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>

Aidan brought up, what I think is, a good point.

<aidan>
Should we separate things a bit?  Many open source projects with a
bit of momentum have a mailing list for users and one for
developers.  Obviously, anyone can sign up for both, but the intent
is that discussions of code/patches/etc live on the developers
mailing list, so that new users to the project don't unsubscribe from
the mailing list because the traffic flow is too low level.
</aidan>

People developing web apps _with_ Nitro might not care about
development _of_ Nitro. Having the dev discussion on general might be
enough to chase them off. So, to the general population of this list,
how do you feel? Would you rather see patches, etc. moved off of the
general list? Any opinions either way?

Bryan

Aidan's statement for your consideration:



From james_b at neurogami.com  Thu Mar  2 19:49:37 2006
From: james_b at neurogami.com (James Britt)
Date: Thu, 02 Mar 2006 17:49:37 -0700
Subject: [Nitro] Should we create a nitro-dev mailing list?
In-Reply-To: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
Message-ID: <440792A1.20507@neurogami.com>

No.  Traffic is low enough.  I scan posts quickly, and some of the dev 
stuff suggests how code is supposed to be used.  It's helpful.



-- 
James Britt


From bryan.a.soto at gmail.com  Thu Mar  2 20:11:00 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 17:11:00 -0800
Subject: [Nitro] PATCHSET: tests
Message-ID: <bd93db0d0603021710y4ad5324btbd8f3a772a8eda94@mail.gmail.com>

* test_add_og_tc_sti
  Adds a test case to ensure STI parents and children return the same
fields when creating a database table.


* test_fix_tc_session
  Added back og cache test. Conditional run of memcached test.


* test_fix_tc_property
  Slightly modifies Rob's patch to maintain backwards compatibility.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tests.zip
Type: application/zip
Size: 29005 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060302/2dbac73a/attachment.zip 

From bakki.kudva at gmail.com  Thu Mar  2 20:41:55 2006
From: bakki.kudva at gmail.com (Bakki Kudva)
Date: Thu, 2 Mar 2006 20:41:55 -0500
Subject: [Nitro] Should we create a nitro-dev mailing list?
In-Reply-To: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
Message-ID: <f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>

May I ask how many of the subscribers are users vs. developers?

 It might make more sense to split if traffic is too heavy and too
much newbie stuff ( me for isntance :) or application issues is
clogging up the list from a developer point of view.

On the other hand dev discussion is helpful to app developers and
newbies to understand the architecture. I for one have enjoyed all the
interchanges among core developers and dug into the source to see what
they were talking about.

As long as both lists are open it won't matter.

-bakki

On 3/2/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Aidan brought up, what I think is, a good point.
>
> <aidan>
> Should we separate things a bit?  Many open source projects with a
> bit of momentum have a mailing list for users and one for
> developers.  Obviously, anyone can sign up for both, but the intent
> is that discussions of code/patches/etc live on the developers
> mailing list, so that new users to the project don't unsubscribe from
> the mailing list because the traffic flow is too low level.
> </aidan>
>
> People developing web apps _with_ Nitro might not care about
> development _of_ Nitro. Having the dev discussion on general might be
> enough to chase them off. So, to the general population of this list,
> how do you feel? Would you rather see patches, etc. moved off of the
> general list? Any opinions either way?
>
> Bryan
>
> Aidan's statement for your consideration:
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From aidan at yoyo.org  Thu Mar  2 22:57:39 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Fri, 3 Mar 2006 14:57:39 +1100
Subject: [Nitro] PATCHSET: tests
In-Reply-To: <bd93db0d0603021710y4ad5324btbd8f3a772a8eda94@mail.gmail.com>
References: <bd93db0d0603021710y4ad5324btbd8f3a772a8eda94@mail.gmail.com>
Message-ID: <C1E58FF0-84D2-4117-A298-842B51FA9E5D@yoyo.org>

Bryan,

In your first test, the top line is missing the '..' needed to load  
the CONFIG.rb correctly.

require File.join(File.dirname(__FILE__), '..', 'CONFIG.rb')

test_fix_tc_property looks fine.  I don't have the knowledge to  
comment on the tc_session fix.

Aidan
---
http://www.infurious.com


From james_b at neurogami.com  Thu Mar  2 23:20:56 2006
From: james_b at neurogami.com (James Britt)
Date: Thu, 02 Mar 2006 21:20:56 -0700
Subject: [Nitro] Should we create a nitro-dev mailing list?
In-Reply-To: <f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
	<f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>
Message-ID: <4407C428.8000402@neurogami.com>

Bakki Kudva wrote:
> May I ask how many of the subscribers are users vs. developers?
>

In my Nitro experience, users sometimes become developers in order to 
see something get fixed in a timely manner.



:)



-- 
James Britt

http://web2.0validator.com    - We're the Dot in Web 2.0
http://refreshingcities.org   - Design, technology, usability
http://yourelevatorpitch.com  - Finding Business Focus
http://www.jamesbritt.com     - Playing with Better Toys


From bryan.a.soto at gmail.com  Fri Mar  3 02:39:09 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 2 Mar 2006 23:39:09 -0800
Subject: [Nitro] PATCHSET: tests
In-Reply-To: <C1E58FF0-84D2-4117-A298-842B51FA9E5D@yoyo.org>
References: <bd93db0d0603021710y4ad5324btbd8f3a772a8eda94@mail.gmail.com>
	<C1E58FF0-84D2-4117-A298-842B51FA9E5D@yoyo.org>
Message-ID: <bd93db0d0603022339g5b7e300cjcebe258d1dd1915a@mail.gmail.com>

On 3/2/06, Aidan Rogers <aidan at yoyo.org> wrote:
> Bryan,
>
> In your first test, the top line is missing the '..' needed to load
> the CONFIG.rb correctly.
>
> require File.join(File.dirname(__FILE__), '..', 'CONFIG.rb')
>

Yes, after I finished it, I thought it was more appropriate in the
store subdirectory.

Attached corrects and should replace previous. Thanks for catching.

> test_fix_tc_property looks fine.  I don't have the knowledge to
> comment on the tc_session fix.
>
> Aidan
> ---
> http://www.infurious.com
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test_add_og_tc_sti.zip
Type: application/zip
Size: 9872 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060303/ba79180e/attachment.zip 

From bryan.a.soto at gmail.com  Fri Mar  3 03:00:15 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 3 Mar 2006 00:00:15 -0800
Subject: [Nitro] Should we create a nitro-dev mailing list?
In-Reply-To: <4407C428.8000402@neurogami.com>
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
	<f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>
	<4407C428.8000402@neurogami.com>
Message-ID: <bd93db0d0603030000q6243e0c7jc6f6b3990c25bb9a@mail.gmail.com>

On 3/2/06, James Britt <james_b at neurogami.com> wrote:
> Bakki Kudva wrote:
> > May I ask how many of the subscribers are users vs. developers?
> >
>
> In my Nitro experience, users sometimes become developers in order to
> see something get fixed in a timely manner.
>
> :)

Okay, that one was pretty good. :)



From george.moschovitis at gmail.com  Fri Mar  3 03:01:21 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 10:01:21 +0200
Subject: [Nitro] Powered by Nitro button
In-Reply-To: <44073C80.2090108@neurogami.com>
References: <3312293ba97892d0b97a76da9170fe86@oggu.de>
	<afd80b540603020437l7738f36l199b7437b483002d@mail.gmail.com>
	<44073C80.2090108@neurogami.com>
Message-ID: <afd80b540603030001m4ef07c95pbf5606878dc53326@mail.gmail.com>

the one on www.nitrohq.com

-g.

On 3/2/06, James Britt <james_b at neurogami.com> wrote:
> George Moschovitis wrote:
> > This is the original logo I have created for nitro. I would suggest
> > that you use the new logo though. Some buttons will be available soon.
> >
>
> Which is the new logo?
>
> James
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Fri Mar  3 03:11:33 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 3 Mar 2006 00:11:33 -0800
Subject: [Nitro] Should we create a nitro-dev mailing list?
In-Reply-To: <f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
	<f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>
Message-ID: <bd93db0d0603030011k46957a55r2006acd6841d887d@mail.gmail.com>

On 3/2/06, Bakki Kudva <bakki.kudva at gmail.com> wrote:
> May I ask how many of the subscribers are users vs. developers?
>
>  It might make more sense to split if traffic is too heavy and too
> much newbie stuff ( me for isntance :) or application issues is
> clogging up the list from a developer point of view.
>

Well, I don't think we have too much traffic. And personally, I'd
rather have more newbie stuff. ;)

> On the other hand dev discussion is helpful to app developers and
> newbies to understand the architecture. I for one have enjoyed all the
> interchanges among core developers and dug into the source to see what
> they were talking about.
>

That's actually what I was concerned with personally. I wanted to make
sure people didn't dread seeing all those patch emails in their
inboxes.

> As long as both lists are open it won't matter.
>
> -bakki



From james_b at neurogami.com  Fri Mar  3 03:28:43 2006
From: james_b at neurogami.com (James Britt)
Date: Fri, 03 Mar 2006 01:28:43 -0700
Subject: [Nitro] Powered by Nitro button
In-Reply-To: <afd80b540603030001m4ef07c95pbf5606878dc53326@mail.gmail.com>
References: <3312293ba97892d0b97a76da9170fe86@oggu.de>	<afd80b540603020437l7738f36l199b7437b483002d@mail.gmail.com>	<44073C80.2090108@neurogami.com>
	<afd80b540603030001m4ef07c95pbf5606878dc53326@mail.gmail.com>
Message-ID: <4407FE3B.4080606@neurogami.com>

George Moschovitis wrote:
> the one on www.nitrohq.com

Oh.

Not terribly striking, and when shrunk down to a button is 
indistinguishable from any plain san-serif font.

I suggest giving some thought to a more eye-catching logo.


James


From james_b at neurogami.com  Fri Mar  3 03:30:23 2006
From: james_b at neurogami.com (James Britt)
Date: Fri, 03 Mar 2006 01:30:23 -0700
Subject: [Nitro] Should we create a nitro-dev mailing list?
In-Reply-To: <bd93db0d0603030011k46957a55r2006acd6841d887d@mail.gmail.com>
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>	<f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>
	<bd93db0d0603030011k46957a55r2006acd6841d887d@mail.gmail.com>
Message-ID: <4407FE9F.9060801@neurogami.com>

Bryan Soto wrote:

> ...
> That's actually what I was concerned with personally. I wanted to make
> sure people didn't dread seeing all those patch emails in their
> inboxes.
> 

If people are good about subject tags, then filtering and such is trivial.



James


From george.moschovitis at gmail.com  Fri Mar  3 03:41:35 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 10:41:35 +0200
Subject: [Nitro] Should we create a nitro-dev mailing list?
In-Reply-To: <4407FE9F.9060801@neurogami.com>
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com>
	<f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com>
	<bd93db0d0603030011k46957a55r2006acd6841d887d@mail.gmail.com>
	<4407FE9F.9060801@neurogami.com>
Message-ID: <afd80b540603030041n1e827dc8w8134e21d92dfd6ed@mail.gmail.com>

For the moment I prefer a single mailing list.

regards,
George.


On 3/3/06, James Britt <james_b at neurogami.com> wrote:
> Bryan Soto wrote:
>
> > ...
> > That's actually what I was concerned with personally. I wanted to make
> > sure people didn't dread seeing all those patch emails in their
> > inboxes.
> >
>
> If people are good about subject tags, then filtering and such is trivial.
>
>
>
> James
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar  3 03:44:38 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 10:44:38 +0200
Subject: [Nitro] nitro rake test
Message-ID: <afd80b540603030044ue25d9f6icf4245d6095fcbeb@mail.gmail.com>

Dear devs,

when I run

rake test in the nitro dir I get the following:

D, [2006-03-03T10:42:59.619772 #9744] DEBUG -- : Using Memory sessions.
/usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb:5:in
` load': no such file to load -- test/nitro/tc_session.rb (LoadError)
        from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
der.rb:5
        from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
der.rb:5
rake aborted!

and the file test/nitro/tc_session.rb *exists*. Anyone has any ideas
what is causing this?

thanks,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar  3 03:45:37 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 10:45:37 +0200
Subject: [Nitro] updating AUTHORS
Message-ID: <afd80b540603030045l1c29b345q72152673e5924955@mail.gmail.com>

Dear devs,

I am wondering if I have missed the name of contributor from the
AUTHORS file. If so, please let me know!

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar  3 05:09:04 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 12:09:04 +0200
Subject: [Nitro] Patches...
Message-ID: <afd80b540603030209n61fbd239x79763a3f6a0fd75c@mail.gmail.com>

Please continue to send patches to this list with a [PATCH] Prefix in
the subject.
It is very useful to keep track of all the changes.

thanks,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From zimba.tm at gmail.com  Fri Mar  3 05:37:08 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 3 Mar 2006 11:37:08 +0100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <bd93db0d0603021049n4f03e25dj3309b93bf722e942@mail.gmail.com>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<200603021434.35669.zimba.tm@gmail.com>
	<bd93db0d0603021049n4f03e25dj3309b93bf722e942@mail.gmail.com>
Message-ID: <200603031137.08705.zimba.tm@gmail.com>

On Thursday 02 March 2006 19:49, Bryan Soto wrote:
> How about Mysql and Postgres then? If we do those, perhaps we can hack
> something up to make the Og test suite run against all three. This
> would definitely be worth automating as it'd give pretty good feedback
> on breakage from patches.

Ok I have installed and fired-up both on the system. You can get access with 
no password (for mysql at least)

> How about a full test run every day. Nitro, Glue and Og (all three
> major databases). Then have it post results errors and failures. Once
> a day would provide pretty quick feedback.
>
> Possible?

Yes, my crontab already build the docs daily.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From zimba.tm at gmail.com  Fri Mar  3 05:41:51 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 3 Mar 2006 11:41:51 +0100
Subject: [Nitro] So what does it all mean.
In-Reply-To: <865FC226-4F59-47DF-B6FA-A6FD7D563641@yoyo.org>
References: <bd93db0d0603011609s5cad8953p5076e135d6e19dab@mail.gmail.com>
	<bd93db0d0603021049n4f03e25dj3309b93bf722e942@mail.gmail.com>
	<865FC226-4F59-47DF-B6FA-A6FD7D563641@yoyo.org>
Message-ID: <200603031141.51266.zimba.tm@gmail.com>

On Thursday 02 March 2006 21:57, Aidan Rogers wrote:
> Ideally I'd like to install DamageControl (http://
> dev.buildpatterns.com/trac/wiki/DamageControl) to run continuous
> integration.  i.e. whenever we check in a patch, all tests are run
> and the build breaks if the tests don't all pass.  DamageControl's
> support for darcs is a bit flaky, but it uses rscm (http://
> rscm.rubyforge.org) for repository access, which shouldn't take one
> of us long to patch and fix.  My buddy Aslak runs both projects, so
> if we provide the code it can get included quickly enough.  Sounds
> like I'm volunteering, so if anyone else would like to help let me
> know :-)

Let me know what you need to install it on devlab

> Also, I'd like to make a recommendation that we don't allow any
> further check-ins unless accompanied by rdoc and unit tests.  Comments?

Hmm. It might be a little too constraining. I'd better use thedev + 
stabilization phases and ask for test when new features are added.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From fabian at oggu.de  Fri Mar  3 08:35:51 2006
From: fabian at oggu.de (Fabian Buch)
Date: Fri, 3 Mar 2006 14:35:51 +0100
Subject: [Nitro] [patch] RFC typo <-> English, of referrer <-> referer
Message-ID: <a66cb14ae1a6396680684d82fd4a7257@oggu.de>

Added an alias_method :redirect_referrer, :redirect_referer

Referrer is written with rr in English, but referer (with one r) became 
standard, because of a typo in the original RFC. But not e
veryone might know that, so might write it with two r, which shouldn't 
lead to the app not working.

Fabian



From mischa.kroon at gmail.com  Fri Mar  3 08:47:03 2006
From: mischa.kroon at gmail.com (Mischa Kroon)
Date: Fri, 3 Mar 2006 14:47:03 +0100
Subject: [Nitro] Should we create a nitro-dev mailing list?
References: <bd93db0d0603021636s17ce1af7p80fa0a4e76e48f97@mail.gmail.com><f7f43c920603021741h62a4e23j9c5012c73be31e48@mail.gmail.com><bd93db0d0603030011k46957a55r2006acd6841d887d@mail.gmail.com><4407FE9F.9060801@neurogami.com>
	<afd80b540603030041n1e827dc8w8134e21d92dfd6ed@mail.gmail.com>
Message-ID: <001801c63ec8$f517b230$0a01a8c0@mischabak>

I would like to cast a tiny vote for splitting it up :)

Devs would still be looking at both lists probably.
But it's a bit nicer for people who just want to get stuff done through 
nitro.

Or who are totally new to nitro.

This list the way it is now is quite intimidating, and for a Ruby noob like 
myself it's quite hard to find points to get started or to find starting 
points.

For some experimental sites or stuff like that ...
So it's a tad more noob friendly to go with seperate lists.

For the hardcore the 1 list is a tad more convienent.
( which is also the more vocal part of this list offcourse )

Just my 2 bits :)

Mischa

----- Original Message ----- 
From: "George Moschovitis" <george.moschovitis at gmail.com>
To: "General discussion about Nitro" <nitro-general at rubyforge.org>
Sent: Friday, March 03, 2006 9:41 AM
Subject: Re: [Nitro] Should we create a nitro-dev mailing list?


> For the moment I prefer a single mailing list.
>
> regards,
> George.



From zimba.tm at gmail.com  Fri Mar  3 09:06:33 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 3 Mar 2006 15:06:33 +0100
Subject: [Nitro] [devlab] Trac auth, databases & Help
Message-ID: <200603031506.33816.zimba.tm@gmail.com>

Hi list,

here are the today news about devlab.

== Trac auth

Core developpers now can authenticate themselves with the same password as ssh 
access into trac.

I don't know if it is userfull but it lets us move to a more restricted 
environment if spam or other buggers start to appear

== Databases

Mysql + Postgresql and Sqlite3 are now available and open to every core 
developper. Maybe except for postgresql, there are no password and I consider 
that you are mature enough to not brake things up just for the fun.

This is a step forward to automated testing. Let me know if we need any other 
databases (kirby ?)

== Help

= Design

So first of all, my offer for some graphist to make a nice design for devlab 
is still open ;-)

= Darcs

I would like to split the repository in different parts. Nitro, Og, Glue, 
Spark, Flare, Gen and move example into Nitro. And have a trac instance for 
each one of them. 

In my opinion, it would help to make the separation between the different 
things clearer.

What do you think ?

= Damage Control

It would be nice to install Damage Control into the host. I've installed all 
the required dependencies but didn't get further.
---- log ----
ruby init.rb --data-dir ~/damagecontrol_data
=> /var/devlab/damagecontrol_data/dc.yml not found, copying 
from /var/devlab/damagecontrol/config/dc.yml
=> Booting lighttpd (use 'script/server webrick' to force WEBrick)
server: invalid option: --port
=> Rails application started on http://0.0.0.0:3000
=> Call with -d to detach
=> Ctrl-C to shutdown server (see config/lighttpd.conf for options)
2006-03-03 16:42:29: (config.c.648) line: 8 pos: 56 invalid value field 
2006-03-03 16:42:29: (config.c.730) configfile parser failed 
Exiting
Couldn't find any process 
matching: /var/devlab/damagecontrol/public/dispatch.fcgi
devlab at oree ~/damagecontrol $ Failed to start process 
#<DamageControl::Process::Builder:0xb7ed9964>
----- end of log -----


-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From george.moschovitis at gmail.com  Fri Mar  3 09:12:06 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 16:12:06 +0200
Subject: [Nitro] [devlab] Trac auth, databases & Help
In-Reply-To: <200603031506.33816.zimba.tm@gmail.com>
References: <200603031506.33816.zimba.tm@gmail.com>
Message-ID: <afd80b540603030612l24af7585t41cc90bff7be80b4@mail.gmail.com>

> I would like to split the repository in different parts. Nitro, Og, Glue,
> Spark, Flare, Gen and move example into Nitro. And have a trac instance for
> each one of them.
>
> In my opinion, it would help to make the separation between the different
> things clearer.
>
> What do you think ?

Please no! Can you please keep one repository for the momen? We can
reconsider this in 1-2 months and change if needed.

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar  3 09:20:08 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 16:20:08 +0200
Subject: [Nitro] [patch] RFC typo <-> English, of referrer <-> referer
In-Reply-To: <a66cb14ae1a6396680684d82fd4a7257@oggu.de>
References: <a66cb14ae1a6396680684d82fd4a7257@oggu.de>
Message-ID: <afd80b540603030620p41749cf7gd30f2e4ea96502d@mail.gmail.com>

You forgot the patch ;-)

-g.

On 3/3/06, Fabian Buch <fabian at oggu.de> wrote:
> Added an alias_method :redirect_referrer, :redirect_referer
>
> Referrer is written with rr in English, but referer (with one r) became
> standard, because of a typo in the original RFC. But not e
> veryone might know that, so might write it with two r, which shouldn't
> lead to the app not working.
>
> Fabian
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From zimba.tm at gmail.com  Fri Mar  3 09:30:22 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 3 Mar 2006 15:30:22 +0100
Subject: [Nitro] [devlab] Trac auth, databases & Help
In-Reply-To: <afd80b540603030612l24af7585t41cc90bff7be80b4@mail.gmail.com>
References: <200603031506.33816.zimba.tm@gmail.com>
	<afd80b540603030612l24af7585t41cc90bff7be80b4@mail.gmail.com>
Message-ID: <200603031530.22130.zimba.tm@gmail.com>

On Friday 03 March 2006 15:12, George Moschovitis wrote:
> > I would like to split the repository in different parts. Nitro, Og, Glue,
> > Spark, Flare, Gen and move example into Nitro. And have a trac instance
> > for each one of them.
> >
> > In my opinion, it would help to make the separation between the different
> > things clearer.
> >
> > What do you think ?
>
> Please no! Can you please keep one repository for the momen? We can
> reconsider this in 1-2 months and change if needed.

Sure, it was only a question. 

I would prefer to setup de system now so that things don't get changed too 
much later but it's like you want.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From george.moschovitis at gmail.com  Fri Mar  3 09:36:28 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 16:36:28 +0200
Subject: [Nitro] [devlab] Trac auth, databases & Help
In-Reply-To: <200603031530.22130.zimba.tm@gmail.com>
References: <200603031506.33816.zimba.tm@gmail.com>
	<afd80b540603030612l24af7585t41cc90bff7be80b4@mail.gmail.com>
	<200603031530.22130.zimba.tm@gmail.com>
Message-ID: <afd80b540603030636i45cb3904uef74eb082464fc1c@mail.gmail.com>

please lets keep it this way for the moment.

one more request. lets keep the repo.nitrohq.com repo for 1-2 more
days till I prepare version 0.29.0.

Then I will switch the domain to your repository. I will be missing
for some more days (but this time not so long).

best regards,
George.



On 3/3/06, zimba.tm <zimba.tm at gmail.com> wrote:
> On Friday 03 March 2006 15:12, George Moschovitis wrote:
> > > I would like to split the repository in different parts. Nitro, Og, Glue,
> > > Spark, Flare, Gen and move example into Nitro. And have a trac instance
> > > for each one of them.
> > >
> > > In my opinion, it would help to make the separation between the different
> > > things clearer.
> > >
> > > What do you think ?
> >
> > Please no! Can you please keep one repository for the momen? We can
> > reconsider this in 1-2 months and change if needed.
>
> Sure, it was only a question.
>
> I would prefer to setup de system now so that things don't get changed too
> much later but it's like you want.
>
> --
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar  3 09:37:26 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 16:37:26 +0200
Subject: [Nitro] Patches for 0.29.0
Message-ID: <afd80b540603030637n5bc0e2acs582f842749ec32da@mail.gmail.com>

Dear devs,

If anyone has any patches for version 0.29.0 now is the right time to
post them to the list. I aim to release the 0.29.0 gems on Monday.

best regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From fabian at oggu.de  Fri Mar  3 09:43:54 2006
From: fabian at oggu.de (Fabian Buch)
Date: Fri, 3 Mar 2006 15:43:54 +0100
Subject: [Nitro] [patch] RFC typo <-> English, of referrer <-> referer
In-Reply-To: <afd80b540603030620p41749cf7gd30f2e4ea96502d@mail.gmail.com>
References: <a66cb14ae1a6396680684d82fd4a7257@oggu.de>
	<afd80b540603030620p41749cf7gd30f2e4ea96502d@mail.gmail.com>
Message-ID: <bc4d52828f163cc197b3fc7d00d0103b@oggu.de>

oops, thanks, attached now

Am 03.03.2006 um 15:20 schrieb George Moschovitis:
> You forgot the patch ;-)
>
> -g.
>
> On 3/3/06, Fabian Buch <fabian at oggu.de> wrote:
>> Added an alias_method :redirect_referrer, :redirect_referer
>>
>> Referrer is written with rr in English, but referer (with one r) 
>> became
>> standard, because of a typo in the original RFC. But not e
>> veryone might know that, so might write it with two r, which shouldn't
>> lead to the app not working.
>>
>> Fabian

-------------- next part --------------
A non-text attachment was scrubbed...
Name: referrer_alias.patch
Type: application/octet-stream
Size: 25958 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060303/39704221/attachment.obj 
-------------- next part --------------


From zimba.tm at gmail.com  Fri Mar  3 09:59:35 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 3 Mar 2006 15:59:35 +0100
Subject: [Nitro] [devlab] Trac auth, databases & Help
In-Reply-To: <afd80b540603030636i45cb3904uef74eb082464fc1c@mail.gmail.com>
References: <200603031506.33816.zimba.tm@gmail.com>
	<200603031530.22130.zimba.tm@gmail.com>
	<afd80b540603030636i45cb3904uef74eb082464fc1c@mail.gmail.com>
Message-ID: <200603031559.35966.zimba.tm@gmail.com>

On Friday 03 March 2006 15:36, George Moschovitis wrote:
> please lets keep it this way for the moment.
>
> one more request. lets keep the repo.nitrohq.com repo for 1-2 more
> days till I prepare version 0.29.0.

okkidokie

> Then I will switch the domain to your repository. I will be missing
> for some more days (but this time not so long).

What would you think of having a dedicated virtual host for nitrohq.com ? I am 
thinking for this since some time. I could probably convince my friend to 
setup one. You and I would get root access.

The only problem is that the bandwith costs around 5$ per Gigaoctet. If it 
isn't too high I can take the costs and otherwise we can decentralize files 
to rubyforge or somewhere else.

But it would be really nice. You look pretty much occupied and other people at 
Navel too. I would be happy to be helpful when you're not there.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From aglarond at gmail.com  Fri Mar  3 10:37:30 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Fri, 3 Mar 2006 16:37:30 +0100
Subject: [Nitro] [devlab] Trac auth, databases & Help
In-Reply-To: <200603031506.33816.zimba.tm@gmail.com>
References: <200603031506.33816.zimba.tm@gmail.com>
Message-ID: <55c107bf0603030737m435ab9efgb40775bce39f4ba2@mail.gmail.com>

Hi Zimba,

On 3/3/06, zimba.tm <zimba.tm at gmail.com> wrote:

> == Databases
>
> Mysql + Postgresql and Sqlite3 are now available and open to every core
> developper. Maybe except for postgresql, there are no password and I consider
> that you are mature enough to not brake things up just for the fun.
>

To allow local PostgreSQL users access without a password, place the
following into your pg_hba.conf file:

local   all             <username>                      trust

> This is a step forward to automated testing. Let me know if we need any other
> databases (kirby ?)
>

I would vote for KirbyBase being installed as well.  It's such a nice,
flat-file database, it would be a shame for compatibility to break at
some random time in the future.

- Dimitri

P.S.  Where in Switzerland are you? (Can be answered privately, of course.)



From george.moschovitis at gmail.com  Fri Mar  3 10:50:33 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 17:50:33 +0200
Subject: [Nitro] [devlab] Trac auth, databases & Help
In-Reply-To: <200603031559.35966.zimba.tm@gmail.com>
References: <200603031506.33816.zimba.tm@gmail.com>
	<200603031530.22130.zimba.tm@gmail.com>
	<afd80b540603030636i45cb3904uef74eb082464fc1c@mail.gmail.com>
	<200603031559.35966.zimba.tm@gmail.com>
Message-ID: <afd80b540603030750l2d224aa9m6630b8d319513c4b@mail.gmail.com>

> okkidokie

thanks!

> But it would be really nice. You look pretty much occupied and other people at
> Navel too. I would be happy to be helpful when you're not there.

We (Navel) can handle this, we are a bit busy at the moment, but be
assured, we will devote more time and energy in nitro and nitrohq.com
real soon ;-) But, no more now, we *love* surprising people ;-)

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar  3 11:43:44 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 3 Mar 2006 18:43:44 +0200
Subject: [Nitro] About the new Control system
In-Reply-To: <200603021258.39943.zimba.tm@gmail.com>
References: <200603021258.39943.zimba.tm@gmail.com>
Message-ID: <afd80b540603030843r3b3866afjf6f17b9043055394@mail.gmail.com>

Can you provide an example that could not be implemented with
Elements? Ie I could easily implement <Calendar > with an element.

thanks,
George.



From m.fellinger at gmail.com  Fri Mar  3 12:10:47 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Sat, 4 Mar 2006 02:10:47 +0900
Subject: [Nitro] [PATCH] Fixed the template-root for flare
Message-ID: <9c00d3e00603030910j139e7967td0aa9b8fa8bdee3b@mail.gmail.com>

Hello List,

I just made a bundle that makes the template-root for flare more
standard, moving it from /template to /templates and adding a
Template.root = "templates" in the run.rb
I could have left this one out since it's the standard, but i'd rather
go sure...

~~~~manveru
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fixed_template_root_flare.tar.bz2
Type: application/x-bzip2
Size: 10815 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060303/e3ce6921/attachment.bz2 

From zimba.tm at gmail.com  Fri Mar  3 12:23:26 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 3 Mar 2006 18:23:26 +0100
Subject: [Nitro] [flame] Bad things in Nitro and Og
Message-ID: <200603031823.26197.zimba.tm@gmail.com>

Hello list, hello George :-)

this is not an ordinary email. I'm not going to be kind. I'm going to 
criticize.

== General

= Where is the doc ?

well this is a know subject. This topic also goes to rdoc with method, class 
and file definitions. There is also no global overview that lets you know how 
the system works generally without diving into the code.

= Bad fixme and todos

there are lots FIXME and TODO in the code but they are not explained well 
enough for other coders to participate

= Path handling could be better

Some parts in the code define where things are at loading. It would be much 
better to use the $PATH method where files are looked-up on need. I'm talking 
about template paths particularily.

= Module namespaces are not always respected

Generally, it is bad to remove the namespace of the class's name. It would be 
better to have a generic way of handling :
 * Class <=> method name   (see facet/class/method_name)
 * Class <=> path  (see facet/class/unix_path)
 * others

== Nitro

= It takes hours to start

I don't know why it takes so long to start. Probably because admin action's 
templates are generated at start but it is not the only problem.

= Reloading is not handled well in all cases

Avoid constants because they generate warning when reloaded. There is also 
other problems with variables that are defined on loading. That should also 
be avoided.

= Everything is magic

It is hard to understand what is going on in the code because lots of things 
have to do with inheritance, extension. Plus lot of facets methods are used 
which makes it longer to understand what it is supposed to do. I don't know 
if it's really necessary to use all those facets methods. I would prefer that 
the code sticks to standard ruby where possible.

= Views are tied to one template extension (.xthml by default)

It makes it harder to implement different kind of sub-html protocols. For 
example .xul and .rss. It would be nice if each extension would have a 
different transformation path.

== Og

= No custom types

Right now, it is not possible to add custom types into Og easily. Let's say I 
want to store a File. I don't want to create a new class called WebFile. I 
would like to tell Og how it should store and restore it.

== Enhancements

= stored settings

It would be nice if this would be possible. Setting already supports a yaml 
export. This could be generalized.

== End

Don't forget, this is only my opinion. I can be wrong and am subject to 
changes too.

I have other griefs but this is enough for today :-p

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From zimba.tm at gmail.com  Fri Mar  3 12:50:10 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 3 Mar 2006 18:50:10 +0100
Subject: [Nitro] About the new Control system
In-Reply-To: <afd80b540603030843r3b3866afjf6f17b9043055394@mail.gmail.com>
References: <200603021258.39943.zimba.tm@gmail.com>
	<afd80b540603030843r3b3866afjf6f17b9043055394@mail.gmail.com>
Message-ID: <200603031850.10543.zimba.tm@gmail.com>

On Friday 03 March 2006 17:43, George Moschovitis wrote:
> Can you provide an example that could not be implemented with
> Elements? Ie I could easily implement <Calendar > with an element.

Maybe, I never understood how Element work exactly. How do you pass ruby 
values to it that are not strings ?

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From bryan.a.soto at gmail.com  Fri Mar  3 16:36:08 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 3 Mar 2006 13:36:08 -0800
Subject: [Nitro] [flame] Bad things in Nitro and Og
In-Reply-To: <200603031823.26197.zimba.tm@gmail.com>
References: <200603031823.26197.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>

I'll respond where I think I can contribute something...

On 3/3/06, zimba.tm <zimba.tm at gmail.com> wrote:
> Hello list, hello George :-)
>
> this is not an ordinary email. I'm not going to be kind. I'm going to
> criticize.
>
> == General
>
> = Where is the doc ?
>
> well this is a know subject. This topic also goes to rdoc with method, class
> and file definitions. There is also no global overview that lets you know how
> the system works generally without diving into the code.
>

I suppose that's because the set of people who enjoy writing docs and
the set of people capable writing the docs is small... I actually do
enjoy writing, but the results of the Doc Survey (published on the
Devlab wiki by the way) showed an overwhelming preference for Nitro
specific docs. As people have probably noticed, I'm much more
comfortable with Og. That said, I'm studying Fowler's Patterns of
Enterprise Applications and Gross' Ajax Patterns and Best Practices.
That plus what I learn while debugging, and hopefully be able to
contribute something.

> = Bad fixme and todos
>
> there are lots FIXME and TODO in the code but they are not explained well
> enough for other coders to participate
>
> = Path handling could be better
>
> Some parts in the code define where things are at loading. It would be much
> better to use the $PATH method where files are looked-up on need. I'm talking
> about template paths particularily.
>
> = Module namespaces are not always respected
>
> Generally, it is bad to remove the namespace of the class's name. It would be
> better to have a generic way of handling :
>  * Class <=> method name   (see facet/class/method_name)
>  * Class <=> path  (see facet/class/unix_path)
>  * others
>
> == Nitro
>
> = It takes hours to start
>
> I don't know why it takes so long to start. Probably because admin action's
> templates are generated at start but it is not the only problem.
>

Nitro alone? Or Nitro and Og? On startup, Og makes a lot of database
calls. Basically checking if the database and tables exist, if tables
need fields added or deleted. It also iterates through ObjectSpace
looking for classes to manage, so there's alot to do. Is startup time
a concern with web apps in general? I'd figure response time when
processing a request would be where people would look. Or does it it
make a bad impression?

> = Reloading is not handled well in all cases
>
> Avoid constants because they generate warning when reloaded. There is also
> other problems with variables that are defined on loading. That should also
> be avoided.
>

irb(main):001:0> C = Class
=> Class
irb(main):002:0> C = Class
(irb):2: warning: already initialized constant C
=> Class
irb(main):003:0> C = Class unless defined?(C)
=> nil

Reloading uses the standard Ruby method Kernel#load, so the behaviour
you're seeing is from Ruby. I'm not sure that we should change it.
Perhaps we can create a best practices page on the wiki and note this?

> = Everything is magic
>
> It is hard to understand what is going on in the code because lots of things
> have to do with inheritance, extension. Plus lot of facets methods are used
> which makes it longer to understand what it is supposed to do. I don't know
> if it's really necessary to use all those facets methods. I would prefer that
> the code sticks to standard ruby where possible.
>

And evals and define_method... It's not the simplest code base to
understand, but it's what makes Nitro, Nitro. As to the facets
methods, I think, for the most part, they are necessary and were put
into the facets package because they were considered reusable. If they
weren't in facets, they'd probably just be in glue.

> = Views are tied to one template extension (.xthml by default)
>
> It makes it harder to implement different kind of sub-html protocols. For
> example .xul and .rss. It would be nice if each extension would have a
> different transformation path.
>

That's an interesting idea... I think there was also a thread recently
re: Localization that made the point that the Compiler pipeline wasn't
as flexible as it could be? Begins at:

http://rubyforge.org/pipermail/nitro-general/2006-February/002917.html

Perhaps if the Compiler pipeline could be enhance somehow, it'd make
your template extension idea possible?

> == Og
>
> = No custom types
>
> Right now, it is not possible to add custom types into Og easily. Let's say I
> want to store a File. I don't want to create a new class called WebFile. I
> would like to tell Og how it should store and restore it.
>

Ah, Og Revisited. I'll have to re-read that. :)

> == Enhancements
>
> = stored settings
>
> It would be nice if this would be possible. Setting already supports a yaml
> export. This could be generalized.
>
> == End
>
> Don't forget, this is only my opinion. I can be wrong and am subject to
> changes too.
>
> I have other griefs but this is enough for today :-p
>

I hope tomorrow is better. :)



From transfire at gmail.com  Fri Mar  3 20:25:51 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 4 Mar 2006 01:25:51 +0000
Subject: [Nitro] [flame] Bad things in Nitro and Og
In-Reply-To: <bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
References: <200603031823.26197.zimba.tm@gmail.com>
	<bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
Message-ID: <4b6f054f0603031725i423fa811v139b203c748418c8@mail.gmail.com>

On 3/3/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> It also iterates through ObjectSpace
> looking for classes to manage, so there's alot to do.

Just have to say that this is kind of silly. I mean, is it really so
hard to tell Nitro explitly which classes? IMHO, sometimes the magic
is too much.



From transfire at gmail.com  Fri Mar  3 20:25:51 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 4 Mar 2006 01:25:51 +0000
Subject: [Nitro] [flame] Bad things in Nitro and Og
In-Reply-To: <bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
References: <200603031823.26197.zimba.tm@gmail.com>
	<bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
Message-ID: <4b6f054f0603031725i423fa811v139b203c748418c8@mail.gmail.com>

On 3/3/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> It also iterates through ObjectSpace
> looking for classes to manage, so there's alot to do.

Just have to say that this is kind of silly. I mean, is it really so
hard to tell Nitro explitly which classes? IMHO, sometimes the magic
is too much.



From transfire at gmail.com  Fri Mar  3 20:25:51 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 4 Mar 2006 01:25:51 +0000
Subject: [Nitro] [flame] Bad things in Nitro and Og
In-Reply-To: <bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
References: <200603031823.26197.zimba.tm@gmail.com>
	<bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
Message-ID: <4b6f054f0603031725i423fa811v139b203c748418c8@mail.gmail.com>

On 3/3/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> It also iterates through ObjectSpace
> looking for classes to manage, so there's alot to do.

Just have to say that this is kind of silly. I mean, is it really so
hard to tell Nitro explitly which classes? IMHO, sometimes the magic
is too much.



From transfire at gmail.com  Fri Mar  3 20:25:51 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 4 Mar 2006 01:25:51 +0000
Subject: [Nitro] [flame] Bad things in Nitro and Og
In-Reply-To: <bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
References: <200603031823.26197.zimba.tm@gmail.com>
	<bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
Message-ID: <4b6f054f0603031725i423fa811v139b203c748418c8@mail.gmail.com>

On 3/3/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> It also iterates through ObjectSpace
> looking for classes to manage, so there's alot to do.

Just have to say that this is kind of silly. I mean, is it really so
hard to tell Nitro explitly which classes? IMHO, sometimes the magic
is too much.



From bryan.a.soto at gmail.com  Sat Mar  4 02:13:06 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 3 Mar 2006 23:13:06 -0800
Subject: [Nitro] PATCH: mongrel-mime-type-fixes
Message-ID: <bd93db0d0603032313q5788d498yd1fa319af0830b5@mail.gmail.com>

Thanks to manveru for doing the detective work on this one.

Attached adds proper content type handling to the Mongrel adapter and
fixes a bug where it was reading files as Ascii. Rather problematic
when serving graphics.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mongrel-mime-type-fixes.zip
Type: application/zip
Size: 10114 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060304/ab98ab95/attachment.zip 

From m.fellinger at gmail.com  Sat Mar  4 02:22:53 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Sat, 4 Mar 2006 16:22:53 +0900
Subject: [Nitro] PATCH: mongrel-mime-type-fixes
In-Reply-To: <bd93db0d0603032313q5788d498yd1fa319af0830b5@mail.gmail.com>
References: <bd93db0d0603032313q5788d498yd1fa319af0830b5@mail.gmail.com>
Message-ID: <9c00d3e00603032322t23e312fcyb48c7f2286a05c3f@mail.gmail.com>

Hey Bryan,
thanks for the patch, i really begin to love the new bug-tracker, it'sa perfect way to gain overview and making tickets is a charm.We just have to make sure everybody uses it from now on... also weshould move the remaining bugs from rubyforge... if there is anythingleft
~~~~manveru
On 3/4/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:> Thanks to manveru for doing the detective work on this one.>> Attached adds proper content type handling to the Mongrel adapter and> fixes a bug where it was reading files as Ascii. Rather problematic> when serving graphics.>>> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>>>


From bryan.a.soto at gmail.com  Sat Mar  4 02:33:39 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 3 Mar 2006 23:33:39 -0800
Subject: [Nitro] PATCH: mongrel-mime-type-fixes
In-Reply-To: <9c00d3e00603032322t23e312fcyb48c7f2286a05c3f@mail.gmail.com>
References: <bd93db0d0603032313q5788d498yd1fa319af0830b5@mail.gmail.com>
	<9c00d3e00603032322t23e312fcyb48c7f2286a05c3f@mail.gmail.com>
Message-ID: <bd93db0d0603032333i2eca0898r655ab3e95739c8fa@mail.gmail.com>

On 3/3/06, Michael Fellinger <m.fellinger at gmail.com> wrote:
> Hey Bryan,
> thanks for the patch, i really begin to love the new bug-tracker, it'sa perfect way to gain overview and making tickets is a charm.We just have to make sure everybody uses it from now on... also weshould move the remaining bugs from rubyforge... if there is anythingleft

I added everything recent, which was stuff I had entered. I think
there's a couple of old ones (more than 6 months) I didn't bother
with.

Mongrel does seem to need a bit more work though. It doesn't seem to
be handling cookies, I think. At least trying to log on with the blog
example doesn't do anything.



From bryan.a.soto at gmail.com  Sat Mar  4 03:15:15 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 00:15:15 -0800
Subject: [Nitro] PATCH: blog-example-comment-fix
Message-ID: <bd93db0d0603040015j2e124de2je60225939fd11733@mail.gmail.com>

Attached makes the blog example work and changes some time format
strings so that time and dates actually output.

-----

Problem is outlined at http://devlab.oree.ch/trac/nitrohq/ticket/7 but
basically amounts to this. A template file was named comments.xhtml.
In the dispatcher, while working on the request /comments/save, it
called klass.respond_to_action_or_template 'comments'. Because there
was a template named comments, it considered that a valid action and
tried to run the template, which made all the instance variables nil.
Thus causing the error.

My solution was to rename the file from comments.xhtml to
entry_comments.xhtml. I'm not sure if that's the best solution though.



From bryan.a.soto at gmail.com  Sat Mar  4 03:16:18 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 00:16:18 -0800
Subject: [Nitro] PATCH: blog-example-comment-fix
In-Reply-To: <bd93db0d0603040015j2e124de2je60225939fd11733@mail.gmail.com>
References: <bd93db0d0603040015j2e124de2je60225939fd11733@mail.gmail.com>
Message-ID: <bd93db0d0603040016x7e19dc5ei4c19e49e95b9ac63@mail.gmail.com>

And at least once a week, I have to do this. Again with the attachment.

On 3/4/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Attached makes the blog example work and changes some time format
> strings so that time and dates actually output.
>
> -----
>
> Problem is outlined at http://devlab.oree.ch/trac/nitrohq/ticket/7 but
> basically amounts to this. A template file was named comments.xhtml.
> In the dispatcher, while working on the request /comments/save, it
> called klass.respond_to_action_or_template 'comments'. Because there
> was a template named comments, it considered that a valid action and
> tried to run the template, which made all the instance variables nil.
> Thus causing the error.
>
> My solution was to rename the file from comments.xhtml to
> entry_comments.xhtml. I'm not sure if that's the best solution though.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: blog-example-comment-fix.zip
Type: application/zip
Size: 10603 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060304/f27223c0/attachment.zip 

From bryan.a.soto at gmail.com  Sat Mar  4 03:28:45 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 00:28:45 -0800
Subject: [Nitro] PATCH: scaffolding-classes-defaults
Message-ID: <bd93db0d0603040028n7acbe1b3u6e72e01975f44f33@mail.gmail.com>

When scaffolding a class, if options have already been specified, they
should be updated. Current behaviour uses defaults.

-----

The specific context of this was someone on irc who was specifying
plural names for scaffolded classes and using part/admin. When
part/admin was called, it didn't specify options, and the defaults
overwrote what was previously specified.



From bryan.a.soto at gmail.com  Sat Mar  4 03:29:30 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 00:29:30 -0800
Subject: [Nitro] PATCH: scaffolding-classes-defaults
In-Reply-To: <bd93db0d0603040028n7acbe1b3u6e72e01975f44f33@mail.gmail.com>
References: <bd93db0d0603040028n7acbe1b3u6e72e01975f44f33@mail.gmail.com>
Message-ID: <bd93db0d0603040029u60c648cbgedbbd75adaec68fc@mail.gmail.com>

Twice. Can you tell it's late for me?

On 3/4/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> When scaffolding a class, if options have already been specified, they
> should be updated. Current behaviour uses defaults.
>
> -----
>
> The specific context of this was someone on irc who was specifying
> plural names for scaffolded classes and using part/admin. When
> part/admin was called, it didn't specify options, and the defaults
> overwrote what was previously specified.
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: scaffolding-classes-defaults.zip
Type: application/zip
Size: 9943 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060304/e6d95fb2/attachment.zip 

From bryan.a.soto at gmail.com  Sat Mar  4 03:58:39 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 00:58:39 -0800
Subject: [Nitro] PATCH: mysql-using-sqlutils
Message-ID: <bd93db0d0603040058k515182b5w2966211ad3fd547d@mail.gmail.com>

Attached removes an unnecessary redefinition of quote that is actually
inheirited from SqlUtils.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mysql-using-sqlutils.zip
Type: application/zip
Size: 10052 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060304/246d4770/attachment.zip 

From bryan.a.soto at gmail.com  Sat Mar  4 04:11:51 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 01:11:51 -0800
Subject: [Nitro] Current status.
Message-ID: <bd93db0d0603040111q3459e749iea0baff349480c26@mail.gmail.com>

Okay, I show all tests passing with Nitro and Glue. Og has these problems:

  1) Error:
test_all(TC_Cacheable):
Errno::EINVAL: Invalid argument -
C:/DOCUME~1/Bryan/LOCALS~1/Temp/nitro_file_cache/cache/TC_Cacheable::User:1
    K:/official/glue/lib/glue/cache/file.rb:58:in `initialize'
    K:/official/glue/lib/glue/cache/file.rb:58:in `encode_file'
    K:/official/glue/lib/glue/cache/file.rb:21:in `set'
    ./test/og/../../lib/glue/cacheable.rb:100:in `[]'
    ./test/og/tc_cacheable.rb:51:in `test_all'
    ./test/og/tc_cacheable.rb:44:in `test_all'

  2) Failure:
test_all(TestCaseOgOrderable) [./test/og/mixin/tc_orderable.rb:75]:
<2> expected but was
<1>.

  3) Failure:
test_all(TestOgTaggable) [./test/og/mixin/tc_taggable.rb:47]:
<3> expected but was
<0>.

examples\blog works with the patch I submitted.

So we still need to verify:
examples\flash
examples\flickr
examples\gallery
examples\hello
examples\stream
examples\tiny
examples\wee
examples\why_wiki
spark
flare

Though, perhaps some of the examples aren't useful anymore?

I'll check in tomorrow, later today actually, and continue with
whatever is still outstanding.

Most of the tickets on devlab are cleared up. I've been concentrating
on those. Though that annoying float bug is still outstanding.

Oh, and George, I sent an email to the list with the output from rcov.
It was held for moderator approval due to a size restriction.
Apparently zip doesn't compress very well.

Have fun.



From george.moschovitis at gmail.com  Sat Mar  4 05:25:59 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 4 Mar 2006 12:25:59 +0200
Subject: [Nitro] About the new Control system
In-Reply-To: <200603031850.10543.zimba.tm@gmail.com>
References: <200603021258.39943.zimba.tm@gmail.com>
	<afd80b540603030843r3b3866afjf6f17b9043055394@mail.gmail.com>
	<200603031850.10543.zimba.tm@gmail.com>
Message-ID: <afd80b540603040225n7958efedi455f0e38de1a67ca@mail.gmail.com>

> Maybe, I never understood how Element work exactly. How do you pass ruby
> values to it that are not strings ?


Here is an example:

module Nitro::Element
  class Menu
     def render
        %{
        Hello \#{#{target}.name} | <a
href="profile/\#{#{target}.name}">show profile</a>
        }
     end
   end
end

and use it like this:

<?r user = session[:USER] ?>
...
<Menu target="user" />
...


this is a simple example I just came up with, but I guess you can see
how it is possible
to create Elements that handle run-time parameters.

George.



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Sat Mar  4 05:41:43 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 4 Mar 2006 12:41:43 +0200
Subject: [Nitro] Current status.
In-Reply-To: <bd93db0d0603040111q3459e749iea0baff349480c26@mail.gmail.com>
References: <bd93db0d0603040111q3459e749iea0baff349480c26@mail.gmail.com>
Message-ID: <afd80b540603040241v6765e933nf00070004f209dae@mail.gmail.com>

Bryan,

*thanks* for your *excellent* work!

-g.

On 3/4/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Okay, I show all tests passing with Nitro and Glue. Og has these problems:
>
>   1) Error:
> test_all(TC_Cacheable):
> Errno::EINVAL: Invalid argument -
> C:/DOCUME~1/Bryan/LOCALS~1/Temp/nitro_file_cache/cache/TC_Cacheable::User:1
>     K:/official/glue/lib/glue/cache/file.rb:58:in `initialize'
>     K:/official/glue/lib/glue/cache/file.rb:58:in `encode_file'
>     K:/official/glue/lib/glue/cache/file.rb:21:in `set'
>     ./test/og/../../lib/glue/cacheable.rb:100:in `[]'
>     ./test/og/tc_cacheable.rb:51:in `test_all'
>     ./test/og/tc_cacheable.rb:44:in `test_all'
>
>   2) Failure:
> test_all(TestCaseOgOrderable) [./test/og/mixin/tc_orderable.rb:75]:
> <2> expected but was
> <1>.
>
>   3) Failure:
> test_all(TestOgTaggable) [./test/og/mixin/tc_taggable.rb:47]:
> <3> expected but was
> <0>.
>
> examples\blog works with the patch I submitted.
>
> So we still need to verify:
> examples\flash
> examples\flickr
> examples\gallery
> examples\hello
> examples\stream
> examples\tiny
> examples\wee
> examples\why_wiki
> spark
> flare
>
> Though, perhaps some of the examples aren't useful anymore?
>
> I'll check in tomorrow, later today actually, and continue with
> whatever is still outstanding.
>
> Most of the tickets on devlab are cleared up. I've been concentrating
> on those. Though that annoying float bug is still outstanding.
>
> Oh, and George, I sent an email to the list with the output from rcov.
> It was held for moderator approval due to a size restriction.
> Apparently zip doesn't compress very well.
>
> Have fun.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From aidan at yoyo.org  Sat Mar  4 07:42:09 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Sat, 4 Mar 2006 23:42:09 +1100
Subject: [Nitro] Tutorial?
Message-ID: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>

Hi all,

A wee while back I posted an Og Tutorial (http://www.infurious.com/ 
downloads/) and asked for feedback and suggestions for part two.   
I've not really had any suggestions for the second part, and very  
little feedback also :-)

I'm looking at the results of Bryan's documentation survey, and the  
most requested thing (after a screencast) is a simple wiki tutorial  
for Nitro.  If no-one comes up with any better suggestions, I'm going  
to spend the next few days writing that tutorial.

Aidan
---
http://www.infurious.com




From zimba.tm at gmail.com  Sat Mar  4 09:27:09 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Sat, 4 Mar 2006 15:27:09 +0100
Subject: [Nitro] About the new Control system
In-Reply-To: <afd80b540603040225n7958efedi455f0e38de1a67ca@mail.gmail.com>
References: <200603021258.39943.zimba.tm@gmail.com>
	<200603031850.10543.zimba.tm@gmail.com>
	<afd80b540603040225n7958efedi455f0e38de1a67ca@mail.gmail.com>
Message-ID: <200603041527.09635.zimba.tm@gmail.com>

On Saturday 04 March 2006 11:25, George Moschovitis wrote:
> > Maybe, I never understood how Element work exactly. How do you pass ruby
> > values to it that are not strings ?
>
> Here is an example:
>
> module Nitro::Element
>   class Menu
>      def render
>         %{
>         Hello \#{#{target}.name} | <a
> href="profile/\#{#{target}.name}">show profile</a>
>         }
>      end
>    end
> end
>
> and use it like this:
>
> <?r user = session[:USER] ?>
> ...
> <Menu target="user" />
> ...
>
>
> this is a simple example I just came up with, but I guess you can see
> how it is possible
> to create Elements that handle run-time parameters.

Ok i get it.

" \#{#{target}.name}" is really not practical. Couldn't we find a better way ?

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From mneumann at ntecs.de  Sat Mar  4 10:15:46 2006
From: mneumann at ntecs.de (Michael Neumann)
Date: Sat, 4 Mar 2006 16:15:46 +0100
Subject: [Nitro] About the new Control system
In-Reply-To: <200603041527.09635.zimba.tm@gmail.com>
References: <200603021258.39943.zimba.tm@gmail.com>
	<200603031850.10543.zimba.tm@gmail.com>
	<afd80b540603040225n7958efedi455f0e38de1a67ca@mail.gmail.com>
	<200603041527.09635.zimba.tm@gmail.com>
Message-ID: <4E9828BE-2D68-4592-809D-8035039CEB77@ntecs.de>


Am 04.03.2006 um 15:27 schrieb zimba.tm:

> On Saturday 04 March 2006 11:25, George Moschovitis wrote:
>>> Maybe, I never understood how Element work exactly. How do you  
>>> pass ruby
>>> values to it that are not strings ?
>>
>> Here is an example:
>>
>> module Nitro::Element
>>   class Menu
>>      def render
>>         %{
>>         Hello \#{#{target}.name} | <a
>> href="profile/\#{#{target}.name}">show profile</a>
>>         }
>>      end
>>    end
>> end
>>
>> and use it like this:
>>
>> <?r user = session[:USER] ?>
>> ...
>> <Menu target="user" />
>> ...
>>
>>
>> this is a simple example I just came up with, but I guess you can see
>> how it is possible
>> to create Elements that handle run-time parameters.
>
> Ok i get it.
>
> " \#{#{target}.name}" is really not practical. Couldn't we find a  
> better way ?

Maybe

"#{var :target}"

where var() is a method.

Regards,

   Michael


From bryan.a.soto at gmail.com  Sun Mar  5 01:30:03 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 22:30:03 -0800
Subject: [Nitro] PATCH: examples-ui-patch
Message-ID: <bd93db0d0603042230u3e56cd51n53772506c6e5b3ed@mail.gmail.com>

Attached adds some begin/rescues to examples that require external
componets like wee, rmagick, etc.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: examples-ui-patch.zip
Type: application/zip
Size: 10090 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060305/fff596e9/attachment.zip 

From bryan.a.soto at gmail.com  Sun Mar  5 01:34:43 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 22:34:43 -0800
Subject: [Nitro] PATCH: stringio-patch-modification
Message-ID: <bd93db0d0603042234o50808694kcc148753ddc529c7@mail.gmail.com>

Attached modifies my earlier patch that broke the examples with file uploads.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: stringio-patch-modification.zip
Type: application/zip
Size: 9925 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060305/d30931fa/attachment.zip 

From bryan.a.soto at gmail.com  Sun Mar  5 02:33:46 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 23:33:46 -0800
Subject: [Nitro] Current status.
In-Reply-To: <bd93db0d0603040111q3459e749iea0baff349480c26@mail.gmail.com>
References: <bd93db0d0603040111q3459e749iea0baff349480c26@mail.gmail.com>
Message-ID: <bd93db0d0603042333p6613cf66j336abf1525eb86aa@mail.gmail.com>

On 3/4/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:

Re: Og test suite, I believe the first error is a permission problem.
I seem to recall getting it on my Linux box because the directory was
owned by Apache. Can someone verify that? The other two though are
real failures.

I've verified these examples:

> examples\blog works with the patch I submitted.
> examples\flickr
> examples\gallery works with the patch I submitted.
> examples\hello
> examples\tiny works with the patch I submitted.
> examples\wee
> examples\why_wiki
> spark
> flare

I can't do this one:
> examples\flash # can't install ming from Rubyforge. This WinXP box has no C compiler

And this one is broken. Cgi streaming doesn't work.
> examples\stream

One new ticket:

http://devlab.oree.ch/trac/nitrohq/ticket/9 submitted. Patch to be sent shortly.

There's still that annoying float bug. I haven't been able to track
that one down.

> Oh, and George, I sent an email to the list with the output from rcov.
> It was held for moderator approval due to a size restriction.
> Apparently zip doesn't compress very well.

Just a ping on this one. It can wait till after the release obviously.



From bryan.a.soto at gmail.com  Sun Mar  5 02:50:19 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 23:50:19 -0800
Subject: [Nitro] Current status.
In-Reply-To: <bd93db0d0603042333p6613cf66j336abf1525eb86aa@mail.gmail.com>
References: <bd93db0d0603040111q3459e749iea0baff349480c26@mail.gmail.com>
	<bd93db0d0603042333p6613cf66j336abf1525eb86aa@mail.gmail.com>
Message-ID: <bd93db0d0603042350w283188bem1b2b718f9aab2b99@mail.gmail.com>

> One new ticket:
>
> http://devlab.oree.ch/trac/nitrohq/ticket/9 submitted. Patch to be sent shortly.

Skip that. I updated the ticket. I think this one should wait till
after the release.



From bryan.a.soto at gmail.com  Sun Mar  5 02:56:40 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 4 Mar 2006 23:56:40 -0800
Subject: [Nitro] nitro rake test
In-Reply-To: <afd80b540603030044ue25d9f6icf4245d6095fcbeb@mail.gmail.com>
References: <afd80b540603030044ue25d9f6icf4245d6095fcbeb@mail.gmail.com>
Message-ID: <bd93db0d0603042356p34b3d7d2gbadacc1348ce6cae@mail.gmail.com>

Are you still getting this after my patch?

On 3/3/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Dear devs,
>
> when I run
>
> rake test in the nitro dir I get the following:
>
> D, [2006-03-03T10:42:59.619772 #9744] DEBUG -- : Using Memory sessions.
> /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb:5:in
> ` load': no such file to load -- test/nitro/tc_session.rb (LoadError)
>         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> der.rb:5
>         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> der.rb:5
> rake aborted!
>
> and the file test/nitro/tc_session.rb *exists*. Anyone has any ideas
> what is causing this?
>
> thanks,
> George.
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From george.moschovitis at gmail.com  Sun Mar  5 03:35:56 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 5 Mar 2006 09:35:56 +0100
Subject: [Nitro] Tutorial?
In-Reply-To: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
References: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
Message-ID: <afd80b540603050035p7ba57dael22be1d6079b0ee1f@mail.gmail.com>

I am sure *everyone* would love to see your tutorial! Go ahead! ;-)

-g.

On 3/4/06, Aidan Rogers <aidan at yoyo.org> wrote:
> Hi all,
>
> A wee while back I posted an Og Tutorial (http://www.infurious.com/
> downloads/) and asked for feedback and suggestions for part two.
> I've not really had any suggestions for the second part, and very
> little feedback also :-)
>
> I'm looking at the results of Bryan's documentation survey, and the
> most requested thing (after a screencast) is a simple wiki tutorial
> for Nitro.  If no-one comes up with any better suggestions, I'm going
> to spend the next few days writing that tutorial.
>
> Aidan
> ---
> http://www.infurious.com
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Sun Mar  5 03:37:30 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 5 Mar 2006 09:37:30 +0100
Subject: [Nitro] About the new Control system
In-Reply-To: <4E9828BE-2D68-4592-809D-8035039CEB77@ntecs.de>
References: <200603021258.39943.zimba.tm@gmail.com>
	<200603031850.10543.zimba.tm@gmail.com>
	<afd80b540603040225n7958efedi455f0e38de1a67ca@mail.gmail.com>
	<200603041527.09635.zimba.tm@gmail.com>
	<4E9828BE-2D68-4592-809D-8035039CEB77@ntecs.de>
Message-ID: <afd80b540603050037k411841c9m590a7272279d3359@mail.gmail.com>

>
> Maybe
>
> "#{var :target}"
>

*Interesting* idea! let me investigate this...

thanks,
-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Sun Mar  5 03:38:43 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 5 Mar 2006 09:38:43 +0100
Subject: [Nitro] Current status.
In-Reply-To: <bd93db0d0603042333p6613cf66j336abf1525eb86aa@mail.gmail.com>
References: <bd93db0d0603040111q3459e749iea0baff349480c26@mail.gmail.com>
	<bd93db0d0603042333p6613cf66j336abf1525eb86aa@mail.gmail.com>
Message-ID: <afd80b540603050038n464224fex7badd72b16017ade@mail.gmail.com>

> And this one is broken. Cgi streaming doesn't work.
> > examples\stream

will investigate...

> There's still that annoying float bug. I haven't been able to track
> that one down.

oh will fix that, dont worry ;-)

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Sun Mar  5 03:39:15 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 5 Mar 2006 09:39:15 +0100
Subject: [Nitro] nitro rake test
In-Reply-To: <bd93db0d0603042356p34b3d7d2gbadacc1348ce6cae@mail.gmail.com>
References: <afd80b540603030044ue25d9f6icf4245d6095fcbeb@mail.gmail.com>
	<bd93db0d0603042356p34b3d7d2gbadacc1348ce6cae@mail.gmail.com>
Message-ID: <afd80b540603050039p105db54dvd44d073d3c485d4f@mail.gmail.com>

Will try this a bit later and let you know...

keep sending those patches man ;-)

-g.

On 3/5/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Are you still getting this after my patch?
>
> On 3/3/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Dear devs,
> >
> > when I run
> >
> > rake test in the nitro dir I get the following:
> >
> > D, [2006-03-03T10:42:59.619772 #9744] DEBUG -- : Using Memory sessions.
> > /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb:5:in
> > ` load': no such file to load -- test/nitro/tc_session.rb (LoadError)
> >         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> > der.rb:5
> >         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> > der.rb:5
> > rake aborted!
> >
> > and the file test/nitro/tc_session.rb *exists*. Anyone has any ideas
> > what is causing this?
> >
> > thanks,
> > George.
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From kashia at vfemail.net  Sun Mar  5 06:37:44 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sun, 05 Mar 2006 12:37:44 +0100
Subject: [Nitro] Tutorial?
In-Reply-To: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
References: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
Message-ID: <op.s5xuk6swg3glhz@jo.local>

Hi,

> A wee while back I posted an Og Tutorial (http://www.infurious.com/
> downloads/) and asked for feedback and suggestions for part two.
> I've not really had any suggestions for the second part, and very
> little feedback also :-)

Finally got around to reading your first tutorial, I think it's a
very good introduction for people who already know about databases,
web-frameworks and Ruby.
For real newbies I think it'd be a bit heavy at first, the theme of
the application you're building isn't really friendly to them I
think.

But overall, I think it is a quite good instruction for someone
who is already into the business and wants to learn a web-framework
or wants to switch from Rails.

I think we have to place this tutorial online as html, so one can
look at it without downloading a zip file, which may keep some
people from looking at it.

For the second part I would suggest embedding that Og model in a
Nitro application.
I really liked the pictures from the first part, they are a really
nice idea, we need more diagrams :D

Kash

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From nusgnaf at gmail.com  Sun Mar  5 08:52:04 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Sun, 5 Mar 2006 13:52:04 +0000
Subject: [Nitro] [Bug] Memory based session cache does not release memory
Message-ID: <716700c90603050552u673ab3abqa61f7e6da537627f@mail.gmail.com>

Hi, devs on the list. I develop using scgi+lighttpd configuration,
with memory sessions. I use ab from apache package to benchmark. find
the scgi session continuously eat up memory. On a P4 630 machine with
1gb memory, a 20,0000 run consumes almost 200Mb memory. I setup in
conf/live.rb the following parameter:

Nitro::Session.cookie_expires = true
Nitro::Session.keepalive = 2.minutes

and hook some call back like

Nitro::Session.cache.hash.size
Nitro::Session.cache.gc!
Nitro::Session.cache.hash.size

I can see the hash size of  @hash variable of Nitro::Session.cache
shrink from several thounds of to 0 as time pass the keepalive time,
but the ruby consumed memory doesn't shrink. Any better way to debug
the session memory leak?



From bryan.a.soto at gmail.com  Sun Mar  5 12:35:11 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 5 Mar 2006 09:35:11 -0800
Subject: [Nitro] PATCH: Fwd:  Og : Ez clause with arrays
Message-ID: <bd93db0d0603050935l4f5da21dm7217fd885b5fd729@mail.gmail.com>

Hi George,

Did you see this one?

---------- Forwarded message ----------
From: bicou <rf at bicou.com>
Date: Mar 2, 2006 12:16 AM
Subject: [Nitro] Og : Ez clause with arrays
To: nitro-general at rubyforge.org


Found a bug with the ez === operator, something like :

topics = Topic.find { |t|
       t.name === ['NAME1','NAME2','NAME3']
}

would be written as :

SELECT [...] WHERE name IN (---
- NAME1
- NAME2
... (yaml is escaping the array)


I've made a patch with a new prepared statement like operator :

WHERE name IN (?*)

with ?* replaced with the array as a list of quoted values
instead of the yaml dump.

patch attached

_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ez-where-in-array.diff
Type: application/octet-stream
Size: 1437 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060305/9bacc340/attachment.obj 

From bryan.a.soto at gmail.com  Sun Mar  5 13:20:51 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 5 Mar 2006 10:20:51 -0800
Subject: [Nitro] nitro rake test
In-Reply-To: <afd80b540603050039p105db54dvd44d073d3c485d4f@mail.gmail.com>
References: <afd80b540603030044ue25d9f6icf4245d6095fcbeb@mail.gmail.com>
	<bd93db0d0603042356p34b3d7d2gbadacc1348ce6cae@mail.gmail.com>
	<afd80b540603050039p105db54dvd44d073d3c485d4f@mail.gmail.com>
Message-ID: <bd93db0d0603051020q17b831e0yfcf7b93f64fd95b5@mail.gmail.com>

I'm just curious, because I ran into those same errors while running
rcov. I'm not exactly sure why they stopped...

On 3/5/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Will try this a bit later and let you know...
>
> keep sending those patches man ;-)
>
> -g.
>
> On 3/5/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> > Are you still getting this after my patch?
> >
> > On 3/3/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > > Dear devs,
> > >
> > > when I run
> > >
> > > rake test in the nitro dir I get the following:
> > >
> > > D, [2006-03-03T10:42:59.619772 #9744] DEBUG -- : Using Memory sessions.
> > > /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb:5:in
> > > ` load': no such file to load -- test/nitro/tc_session.rb (LoadError)
> > >         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> > > der.rb:5
> > >         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> > > der.rb:5
> > > rake aborted!
> > >
> > > and the file test/nitro/tc_session.rb *exists*. Anyone has any ideas
> > > what is causing this?
> > >
> > > thanks,
> > > George.
> > >
> > >
> > > --
> > > http://www.gmosx.com
> > > http://www.navel.gr
> > > http://www.nitrohq.com
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From bryan.a.soto at gmail.com  Sun Mar  5 13:24:36 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 5 Mar 2006 10:24:36 -0800
Subject: [Nitro] Fwd: rcov output.
In-Reply-To: <bd93db0d0603031620qb8496a8me5e150e2419ff7b0@mail.gmail.com>
References: <bd93db0d0603031620qb8496a8me5e150e2419ff7b0@mail.gmail.com>
Message-ID: <bd93db0d0603051024k6fb109d0h173f55cc12b1d0e0@mail.gmail.com>

I'm resending this as a tar.bz2. To Windows users, there is a zip
archive, but it's currently held up in moderation due to it's size.

Sorry.

---------- Forwarded message ----------
From: Bryan Soto <bryan.a.soto at gmail.com>
Date: Mar 3, 2006 4:20 PM
Subject: rcov output.
To: General discussion about Nitro <nitro-general at rubyforge.org>


Hi all,

Attached is the output of rcov* as run with the magical incantation:

/mnt/zip/official $ RUBYOPT='-rubygems
-r/mnt/zip/official/glycerin.rb' rcov -o /mnt/zip/official/coverage
`find glue/test -name tc\*.rb` `find og/test -name tc\*.rb` `find
nitro/test -name tc\*.rb`

This is run against the repo.nitrohq.com and has all the latest
patches. It should tell us how well everything is covered by tests.
But in Mauricio's own words, "Code coverage shouldn't be abused (in
few words, C0 coverage guarantees nothing) but it's still useful for
testing: it will at least tell you when your tests need more work, and
most importantly where."

It is also hosted at http://devlab.oree.ch/~bryan/coverage/index.html
but to save zimba's bandwith, I ask that if you intend to use this as
a guide for writing tests, that you download the zip and view it
locally. Or perhaps someone else can host it?

Thanks,

Bryan

* Ruby code coverage by Mauricio Fernandez.
  http://www.eigenclass.org/hiki.rb?rcov
-------------- next part --------------
A non-text attachment was scrubbed...
Name: coverage.tar.bz2
Type: application/x-bzip2
Size: 166157 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060305/82ae9283/attachment.bz2 

From bryan.a.soto at gmail.com  Sun Mar  5 14:22:55 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 5 Mar 2006 11:22:55 -0800
Subject: [Nitro] PATCH: test-fix-tc_cacheable
Message-ID: <bd93db0d0603051122o1fa95970s6804e9dcf470b8b5@mail.gmail.com>

  * test-fix-tc_cacheable
  Fixes to File cache, files with a colon (:) are illegal on win32.
  Fixes to tc_cacheable, sepearates each cache type test into it's own class as
they wouldn't run otherwise.

-----

Also makes file cache be stored in a subdirectory of test so it avoids
permission errors on *nix. This directory is removed in setup.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test-fix-tc_cacheable.zip
Type: application/zip
Size: 10781 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060305/dcdd5064/attachment.zip 

From manveru at weez.co.jp  Sun Mar  5 21:13:52 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Mon, 6 Mar 2006 11:13:52 +0900
Subject: [Nitro] Tutorial?
In-Reply-To: <op.s5xuk6swg3glhz@jo.local>
References: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
	<op.s5xuk6swg3glhz@jo.local>
Message-ID: <200603061113.52193.manveru@weez.co.jp>

Hey,

Regarding the 'more friendly theme' i would like to propose a guestbook - i 
think it's one of the things newcomers really would like to do, and it is a 
charm to do in nitro.
Basically it's like a perfect sunday-cake where you use nitro instead of germ, 
wich makes it baking super-fast ;)

Anyway, will hack together something like that (hopefully) today. Still we 
need to finish all the other tutorials that lie around at nitrohq - like my 
'Nitro in Flames', wich was quite promising but never found an end.

Also something we should consider, is documenting the examples more, making 
them to tutorials (wich is part of why they are here at all).
We have a screencast for the FlicR one, if somebody could show me how to do 
screencasts i would make another one for the gallery-example or maybe for 
that guestbook i spoke of.

Anyway, it would be nice if we would make more use of oxyliquit 
( http://oxyliquit.manveru.net ), Kashia and Oggu_ really invested some days 
into it and it would be sweet to see it actually used in the real world - 
just give it a try and sign up, add some Tips/Questions and visit it once in 
a while. I think it could become a real base to introduce newcomers and guide 
them through the first troubles. Also it adds some (kind of) voting (to the 
votes we already have) for wich topics should be covered by documentation 
and/or examples/screencasts/tutorials/howtos/whatever...

~~~~manveru

On Sunday 05 March 2006 20:37, Kashia Buch wrote:
> Hi,
>
> > A wee while back I posted an Og Tutorial (http://www.infurious.com/
> > downloads/) and asked for feedback and suggestions for part two.
> > I've not really had any suggestions for the second part, and very
> > little feedback also :-)
>
> Finally got around to reading your first tutorial, I think it's a
> very good introduction for people who already know about databases,
> web-frameworks and Ruby.
> For real newbies I think it'd be a bit heavy at first, the theme of
> the application you're building isn't really friendly to them I
> think.
>
> But overall, I think it is a quite good instruction for someone
> who is already into the business and wants to learn a web-framework
> or wants to switch from Rails.
>
> I think we have to place this tutorial online as html, so one can
> look at it without downloading a zip file, which may keep some
> people from looking at it.
>
> For the second part I would suggest embedding that Og model in a
> Nitro application.
> I really liked the pictures from the first part, they are a really
> nice idea, we need more diagrams :D
>
> Kash


From bryan.a.soto at gmail.com  Sun Mar  5 23:53:59 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 5 Mar 2006 20:53:59 -0800
Subject: [Nitro] PATCH: test-fixes-taggable-override
Message-ID: <bd93db0d0603052053u3546eb30jb862606887a3e4f0@mail.gmail.com>

* test-fixes-taggable-override
Repairs taggable library using tc_taggable as a spec. Also adds a fix
to tc_override, where it wasn't properly restoring the Og.prefix
setting that broke taggable when running in the full suite.

-----

With this, the og test suite is now:

Loaded suite c:/ruby/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader

Started
....................F.....................
Finished in 15.438 seconds.

  1) Failure:
test_all(TestCaseOgOrderable) [./test/og/mixin/tc_orderable.rb:75]:
<2> expected but was
<1>.

42 tests, 203 assertions, 1 failures, 0 errors
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test-fixes-taggable-override.zip
Type: application/zip
Size: 10196 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060305/b714b978/attachment.zip 

From aidan at yoyo.org  Mon Mar  6 00:35:04 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Mon, 6 Mar 2006 16:35:04 +1100
Subject: [Nitro] Patch workflow
Message-ID: <634BA5F0-5A9F-4B18-9F51-A2767D712DBF@yoyo.org>

Can someone verify the sanity of this workflow?  This is for a bug or  
feature being reported in Trac, through to fix -> check-in and  
commitment to the repository.  It's really simple, and gives us a  
reasonable way to reduce things falling through the cracks.

Thanks,

Aidan
---
http://www.infurious.com


-------------- next part --------------
A non-text attachment was scrubbed...
Name: trac.png
Type: image/png
Size: 61852 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060306/820b2a60/attachment.png 

From bryan.a.soto at gmail.com  Mon Mar  6 02:39:50 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 5 Mar 2006 23:39:50 -0800
Subject: [Nitro] PATCH: testfix-orderable
Message-ID: <bd93db0d0603052339t25b8a374u7fb3b7348b70d132@mail.gmail.com>

* testfix-orderable
Repairs orderable to work with test case. Simple fix. It needed an
aspect run on both :og_insert and :og_update. Removed a save from the
test case as it was unnecessary.

-----

And with this, I'm proud to say:

K:\official\glue>rake test
Loaded suite c:/ruby/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader

Started
....................................
Finished in 1.797 seconds.

36 tests, 168 assertions, 0 failures, 0 errors


K:\official\nitro>rake test
Loaded suite c:/ruby/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader

Started
..................................
Finished in 36.984 seconds.

34 tests, 117 assertions, 0 failures, 0 errors


K:\official\og>rake test
Loaded suite c:/ruby/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader

Started
..........................................
Finished in 14.344 seconds.

42 tests, 216 assertions, 0 failures, 0 errors


All tests pass. :)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: testfix-orderable.zip
Type: application/zip
Size: 9958 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060306/e8741329/attachment.zip 

From bryan.a.soto at gmail.com  Mon Mar  6 02:46:22 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 5 Mar 2006 23:46:22 -0800
Subject: [Nitro] Patch workflow
In-Reply-To: <634BA5F0-5A9F-4B18-9F51-A2767D712DBF@yoyo.org>
References: <634BA5F0-5A9F-4B18-9F51-A2767D712DBF@yoyo.org>
Message-ID: <bd93db0d0603052346o5e5ce730pc1b4920814594b6f@mail.gmail.com>

On 3/5/06, Aidan Rogers <aidan at yoyo.org> wrote:
> Can someone verify the sanity of this workflow?  This is for a bug or
> feature being reported in Trac, through to fix -> check-in and
> commitment to the repository.  It's really simple, and gives us a
> reasonable way to reduce things falling through the cracks.
>
> Thanks,

It seems quite sane to me. :)

By the way, what do you use to generate these diagrams? Is it portable
to Windows or *nix? I think sketches would be quite useful for showing
the flow of processing in Nitro or in a general overview of all four
sub-projects.



From bryan.a.soto at gmail.com  Mon Mar  6 03:14:52 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 00:14:52 -0800
Subject: [Nitro] Tutorial?
In-Reply-To: <200603061113.52193.manveru@weez.co.jp>
References: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
	<op.s5xuk6swg3glhz@jo.local> <200603061113.52193.manveru@weez.co.jp>
Message-ID: <bd93db0d0603060014j234995e1mbc31e15d230ab470@mail.gmail.com>

On 3/5/06, Michael Fellinger <manveru at weez.co.jp> wrote:
> Also something we should consider, is documenting the examples more, making
> them to tutorials (wich is part of why they are here at all).
> We have a screencast for the FlicR one, if somebody could show me how to do
> screencasts i would make another one for the gallery-example or maybe for
> that guestbook i spoke of.
>

I think George used vnc2swf. Check out http://www.unixuser.org/~euske/vnc2swf/

> Anyway, it would be nice if we would make more use of oxyliquit
> ( http://oxyliquit.manveru.net ), Kashia and Oggu_ really invested some days
> into it and it would be sweet to see it actually used in the real world -
> just give it a try and sign up, add some Tips/Questions and visit it once in
> a while. I think it could become a real base to introduce newcomers and guide
> them through the first troubles. Also it adds some (kind of) voting (to the
> votes we already have) for wich topics should be covered by documentation
> and/or examples/screencasts/tutorials/howtos/whatever...

I didn't know about this. I've added it to the Trac wiki. Good work all. :)

One problem:
http://oxyliquit.manveru.net/question/1 leads to an error page and
asks me to report it. Please consider it reported. ;)

Maybe it could email questions to the mailing list or provide an RSS
feed? It might help get a quicker response. Especially if the RSS feed
included an answer box.

I've signed up and am now contemplating how to explain what MVC is...
Should be simple, right? ;)



From manveru at weez.co.jp  Mon Mar  6 03:40:20 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Mon, 6 Mar 2006 17:40:20 +0900
Subject: [Nitro] Tutorial?
In-Reply-To: <bd93db0d0603060014j234995e1mbc31e15d230ab470@mail.gmail.com>
References: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
	<200603061113.52193.manveru@weez.co.jp>
	<bd93db0d0603060014j234995e1mbc31e15d230ab470@mail.gmail.com>
Message-ID: <200603061740.20963.manveru@weez.co.jp>

Oh, sorry i did read your post too late.

I had to purge the databse because of an upgrade of oxyliquit (wich caused the 
error you reported). Hope that doesn't bother you too much :)

However, it's up and running again.

~~~~manveru

On Monday 06 March 2006 17:14, Bryan Soto wrote:
> On 3/5/06, Michael Fellinger <manveru at weez.co.jp> wrote:
> > Also something we should consider, is documenting the examples more,
> > making them to tutorials (wich is part of why they are here at all).
> > We have a screencast for the FlicR one, if somebody could show me how to
> > do screencasts i would make another one for the gallery-example or maybe
> > for that guestbook i spoke of.
>
> I think George used vnc2swf. Check out
> http://www.unixuser.org/~euske/vnc2swf/

will try that if i find some time today

>
> > Anyway, it would be nice if we would make more use of oxyliquit
> > ( http://oxyliquit.manveru.net ), Kashia and Oggu_ really invested some
> > days into it and it would be sweet to see it actually used in the real
> > world - just give it a try and sign up, add some Tips/Questions and visit
> > it once in a while. I think it could become a real base to introduce
> > newcomers and guide them through the first troubles. Also it adds some
> > (kind of) voting (to the votes we already have) for wich topics should be
> > covered by documentation and/or
> > examples/screencasts/tutorials/howtos/whatever...
>
> I didn't know about this. I've added it to the Trac wiki. Good work all. :)
>
> One problem:
> http://oxyliquit.manveru.net/question/1 leads to an error page and
> asks me to report it. Please consider it reported. ;)
>
> Maybe it could email questions to the mailing list or provide an RSS
> feed? It might help get a quicker response. Especially if the RSS feed
> included an answer box.
>
> I've signed up and am now contemplating how to explain what MVC is...
> Should be simple, right? ;)
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general


From aidan at yoyo.org  Mon Mar  6 03:59:39 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Mon, 6 Mar 2006 19:59:39 +1100
Subject: [Nitro] Patch workflow
In-Reply-To: <bd93db0d0603052346o5e5ce730pc1b4920814594b6f@mail.gmail.com>
References: <634BA5F0-5A9F-4B18-9F51-A2767D712DBF@yoyo.org>
	<bd93db0d0603052346o5e5ce730pc1b4920814594b6f@mail.gmail.com>
Message-ID: <EB2333A8-26CC-4D6A-89FB-044372B8C1A3@yoyo.org>

On 06/03/2006, at 6:46 PM, Bryan Soto wrote:

> On 3/5/06, Aidan Rogers <aidan at yoyo.org> wrote:
>> Can someone verify the sanity of this workflow?  This is for a bug or
>> feature being reported in Trac, through to fix -> check-in and
>> commitment to the repository.  It's really simple, and gives us a
>> reasonable way to reduce things falling through the cracks.
>>
>> Thanks,
>
> It seems quite sane to me. :)
>
> By the way, what do you use to generate these diagrams? Is it portable
> to Windows or *nix? I think sketches would be quite useful for showing
> the flow of processing in Nitro or in a general overview of all four
> sub-projects.
>

I use OmniGraffle, which is a Mac-only app.  I'm happy to draw as  
many diagrams as necessary to help people understand stuff.  The only  
caveat is someone has to explain it to me first :-)

Aidan
---
http://www.infurious.com




From george.moschovitis at gmail.com  Mon Mar  6 05:38:29 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 6 Mar 2006 12:38:29 +0200
Subject: [Nitro] nitro rake test
In-Reply-To: <bd93db0d0603051020q17b831e0yfcf7b93f64fd95b5@mail.gmail.com>
References: <afd80b540603030044ue25d9f6icf4245d6095fcbeb@mail.gmail.com>
	<bd93db0d0603042356p34b3d7d2gbadacc1348ce6cae@mail.gmail.com>
	<afd80b540603050039p105db54dvd44d073d3c485d4f@mail.gmail.com>
	<bd93db0d0603051020q17b831e0yfcf7b93f64fd95b5@mail.gmail.com>
Message-ID: <afd80b540603060238j71e1fdd2h63af9f1cd584addd@mail.gmail.com>

Rake test still does not work for me :(
I get the same error (tc_session.rb not found)

-g.

On 3/5/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> I'm just curious, because I ran into those same errors while running
> rcov. I'm not exactly sure why they stopped...
>
> On 3/5/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Will try this a bit later and let you know...
> >
> > keep sending those patches man ;-)
> >
> > -g.
> >
> > On 3/5/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> > > Are you still getting this after my patch?
> > >
> > > On 3/3/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > > > Dear devs,
> > > >
> > > > when I run
> > > >
> > > > rake test in the nitro dir I get the following:
> > > >
> > > > D, [2006-03-03T10:42:59.619772 #9744] DEBUG -- : Using Memory sessions.
> > > > /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb:5:in
> > > > ` load': no such file to load -- test/nitro/tc_session.rb (LoadError)
> > > >         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> > > > der.rb:5
> > > >         from /usr/local/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loa
> > > > der.rb:5
> > > > rake aborted!
> > > >
> > > > and the file test/nitro/tc_session.rb *exists*. Anyone has any ideas
> > > > what is causing this?
> > > >
> > > > thanks,
> > > > George.
> > > >
> > > >
> > > > --
> > > > http://www.gmosx.com
> > > > http://www.navel.gr
> > > > http://www.nitrohq.com
> > > >
> > > > _______________________________________________
> > > > Nitro-general mailing list
> > > > Nitro-general at rubyforge.org
> > > > http://rubyforge.org/mailman/listinfo/nitro-general
> > > >
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar  6 05:43:48 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 6 Mar 2006 12:43:48 +0200
Subject: [Nitro] PATCH: testfix-orderable
In-Reply-To: <bd93db0d0603052339t25b8a374u7fb3b7348b70d132@mail.gmail.com>
References: <bd93db0d0603052339t25b8a374u7fb3b7348b70d132@mail.gmail.com>
Message-ID: <afd80b540603060243s6ca4dbfek33858e2ca9da0e5a@mail.gmail.com>

EXCELLENT ;-)


On 3/6/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> * testfix-orderable
> Repairs orderable to work with test case. Simple fix. It needed an
> aspect run on both :og_insert and :og_update. Removed a save from the
> test case as it was unnecessary.
>
> -----
>
> And with this, I'm proud to say:
>
> K:\official\glue>rake test
> Loaded suite c:/ruby/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader
>
> Started
> ....................................
> Finished in 1.797 seconds.
>
> 36 tests, 168 assertions, 0 failures, 0 errors
>
>
> K:\official\nitro>rake test
> Loaded suite c:/ruby/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader
>
> Started
> ..................................
> Finished in 36.984 seconds.
>
> 34 tests, 117 assertions, 0 failures, 0 errors
>
>
> K:\official\og>rake test
> Loaded suite c:/ruby/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader
>
> Started
> ..........................................
> Finished in 14.344 seconds.
>
> 42 tests, 216 assertions, 0 failures, 0 errors
>
>
> All tests pass. :)
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar  6 05:54:04 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 6 Mar 2006 12:54:04 +0200
Subject: [Nitro] Oxyliquit tip
Message-ID: <afd80b540603060254l6bfb66a5jb45b69d213b7795b@mail.gmail.com>

Dear devs,

I had a quick look at glimmer and oxyliquit (2 open source nitro
projects). Here is a first tip:

why dont you use the Mailer functionality to do the validation etc email?

Nitro has excellent support for emails so why reinvent the wheel? ;-)

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar  6 05:56:41 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 6 Mar 2006 12:56:41 +0200
Subject: [Nitro] [flame] Bad things in Nitro and Og
In-Reply-To: <4b6f054f0603031725i423fa811v139b203c748418c8@mail.gmail.com>
References: <200603031823.26197.zimba.tm@gmail.com>
	<bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
	<4b6f054f0603031725i423fa811v139b203c748418c8@mail.gmail.com>
Message-ID: <afd80b540603060256r5d5efcfu7c710fa007b08ad5@mail.gmail.com>

I didnt realize that Nitro is slow at startup. Perhaps my apps have
not many classes or something. Anyway,    the ObjectSpace traversal
thing is easy to fix without giving up the magic. Stay tunned for an
update in 0.30.0

moreover, I will add some more startup time improvements (for example
transformed scripts caching)

but, lets get 0.29.0 out first ;-)


-g.



On 3/4/06, TRANS <transfire at gmail.com> wrote:
> On 3/3/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> > It also iterates through ObjectSpace
> > looking for classes to manage, so there's alot to do.
>
> Just have to say that this is kind of silly. I mean, is it really so
> hard to tell Nitro explitly which classes? IMHO, sometimes the magic
> is too much.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar  6 05:58:05 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 6 Mar 2006 12:58:05 +0200
Subject: [Nitro] PATCH: Fwd: Og : Ez clause with arrays
In-Reply-To: <bd93db0d0603050935l4f5da21dm7217fd885b5fd729@mail.gmail.com>
References: <bd93db0d0603050935l4f5da21dm7217fd885b5fd729@mail.gmail.com>
Message-ID: <afd80b540603060258p33b6430flcdcfbb94353bfa46@mail.gmail.com>

Will add this ;-)

-g.

On 3/5/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Hi George,
>
> Did you see this one?
>
> ---------- Forwarded message ----------
> From: bicou <rf at bicou.com>
> Date: Mar 2, 2006 12:16 AM
> Subject: [Nitro] Og : Ez clause with arrays
> To: nitro-general at rubyforge.org
>
>
> Found a bug with the ez === operator, something like :
>
> topics = Topic.find { |t|
>        t.name === ['NAME1','NAME2','NAME3']
> }
>
> would be written as :
>
> SELECT [...] WHERE name IN (---
> - NAME1
> - NAME2
> ... (yaml is escaping the array)
>
>
> I've made a patch with a new prepared statement like operator :
>
> WHERE name IN (?*)
>
> with ?* replaced with the array as a list of quoted values
> instead of the yaml dump.
>
> patch attached
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From derekmahar.26026218 at bloglines.com  Mon Mar  6 05:59:36 2006
From: derekmahar.26026218 at bloglines.com (derekmahar.26026218 at bloglines.com)
Date: 6 Mar 2006 10:59:36 -0000
Subject: [Nitro] Nitro documentation
Message-ID: <1141642776.717749105.19198.sendItem@bloglines.com>

Where can I find complete and comprehensive Nitro and Ogg documentation? 
I've read an article about Ogg that really piqued my interest, but I don't
know  where to find more information.

Derek


From kashia at vfemail.net  Mon Mar  6 06:40:46 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Mon, 06 Mar 2006 12:40:46 +0100
Subject: [Nitro] Oxyliquit tip
In-Reply-To: <afd80b540603060254l6bfb66a5jb45b69d213b7795b@mail.gmail.com>
References: <afd80b540603060254l6bfb66a5jb45b69d213b7795b@mail.gmail.com>
Message-ID: <op.s5zpd8neg3glhz@jo.local>

Hey George,

> Nitro has excellent support for emails so why reinvent the wheel? ;-)

ATM I don't plan _ANY_ verification for Oxyliquit ;)
I want to make it as easy as possible for new People to enter their
questions, they are logged in right away after registering.

"Easy come, easy go", if people loose their username or password,
either they just register again, or contact me so this gets handled.

BTW: Oxylquit had a feed already built in /feed , enjoy :)


Kash

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From fabian at oggu.de  Mon Mar  6 06:40:26 2006
From: fabian at oggu.de (Fabian Buch)
Date: Mon, 6 Mar 2006 12:40:26 +0100
Subject: [Nitro] Tutorial?
In-Reply-To: <bd93db0d0603060014j234995e1mbc31e15d230ab470@mail.gmail.com>
References: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
	<op.s5xuk6swg3glhz@jo.local>
	<200603061113.52193.manveru@weez.co.jp>
	<bd93db0d0603060014j234995e1mbc31e15d230ab470@mail.gmail.com>
Message-ID: <425755733c85103f070b55c11db74160@oggu.de>


Am 06.03.2006 um 09:14 schrieb Bryan Soto:
> On 3/5/06, Michael Fellinger <manveru at weez.co.jp> wrote:
>> Anyway, it would be nice if we would make more use of oxyliquit
>> ( http://oxyliquit.manveru.net ), Kashia and Oggu_ really invested 
>> some days
>> into it and it would be sweet to see it actually used in the real 
>> world -

> I didn't know about this. I've added it to the Trac wiki. Good work 
> all. :)

The reason we didn't advertise it yet is that we didn't think it's 
feature-complete yet and that it still has annoying bugs. One of them 
being in Nitro (which was the reason for one of the tickets: 
http://devlab.oree.ch/trac/nitrohq/ticket/9), workaround is applied to 
the Nitro on Manveru's server though, so tags like "web2.0" work.

> One problem:
> http://oxyliquit.manveru.net/question/1 leads to an error page and
> asks me to report it. Please consider it reported. ;)

fixed

> Maybe it could email questions to the mailing list or provide an RSS
> feed? It might help get a quicker response. Especially if the RSS feed
> included an answer box.

The latter is on it's way. Should be completely implemented soon, 
whether it's applied to http://oxyliquit.manveru.net today depends on 
Manv.

> I've signed up and am now contemplating how to explain what MVC is...
> Should be simple, right? ;)

Good idea. Should be added as tip.

Fabian (Oggu in IRC)

PS to Manv: if there's content (Questions, Tips) in Nitro Q&A (former 
Oxyliquit), backup the database before dropping..



From guillaume.pierronnet at gmail.com  Mon Mar  6 07:13:36 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Mon, 6 Mar 2006 13:13:36 +0100
Subject: [Nitro] [PATCH] improvements on Glue::FileCache
Message-ID: <6a7d49ca0603060413v3a5aeb69g@mail.gmail.com>

hi list!

* Glue::FileCache improved: escaping bad characters in filenames
* Glue::FileCache: safe locking of files added, making this caching
method safe for multiprocess

Bryan: this patch makes test case tc_session.rb runs well!

regards
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.gz
Type: application/x-gzip
Size: 10079 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060306/8e56f86f/attachment.gz 

From riku.raisanen at walkingwoods.com  Mon Mar  6 07:59:08 2006
From: riku.raisanen at walkingwoods.com (Riku =?iso-8859-1?q?R=E4is=E4nen?=)
Date: Mon, 6 Mar 2006 14:59:08 +0200
Subject: [Nitro] glue::mailer examples
Message-ID: <200603061459.08621.riku.raisanen@walkingwoods.com>

Are there any examples for the usage? and what about sending attachments?

-Riku R?is?nen



From zimba.tm at gmail.com  Mon Mar  6 09:58:10 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Mon, 6 Mar 2006 15:58:10 +0100
Subject: [Nitro] Admin improvements
Message-ID: <200603061558.10912.zimba.tm@gmail.com>

Hi list,

the next thing I would to work on, is to improve the admin. Before I go on, 
here are the ideas I would like to implement :

= Make settings changeable and storable

 * Use the caching system or Og to load/save settings
 * Make the settings modifiable in the web interface
 * Group the settings in different pages, according to their namespace to 
avoid a huge page

= Use rubygems for components

I don't know if it will be straignforward.

 * Use a nitro-specific settings for rubygems (store location, download 
location, ...)
 * Make rubygems (un)installable/uploadable from the web interface
 * Add the ability to instanciate components on specific web paths
 * Upgrade Og so that it supports table installing/upgrading on the fly 
(without relaunching nitro)

= Standalone nitro

Make a nitro executable that will launch nitro with the admin and custom 
options

= Editable

 * Make sources and templates editable from the web interface

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From timh at dirtymonday.net  Mon Mar  6 11:31:13 2006
From: timh at dirtymonday.net (TimH)
Date: Mon, 6 Mar 2006 08:31:13 -0800
Subject: [Nitro] Spark pre text
Message-ID: <20060306083113.7449e959.timh@dirtymonday.net>

For the next version of Spark, can we fix the pre-formatted text coloring by replacing this line in public/style.css

	pre {background: #424242; padding: 10px; overflow: auto;}

with something like this?

	pre {background: #424242; color: #FFF; padding: 10px; overflow: auto;}

--TimH


From vagabond at cataclysm-software.net  Mon Mar  6 11:39:56 2006
From: vagabond at cataclysm-software.net (Andrew Thompson)
Date: Mon, 06 Mar 2006 16:39:56 +0000
Subject: [Nitro] Controller resolution order
Message-ID: <440C65DC.7090200@cataclysm-software.net>

Hi all,

I've been playing with nitro again lately and I have a question about 
how nitro decides which controller to use, for example let's say I have 
this:

Nitro::Server.map= {'/'=> BaseController, 
'/admin/users'=>UserController, '/admin'=>AdminController}

now /admin points to the admin controller, fine. However /admin/users 
ALSO points to the admin controller and looks for a 'users' method on 
it. This seems a bit counter-intuitive, why doesn't Nitro look for a 
controller to match before looking for a method in a controller?

It also does a similar thing for methods with a double underscore (I 
think foo__bar() maps to foo/bar), if I define foo(), and go to foo/bar, 
instead of finding foo__bar(), it passes 'bar' to foo(). This works fine 
if foo() is not defined in the controller however.

The second one isn't such a problem, but the first issue is quite 
annoying...

Vag


From bryan.a.soto at gmail.com  Mon Mar  6 14:36:07 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 11:36:07 -0800
Subject: [Nitro] PATCH: test-fix-tc_session
Message-ID: <bd93db0d0603061136o299aa705o26f9318435ea0169@mail.gmail.com>

* test-fix-tc_session
This must have been the one giving the security error. Fixed.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: test-fix-tc_session.zip
Type: application/zip
Size: 10982 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060306/493ffa5a/attachment.zip 

From george.moschovitis at gmail.com  Mon Mar  6 15:49:58 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 6 Mar 2006 21:49:58 +0100
Subject: [Nitro] [PATCH] improvements on Glue::FileCache
In-Reply-To: <6a7d49ca0603060413v3a5aeb69g@mail.gmail.com>
References: <6a7d49ca0603060413v3a5aeb69g@mail.gmail.com>
Message-ID: <afd80b540603061249r177c80f5t51d12b91cf356222@mail.gmail.com>

thanks, will try this first thing in the morning ;-)
most probably we will have 0.29.0 out tommorow, so if you have any
more patches, please send them now!

-g.

On 3/6/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> hi list!
>
> * Glue::FileCache improved: escaping bad characters in filenames
> * Glue::FileCache: safe locking of files added, making this caching
> method safe for multiprocess
>
> Bryan: this patch makes test case tc_session.rb runs well!
>
> regards
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar  6 15:51:34 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 6 Mar 2006 21:51:34 +0100
Subject: [Nitro] glue::mailer examples
In-Reply-To: <200603061459.08621.riku.raisanen@walkingwoods.com>
References: <200603061459.08621.riku.raisanen@walkingwoods.com>
Message-ID: <afd80b540603061251v3787221dx9dd1b0b0f637c44f@mail.gmail.com>

I think the blog example uses the Mailer.

Btw, the mailer works more or less like the Rails ActionMailer.

-g.

On 3/6/06, Riku R?is?nen <riku.raisanen at walkingwoods.com> wrote:
> Are there any examples for the usage? and what about sending attachments?
>
> -Riku R?is?nen
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Mon Mar  6 16:04:00 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 13:04:00 -0800
Subject: [Nitro] Patch workflow
In-Reply-To: <EB2333A8-26CC-4D6A-89FB-044372B8C1A3@yoyo.org>
References: <634BA5F0-5A9F-4B18-9F51-A2767D712DBF@yoyo.org>
	<bd93db0d0603052346o5e5ce730pc1b4920814594b6f@mail.gmail.com>
	<EB2333A8-26CC-4D6A-89FB-044372B8C1A3@yoyo.org>
Message-ID: <bd93db0d0603061304l47146f66kdd00fdfa48002670@mail.gmail.com>

On 3/6/06, Aidan Rogers <aidan at yoyo.org> wrote:
> I use OmniGraffle, which is a Mac-only app.  I'm happy to draw as
> many diagrams as necessary to help people understand stuff.  The only
> caveat is someone has to explain it to me first :-)

Well, if you're going to be difficult... ;)



From bryan.a.soto at gmail.com  Mon Mar  6 16:21:34 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 13:21:34 -0800
Subject: [Nitro] nitro rake test
In-Reply-To: <afd80b540603060238j71e1fdd2h63af9f1cd584addd@mail.gmail.com>
References: <afd80b540603030044ue25d9f6icf4245d6095fcbeb@mail.gmail.com>
	<bd93db0d0603042356p34b3d7d2gbadacc1348ce6cae@mail.gmail.com>
	<afd80b540603050039p105db54dvd44d073d3c485d4f@mail.gmail.com>
	<bd93db0d0603051020q17b831e0yfcf7b93f64fd95b5@mail.gmail.com>
	<afd80b540603060238j71e1fdd2h63af9f1cd584addd@mail.gmail.com>
Message-ID: <bd93db0d0603061321s63e67abcu350a437a5601f7d1@mail.gmail.com>

On 3/6/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Rake test still does not work for me :(
> I get the same error (tc_session.rb not found)

I ran into the exact same error while running rcov, but it was with
tc_cookie.rb. I renamed the file to tcx_cookie.rb and renamed it back
to tc_cookie.rb and the problem went away. I don't know exactly what
caused it though.

Weird.



From bryan.a.soto at gmail.com  Mon Mar  6 16:23:14 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 13:23:14 -0800
Subject: [Nitro] [flame] Bad things in Nitro and Og
In-Reply-To: <afd80b540603060256r5d5efcfu7c710fa007b08ad5@mail.gmail.com>
References: <200603031823.26197.zimba.tm@gmail.com>
	<bd93db0d0603031336q39bcccadt8b0c7d3da3d8c5a3@mail.gmail.com>
	<4b6f054f0603031725i423fa811v139b203c748418c8@mail.gmail.com>
	<afd80b540603060256r5d5efcfu7c710fa007b08ad5@mail.gmail.com>
Message-ID: <bd93db0d0603061323j4ad7a7ectded6b5d718798833@mail.gmail.com>

On 3/6/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> I didnt realize that Nitro is slow at startup. Perhaps my apps have
> not many classes or something. Anyway,    the ObjectSpace traversal
> thing is easy to fix without giving up the magic. Stay tunned for an
> update in 0.30.0
>

It can be. I think Rob complained about it, but he has 40+ classes, if
I recall correctly.

> moreover, I will add some more startup time improvements (for example
> transformed scripts caching)
>
> but, lets get 0.29.0 out first ;-)
>

Agreed. :)



From bryan.a.soto at gmail.com  Mon Mar  6 16:34:28 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 13:34:28 -0800
Subject: [Nitro] Oxyliquit tip
In-Reply-To: <op.s5zpd8neg3glhz@jo.local>
References: <afd80b540603060254l6bfb66a5jb45b69d213b7795b@mail.gmail.com>
	<op.s5zpd8neg3glhz@jo.local>
Message-ID: <bd93db0d0603061334w65efe1a2he6a526afc4ca987b@mail.gmail.com>

On 3/6/06, Kashia Buch <kashia at vfemail.net> wrote:
> BTW: Oxylquit had a feed already built in /feed , enjoy :)

Mozilla Thunderbird says it's not a valid rss feed. And looking at the
contents, it seems to be a regular web page... It doesn't contain an
rss version number, for example.

I'll see if I can find the source, since g mentioned it's open source.
Patches welcome, after all :)



From bryan.a.soto at gmail.com  Mon Mar  6 16:38:17 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 13:38:17 -0800
Subject: [Nitro] Tutorial?
In-Reply-To: <425755733c85103f070b55c11db74160@oggu.de>
References: <3340E546-3618-4AED-AB98-5A6CE8B5EF06@yoyo.org>
	<op.s5xuk6swg3glhz@jo.local> <200603061113.52193.manveru@weez.co.jp>
	<bd93db0d0603060014j234995e1mbc31e15d230ab470@mail.gmail.com>
	<425755733c85103f070b55c11db74160@oggu.de>
Message-ID: <bd93db0d0603061338w2594f926kcd2538f1ea198556@mail.gmail.com>

On 3/6/06, Fabian Buch <fabian at oggu.de> wrote:
> The reason we didn't advertise it yet is that we didn't think it's
> feature-complete yet and that it still has annoying bugs. One of them
> being in Nitro (which was the reason for one of the tickets:
> http://devlab.oree.ch/trac/nitrohq/ticket/9), workaround is applied to
> the Nitro on Manveru's server though, so tags like "web2.0" work.

Did you make the change to cgi.rb file? I didn't want to make what
might be a big change this late as I thought George was releasing
today, 3/6/2006. But if you guys are already testing, it might be
easier to get in...

bryan



From bryan.a.soto at gmail.com  Mon Mar  6 16:48:35 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 13:48:35 -0800
Subject: [Nitro] PATCH: sparck-style.css Fwd:  Spark pre text
Message-ID: <bd93db0d0603061348t571c78c7u2eaa3a8389ad302@mail.gmail.com>

Much nicer. :)

Patch attached credited to Timh.

---------- Forwarded message ----------
From: TimH <timh at dirtymonday.net>
Date: Mar 6, 2006 8:31 AM
Subject: [Nitro] Spark pre text
To: nitro-general at rubyforge.org


For the next version of Spark, can we fix the pre-formatted text
coloring by replacing this line in public/style.css

        pre {background: #424242; padding: 10px; overflow: auto;}

with something like this?

        pre {background: #424242; color: #FFF; padding: 10px; overflow: auto;}

--TimH
_______________________________________________
Nitro-general mailing list
Nitro-general at rubyforge.org
http://rubyforge.org/mailman/listinfo/nitro-general
-------------- next part --------------
A non-text attachment was scrubbed...
Name: spark-style.css.zip
Type: application/zip
Size: 10945 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060306/140feda6/attachment.zip 

From m.fellinger at gmail.com  Mon Mar  6 16:55:50 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Tue, 7 Mar 2006 06:55:50 +0900
Subject: [Nitro] Admin improvements
In-Reply-To: <200603061558.10912.zimba.tm@gmail.com>
References: <200603061558.10912.zimba.tm@gmail.com>
Message-ID: <9c00d3e00603061355i6630d8e6l2d19ff4fe66e287f@mail.gmail.com>

Well, we could store the Settings in /conf/admin.yaml or something,later reading it with Og and modifying the variables on the fly whilesaving them back to yaml.Now, my next question is... is there such a thing as a YAML-adapter?guess not, in this case we just have to do it manually, wich shouldn'tbe too hard.Anyway, would like to make these changes, also adding/deleting Options(if it was a standard-option it will just be set back to standard, alslong as you don't purge it) on the fly would be nice... kind of what idid with the Localization
anyway, will give this one a stab today, tho i think a web-interfaceto rubygems sounds really tempting, i'm sure i'd mess too much aroundwith it :)
The online-editing-interface would be the cream on my cup ofstrawberries - please make it happen ;) (yes i will help as much as ican)
manveru
On 3/6/06, zimba.tm <zimba.tm at gmail.com> wrote:> Hi list,>> the next thing I would to work on, is to improve the admin. Before I go on,> here are the ideas I would like to implement :>> = Make settings changeable and storable>>  * Use the caching system or Og to load/save settings>  * Make the settings modifiable in the web interface>  * Group the settings in different pages, according to their namespace to> avoid a huge page>> = Use rubygems for components>> I don't know if it will be straignforward.>>  * Use a nitro-specific settings for rubygems (store location, download> location, ...)>  * Make rubygems (un)installable/uploadable from the web interface>  * Add the ability to instanciate components on specific web paths>  * Upgrade Og so that it supports table installing/upgrading on the fly> (without relaunching nitro)>> = Standalone nitro>> Make a nitro executable that will launch nitro with the admin and custom> options>> = Editable>>  * Make sources and templates editable from the web interface>> --> Cheers,>  zimba.tm>> weblog : http://zimba.oree.ch> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From m.fellinger at gmail.com  Mon Mar  6 16:56:50 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Tue, 7 Mar 2006 06:56:50 +0900
Subject: [Nitro] Spark pre text
In-Reply-To: <20060306083113.7449e959.timh@dirtymonday.net>
References: <20060306083113.7449e959.timh@dirtymonday.net>
Message-ID: <9c00d3e00603061356w192976efy3c1eda18dd37c9fb@mail.gmail.com>

just submit a patch please, i did so 3 times and nothing changed :\
On 3/7/06, TimH <timh at dirtymonday.net> wrote:> For the next version of Spark, can we fix the pre-formatted text coloring by replacing this line in public/style.css>>        pre {background: #424242; padding: 10px; overflow: auto;}>> with something like this?>>        pre {background: #424242; color: #FFF; padding: 10px; overflow: auto;}>> --TimH> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From aidan at yoyo.org  Mon Mar  6 17:12:59 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Tue, 7 Mar 2006 09:12:59 +1100
Subject: [Nitro] Nitro documentation
In-Reply-To: <1141642776.717749105.19198.sendItem@bloglines.com>
References: <1141642776.717749105.19198.sendItem@bloglines.com>
Message-ID: <3F4B828D-B43E-4941-A8E7-CC8B351C3966@yoyo.org>

Derek,

There isn't complete and comprehensive documentation in existence  
yet.  Try http://www.infurious.com/downloads/ for an Og tutorial.   
http://www.nitrohq.com and http://trac.nitrohq.com also have some  
documentation on them.

Aidan

On 06/03/2006, at 9:59 PM, derekmahar.26026218 at bloglines.com wrote:

> Where can I find complete and comprehensive Nitro and Ogg  
> documentation?
> I've read an article about Ogg that really piqued my interest, but  
> I don't
> know  where to find more information.
>
> Derek
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>

---
http://www.infurious.com




From aidan at yoyo.org  Mon Mar  6 17:48:29 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Tue, 7 Mar 2006 09:48:29 +1100
Subject: [Nitro] Test driven development for Nitro
Message-ID: <90917A5D-D25F-48A1-8C5F-45AE58CC3F5D@yoyo.org>

Hi all,

I sat down last night to write my first Nitro Element.  I soon  
discovered that there was no simple way to write this using a Test  
Driven Development (TDD) approach.  Looking at some of the Element  
implementations in the various examples/spark/flare, it seems that at  
its simplest an Element is an object which emits HTML.

Any objections to me putting together whatever is necessary to make  
it easy to do TDD/BDD (Behaviour Driven Development) support for  
Nitro?  I've created a ticket in Trac.  http://devlab.oree.ch/trac/ 
nitrohq/ticket/11

Aidan
---
http://www.infurious.com




From bryan.a.soto at gmail.com  Mon Mar  6 18:32:26 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 15:32:26 -0800
Subject: [Nitro] PATCH: testfix-tc_controller
Message-ID: <bd93db0d0603061532k56af9111wcb0869c9f6e18d26@mail.gmail.com>

* testfix-tc_controller
The new template_root stack breaks old code by not honoring the
Controller.template_root method as before. tc_controller.rb failed
because of this. Patch modifies template_for_action method in
compiler.rb:58 to also use the template_root method if defined.

If the change was intentional, I'd recommend this be noted as a
feature that is deprecated and removing it at after notice is given.
0.30.0 perhaps.

-----

This was difficult to track down until I started using Trac and it's
changeset/revision log abilities. Consider me converted. :)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: testfix-tc_controller.zip
Type: application/zip
Size: 11064 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060306/f5091366/attachment.zip 

From kashia at vfemail.net  Mon Mar  6 18:33:37 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Tue, 07 Mar 2006 00:33:37 +0100
Subject: [Nitro] Oxyliquit tip
In-Reply-To: <bd93db0d0603061334w65efe1a2he6a526afc4ca987b@mail.gmail.com>
References: <afd80b540603060254l6bfb66a5jb45b69d213b7795b@mail.gmail.com>
	<op.s5zpd8neg3glhz@jo.local>
	<bd93db0d0603061334w65efe1a2he6a526afc4ca987b@mail.gmail.com>
Message-ID: <op.s50mebp8g3glhz@jo.local>

Yeh,

> On 3/6/06, Kashia Buch <kashia at vfemail.net> wrote:
>> BTW: Oxylquit had a feed already built in /feed , enjoy :)
> I'll see if I can find the source, since g mentioned it's open source.
> Patches welcome, after all :)

Already patched that, but couldn't upload it due to Michael going home :P
Since he moved he doesn't have internet at home anymore, poor guy ^^;

I don't know yet if it's open source... I don't have any ideas about
it and I don't know anything about copyright and licensing.

For now:

What you can do:
  * look at the code
  * work on the code
  * send me patches
  * use small parts of the code in your own project
  * use ideas of it in your own projects
  * if you redistribute it somehow, tell them where you got it from

But please don't:
  * redistibute the work as your own
  * claim it is yours
  * use whole parts in your own projects
  * make a new "Nitro FAQ" with it
  * use anything for commercial projects


Basically everything I'd expect from nice people anyway.
But, over all:  ask me when you want something, I'm very friendly ^_^

'nuff said about unpleasent thingies

I wanted to, somehow introduce the Nitro FAQ to the list somehow, but
since I didn't have any idea how to do that, and Michael just spoiled
it, it doesn't matter now :D

There are already some usefully questions and answers in it now, have
fun using it.

One word about those rating numbers at the end of each answer to your
questions: 0 is unrated, 4 and 5 declare the Question as answered and
will be shown in the "closed" list. If you still feel that there is a
lack of information, you can still answer. The user interface and the
registering/viewing/editing might still be very unstable, please post
anything you get (errors, weird behaviour) right to me, I'll see that
it gets fixed as soon as I can.

Oh, and about editing:

Use textile (I don't like it, but people seem to use it all the time)
and for code use:

<code syntax="ruby">
class Foor
   attr :bar
end
</code>

It is a pleasure to code in Nitro/Og, thank you, I thank everyone who
contibutes to this wonderful project and see the "Oxyliquit" as a way
to contribute to my favourite Web-Framework.

I sure like to talk, don't I ;D

I'll leave it at that, everyone, use the faq, and enjoy it :)

Kash

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From dcnstrct at hushmail.com  Mon Mar  6 18:56:53 2006
From: dcnstrct at hushmail.com (dcnstrct at hushmail.com)
Date: Mon,  6 Mar 2006 18:56:53 -0500
Subject: [Nitro] Og newbie has a question and a few observations.
Message-ID: <20060306235657.F1EE333C23@mailserver5.hushmail.com>

yo,

Is there a way to get a collection of every Og "enchanted" class ?

thanks,

--dcnstrct
dcnstrct at hushmail.com 

p.s.

possible bugs:  Og's sqlite store appears to have a broken 
find_one() method, and Og's kirbybase store has a working 
find_one() method yet find( :limit => 1 ) appears to be broken for 
kirbybase =^o  






Concerned about your privacy? Instantly send FREE secure email, no account required
http://www.hushmail.com/send?l=480

Get the best prices on SSL certificates from Hushmail
https://www.hushssl.com?l=485



From bryan.a.soto at gmail.com  Mon Mar  6 19:11:12 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 16:11:12 -0800
Subject: [Nitro] PATCH: testfix-tc_inheritance
Message-ID: <bd93db0d0603061611g3095a92co94b391c90483018d@mail.gmail.com>

* testfix-tc_inheritance
Repairs implementation of tc_inheritance test. Class.descendents is a
Facet method that is implemented by iterating through ObjectSpace. It
won't always return these descendents in the same order.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: testfix-tc_inheritance.zip
Type: application/zip
Size: 10982 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060306/86b509c7/attachment.zip 

From bryan.a.soto at gmail.com  Mon Mar  6 19:23:06 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 16:23:06 -0800
Subject: [Nitro] Spark pre text
In-Reply-To: <9c00d3e00603061356w192976efy3c1eda18dd37c9fb@mail.gmail.com>
References: <20060306083113.7449e959.timh@dirtymonday.net>
	<9c00d3e00603061356w192976efy3c1eda18dd37c9fb@mail.gmail.com>
Message-ID: <bd93db0d0603061623g702431eei5ef2b210abd58778@mail.gmail.com>

Already submitted one and credited it to TimH. :)

On 3/6/06, Michael Fellinger <m.fellinger at gmail.com> wrote:
> just submit a patch please, i did so 3 times and nothing changed :\
> On 3/7/06, TimH <timh at dirtymonday.net> wrote:> For the next version of Spark, can we fix the pre-formatted text coloring by replacing this line in public/style.css>>        pre {background: #424242; padding: 10px; overflow: auto;}>> with something like this?>>        pre {background: #424242; color: #FFF; padding: 10px; overflow: auto;}>> --TimH> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From bryan.a.soto at gmail.com  Mon Mar  6 19:32:14 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 16:32:14 -0800
Subject: [Nitro] Og newbie has a question and a few observations.
In-Reply-To: <20060306235657.F1EE333C23@mailserver5.hushmail.com>
References: <20060306235657.F1EE333C23@mailserver5.hushmail.com>
Message-ID: <bd93db0d0603061632s42f9937s4b2d6a482029cb8e@mail.gmail.com>

On 3/6/06, dcnstrct at hushmail.com <dcnstrct at hushmail.com> wrote:
> yo,
>
> Is there a way to get a collection of every Og "enchanted" class ?

Do you mean of all the classes currently managed by Og?

In current 0.28.0,
managed_classes = Og::Manager.managers.collect {|m| m.managed_classes}.flatten

In pending 0.29.0 (tomorrow unless delayed),
managed_classes = Og::Manager.managed_classes

> possible bugs:  Og's sqlite store appears to have a broken
> find_one() method, and Og's kirbybase store has a working
> find_one() method yet find( :limit => 1 ) appears to be broken for
> kirbybase =^o

You don't happen to have a small script or testcase to duplicate, do
you? I'd like to check these against the current repo.

Thanks,

Bryan



From dcnstrct at hushmail.com  Mon Mar  6 22:31:05 2006
From: dcnstrct at hushmail.com (dcnstrct at hushmail.com)
Date: Mon,  6 Mar 2006 22:31:05 -0500
Subject: [Nitro] Og newbie has a question and a few observations.
Message-ID: <20060307033111.64CAD33C23@mailserver5.hushmail.com>

Bryan, 

I looked harder at the possible bug and it turns out to be 
a non-bug. 

To sum up what happened:  

FooTable#find( :limit => 1 ) returns an array. 

I incorrectly assumed that FooTable#find_one() would also 
return an array, but it does not.  It now seems obvious to
me that this was the intended behavior.  

thanks very much for your help with my earlier question. 

--dcnstrct 


On Mon, 06 Mar 2006 19:32:14 -0500 Bryan Soto 
<bryan.a.soto at gmail.com> wrote:
>On 3/6/06, dcnstrct at hushmail.com <dcnstrct at hushmail.com> wrote:
>> yo,
>>
>> Is there a way to get a collection of every Og "enchanted" class 
>?
>
>Do you mean of all the classes currently managed by Og?
>
>In current 0.28.0,
>managed_classes = Og::Manager.managers.collect {|m| 
>m.managed_classes}.flatten
>
>In pending 0.29.0 (tomorrow unless delayed),
>managed_classes = Og::Manager.managed_classes
>
>> possible bugs:  Og's sqlite store appears to have a broken
>> find_one() method, and Og's kirbybase store has a working
>> find_one() method yet find( :limit => 1 ) appears to be broken 
>for
>> kirbybase =^o
>
>You don't happen to have a small script or testcase to duplicate, 
>do
>you? I'd like to check these against the current repo.
>
>Thanks,
>
>Bryan
>
>_______________________________________________
>Nitro-general mailing list
>Nitro-general at rubyforge.org
>http://rubyforge.org/mailman/listinfo/nitro-general



Concerned about your privacy? Instantly send FREE secure email, no account required
http://www.hushmail.com/send?l=480

Get the best prices on SSL certificates from Hushmail
https://www.hushssl.com?l=485



From bryan.a.soto at gmail.com  Tue Mar  7 01:37:32 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 22:37:32 -0800
Subject: [Nitro] Controller resolution order
In-Reply-To: <440C65DC.7090200@cataclysm-software.net>
References: <440C65DC.7090200@cataclysm-software.net>
Message-ID: <bd93db0d0603062237o1bc293dbw51a03e6af69492c6@mail.gmail.com>

On 3/6/06, Andrew Thompson <vagabond at cataclysm-software.net> wrote:
> Nitro::Server.map= {'/'=> BaseController,
> '/admin/users'=>UserController, '/admin'=>AdminController}
>
> now /admin points to the admin controller, fine. However /admin/users
> ALSO points to the admin controller and looks for a 'users' method on
> it. This seems a bit counter-intuitive, why doesn't Nitro look for a
> controller to match before looking for a method in a controller?
>
> It also does a similar thing for methods with a double underscore (I
> think foo__bar() maps to foo/bar), if I define foo(), and go to foo/bar,
> instead of finding foo__bar(), it passes 'bar' to foo(). This works fine
> if foo() is not defined in the controller however.
>

Hmm... you make a good point. I suppose George would probably be the
person to ask as to why current behaviour is what it is. Any opinions
on whether this is a bug or a feature? :)

Bryan



From bryan.a.soto at gmail.com  Tue Mar  7 01:45:31 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 22:45:31 -0800
Subject: [Nitro] [Bug] Memory based session cache does not release memory
In-Reply-To: <716700c90603050552u673ab3abqa61f7e6da537627f@mail.gmail.com>
References: <716700c90603050552u673ab3abqa61f7e6da537627f@mail.gmail.com>
Message-ID: <bd93db0d0603062245w51f55061j8747c1012ca23b85@mail.gmail.com>

On 3/5/06, Fang Sun <nusgnaf at gmail.com> wrote:
> Hi, devs on the list. I develop using scgi+lighttpd configuration,
> with memory sessions. I use ab from apache package to benchmark. find
> the scgi session continuously eat up memory. On a P4 630 machine with
> 1gb memory, a 20,0000 run consumes almost 200Mb memory. I setup in
> conf/live.rb the following parameter:
>
> I can see the hash size of  @hash variable of Nitro::Session.cache
> shrink from several thounds of to 0 as time pass the keepalive time,
> but the ruby consumed memory doesn't shrink. Any better way to debug
> the session memory leak?

Have you confirmed that the leak goes away if you change the session
type? Say og, file, drb or the new memcached that will be available in
the upcoming release? I'm just wondering since you say the cache size
shrinks to zero. Perhaps you have something holding onto references
somewhere else?

Bryan



From dylanb at digitalvalence.com  Tue Mar  7 01:45:51 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Tue, 07 Mar 2006 00:45:51 -0600
Subject: [Nitro]  Kirbybase/inheritance/schema evolution
In-Reply-To: <afd80b540603060243s6ca4dbfek33858e2ca9da0e5a@mail.gmail.com>
References: <bd93db0d0603052339t25b8a374u7fb3b7348b70d132@mail.gmail.com>
	<afd80b540603060243s6ca4dbfek33858e2ca9da0e5a@mail.gmail.com>
Message-ID: <440D2C1F.3080402@digitalvalence.com>

I recently started implementing a desktop application using OG to 
connect to KirbyBase as the back end.  While implementing my initial 
model tests I came across a couple of problems with the schema 
inheritance and schema evolution features.

1.  Schema inheritance creates extra columns for each subclass 
property.  Say we have a super class, Person, which has two children, 
Manager and Programmer.  If Manager has an employees property and 
Programmer has a cupsOfCoffee property schema evolution will drop the 
column for whichever property came first.  This may be a general 
inheritance and evolution problem, not just specific to Kirbybase.  I 
haven't checked the other stores for more advanced methods of handling 
this, such as using descendents.
2.  The Kirby store currently works with the Root class for the table 
only, and so if I create a Manager and an Employee and save them both 
referencing Person[1] will bring back the class as an Employee.  What is 
worse, the extra table column will cause an error to be thrown in the 
kb_create method as there are more columns than expected.
3. The evolution process attempts to drop the required recno column.
4. The ogtype column was not being populated.  It would seem that adding 
the ogtype column to the module SchemaInheritanceBase fixes this.  It  
seems like the system has extra code to handle this property on each 
database access that would use it.  Wouldn't it be better to add the 
property using a little bit of magic ?


So, I've fixed these problems.  Took a little doing, but I revamped the 
kirby store to use the Struct mechanism and automatically create the 
return type based on either the class or the ogtype column. 
set_schema_inheritance in entity has been altered, the class now over 
writes the inherited function, to allow for subclass tracking.  Every 
time schema evolution occurs I get the full list of properties for the 
root and all of its subclasses (this isn't yet recursive, I will fix 
that next, maybe use descendents (how fast is this ?)) and use that to 
determine the necessary table columns.   I'm in the middle of patching 
this so that it takes the field_type from the target subclass, rather 
than searching the superclass and retrieving nil.

Should I submit this when I have these few final fixes in ?  How would I 
go about that ? I have darcs up and running, but the changes are pretty 
drastic and a whole bunch of non-changes are being reported as changed 
because the line numbers are off.  Might be an Eclipse issue, but it 
shows up in the command line tool now as well.  What tool should I use 
to reapply these changes in windows without screwing up the spacing ?  I 
don't use VI.  The kirby.rb file is different enough that I don't think 
I could replicate the changes by hand without missing something.  Should 
I just send these along ?

I also found other issues:

1. If an object is added to a has_many the persistence seems to be 
automatic, not requiring the primary object to be saved.  Since 
persistence relies on id the object will be saved with no id in the 
foreign key column, and hence it will not be pulled back with the object 
that you attached it to, even if you the parent object.
2. There seems to be no delete cascading; I'm guessing either I just 
don't know how to specify this, or it isn't present at all.
3.  I ran into issues with specifying the model classes and then 
configuring the database(with Mysql as well), rather than configuring 
the db, creating the model classes, and then calling manage_classes 
explicitly.  A little unpredictable.  I'm going to take this one off the 
table until I can come up with a better description than 'issues' though.

I haven't had time to look at these in any real depth yet.  Maybe 
someone else has ?

Other than these things I'm having a blast poking about with the 
framework.  I really like the feature set; now if only it were a little 
smoother traveling off the beaten path.  Certainly beats the hell out of 
Hibernate.  I've been thinking of implementing one class per table 
inheritance, but I'll leave that for another day since I am losing 
coherency at this time of night.

thanks for the hard work,
Dylan


From bryan.a.soto at gmail.com  Tue Mar  7 02:06:28 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 6 Mar 2006 23:06:28 -0800
Subject: [Nitro] Best way to pass parameter to Nitro::Server?
In-Reply-To: <716700c90603020045u71bd4519v1ce0fb1883ee17c4@mail.gmail.com>
References: <716700c90603020045u71bd4519v1ce0fb1883ee17c4@mail.gmail.com>
Message-ID: <bd93db0d0603062306h4527ef1dh9e1d01333384949c@mail.gmail.com>

On 3/2/06, Fang Sun <nusgnaf at gmail.com> wrote:
> Hi, I am working on imporve SCGI adapters, and am planning to port
> goodies like DRb control and scgi cluster from Zed's scgi-rails
> package.
>
> Right now we use Nitro::Runner to fireup the scgi server. and hard
> code many default setting in SCGI.start method. What's the prefered to
> way to pass parameters to adapters in Nitro?

Hi,

Please forgive the late response. Perhaps using a setting, as in
Nitro::Server? Or maybe command line arguments as in Nitro::Runner?

I'm not familiar with scgi-rails. Maybe if you could describe some of
the parameters you need to pass, we could figure something out?



From dylanb at digitalvalence.com  Tue Mar  7 02:14:14 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Tue, 07 Mar 2006 01:14:14 -0600
Subject: [Nitro] Kirbybase/inheritance/schema evolution
In-Reply-To: <440D2C1F.3080402@digitalvalence.com>
References: <bd93db0d0603052339t25b8a374u7fb3b7348b70d132@mail.gmail.com>	<afd80b540603060243s6ca4dbfek33858e2ca9da0e5a@mail.gmail.com>
	<440D2C1F.3080402@digitalvalence.com>
Message-ID: <440D32C6.20900@digitalvalence.com>

Changing to descendents(sp?) fixes all the obvious problems with the 
inheritance stuff.  Most of the kirbybase issues seem to stem from 
having to override core sql.rb methods like fields_for_class. 

Now I really do have to sleep.  After this last little change...

cheers,

Dylan


From george.moschovitis at gmail.com  Tue Mar  7 03:07:21 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 10:07:21 +0200
Subject: [Nitro] [PATCH] improvements on Glue::FileCache
In-Reply-To: <afd80b540603061249r177c80f5t51d12b91cf356222@mail.gmail.com>
References: <6a7d49ca0603060413v3a5aeb69g@mail.gmail.com>
	<afd80b540603061249r177c80f5t51d12b91cf356222@mail.gmail.com>
Message-ID: <afd80b540603070007o6d6a2e88s75cf1ef4ab6cf2bd@mail.gmail.com>

there where some confilicts in the first patch, can you resend against
the latest repo please?

-g.

On 3/6/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> thanks, will try this first thing in the morning ;-)
> most probably we will have 0.29.0 out tommorow, so if you have any
> more patches, please send them now!
>
> -g.
>
> On 3/6/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > hi list!
> >
> > * Glue::FileCache improved: escaping bad characters in filenames
> > * Glue::FileCache: safe locking of files added, making this caching
> > method safe for multiprocess
> >
> > Bryan: this patch makes test case tc_session.rb runs well!
> >
> > regards
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
> >
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar  7 03:12:23 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 10:12:23 +0200
Subject: [Nitro] Test driven development for Nitro
In-Reply-To: <90917A5D-D25F-48A1-8C5F-45AE58CC3F5D@yoyo.org>
References: <90917A5D-D25F-48A1-8C5F-45AE58CC3F5D@yoyo.org>
Message-ID: <afd80b540603070012q31351d1et8f41197d2c8f2da2@mail.gmail.com>

Have you seen the code in

lib/nitro/test
lib/og/test

perhaps you can improve there?

In any case, I have some nice ideas how to improve the support for
Testing in Nitro, but I am running out of time now and cannot write a
proper explanation. So, perhaps you could give it a try, and send your
improvements and see what comes up. Perhaps you have some better ideas
than I have ;-)


best regards,
George.



On 3/7/06, Aidan Rogers <aidan at yoyo.org> wrote:
> Hi all,
>
> I sat down last night to write my first Nitro Element.  I soon
> discovered that there was no simple way to write this using a Test
> Driven Development (TDD) approach.  Looking at some of the Element
> implementations in the various examples/spark/flare, it seems that at
> its simplest an Element is an object which emits HTML.
>
> Any objections to me putting together whatever is necessary to make
> it easy to do TDD/BDD (Behaviour Driven Development) support for
> Nitro?  I've created a ticket in Trac.  http://devlab.oree.ch/trac/
> nitrohq/ticket/11
>
> Aidan
> ---
> http://www.infurious.com
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar  7 03:16:26 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 10:16:26 +0200
Subject: [Nitro] PATCH: testfix-tc_controller
In-Reply-To: <bd93db0d0603061532k56af9111wcb0869c9f6e18d26@mail.gmail.com>
References: <bd93db0d0603061532k56af9111wcb0869c9f6e18d26@mail.gmail.com>
Message-ID: <afd80b540603070016v7175c894t3e67faaa7748a73a@mail.gmail.com>

Hm, I prefer to *not* include this patch. The change was intended, I
think the new template_root stack system works much better now. mixing
the old/new aproach like you do in your patch is not good I think.

have a look at the admin part controller to see how to convert the old
method to the new

def self.setup_template_root(path)
end

method.

A bit later I will post s message describing the rationale behind this change.


regards,
George.



On 3/7/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> * testfix-tc_controller
> The new template_root stack breaks old code by not honoring the
> Controller.template_root method as before. tc_controller.rb failed
> because of this. Patch modifies template_for_action method in
> compiler.rb:58 to also use the template_root method if defined.
>
> If the change was intentional, I'd recommend this be noted as a
> feature that is deprecated and removing it at after notice is given.
> 0.30.0 perhaps.
>
> -----
>
> This was difficult to track down until I started using Trac and it's
> changeset/revision log abilities. Consider me converted. :)
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar  7 03:19:18 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 10:19:18 +0200
Subject: [Nitro] Spark pre text
In-Reply-To: <bd93db0d0603061623g702431eei5ef2b210abd58778@mail.gmail.com>
References: <20060306083113.7449e959.timh@dirtymonday.net>
	<9c00d3e00603061356w192976efy3c1eda18dd37c9fb@mail.gmail.com>
	<bd93db0d0603061623g702431eei5ef2b210abd58778@mail.gmail.com>
Message-ID: <afd80b540603070019r741a0e51pba3948b526246e72@mail.gmail.com>

Added the patch in the repo, thanks!

-g.

On 3/7/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Already submitted one and credited it to TimH. :)
>
> On 3/6/06, Michael Fellinger <m.fellinger at gmail.com> wrote:
> > just submit a patch please, i did so 3 times and nothing changed :\
> > On 3/7/06, TimH <timh at dirtymonday.net> wrote:> For the next version of Spark, can we fix the pre-formatted text coloring by replacing this line in public/style.css>>        pre {background: #424242; padding: 10px; overflow: auto;}>> with something like this?>>        pre {background: #424242; color: #FFF; padding: 10px; overflow: auto;}>> --TimH> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From aidan at yoyo.org  Tue Mar  7 03:52:50 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Tue, 7 Mar 2006 19:52:50 +1100
Subject: [Nitro] Test driven development for Nitro
In-Reply-To: <afd80b540603070012q31351d1et8f41197d2c8f2da2@mail.gmail.com>
References: <90917A5D-D25F-48A1-8C5F-45AE58CC3F5D@yoyo.org>
	<afd80b540603070012q31351d1et8f41197d2c8f2da2@mail.gmail.com>
Message-ID: <06DE7658-9E7E-4D12-B878-513A7915DFF4@yoyo.org>

On 07/03/2006, at 7:12 PM, George Moschovitis wrote:

> Have you seen the code in
>
> lib/nitro/test
> lib/og/test
>
> perhaps you can improve there?

Nope, not seen that.  I'll take a look at it now.

Aidan
---
http://www.infurious.com




From george.moschovitis at gmail.com  Tue Mar  7 04:10:46 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 11:10:46 +0200
Subject: [Nitro] Test driven development for Nitro
In-Reply-To: <06DE7658-9E7E-4D12-B878-513A7915DFF4@yoyo.org>
References: <90917A5D-D25F-48A1-8C5F-45AE58CC3F5D@yoyo.org>
	<afd80b540603070012q31351d1et8f41197d2c8f2da2@mail.gmail.com>
	<06DE7658-9E7E-4D12-B878-513A7915DFF4@yoyo.org>
Message-ID: <afd80b540603070110u79157c8bx28c62dddb62dc40e@mail.gmail.com>

Also have a look at:

spark/test/*

for example usage...

-g.


On 3/7/06, Aidan Rogers <aidan at yoyo.org> wrote:
> On 07/03/2006, at 7:12 PM, George Moschovitis wrote:
>
> > Have you seen the code in
> >
> > lib/nitro/test
> > lib/og/test
> >
> > perhaps you can improve there?
>
> Nope, not seen that.  I'll take a look at it now.
>
> Aidan
> ---
> http://www.infurious.com
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar  7 04:12:14 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 11:12:14 +0200
Subject: [Nitro] Og tests...
Message-ID: <afd80b540603070112r372d1d20p7dab942a5ea72ea7@mail.gmail.com>

Bryan
I am sorry to dissapoint you but rake test does not run all tests (it
doesnt run tests in test/mixin for example). Saddly test like
tc_taggable do not pass (and tags seem a bit f*cked up). Anyway, I am
working on this...

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From guillaume.pierronnet at gmail.com  Tue Mar  7 05:27:33 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Tue, 7 Mar 2006 11:27:33 +0100
Subject: [Nitro] [PATCH] improvements on Glue::FileCache
In-Reply-To: <afd80b540603070007o6d6a2e88s75cf1ef4ab6cf2bd@mail.gmail.com>
References: <6a7d49ca0603060413v3a5aeb69g@mail.gmail.com>
	<afd80b540603061249r177c80f5t51d12b91cf356222@mail.gmail.com>
	<afd80b540603070007o6d6a2e88s75cf1ef4ab6cf2bd@mail.gmail.com>
Message-ID: <6a7d49ca0603070227h659c408bx@mail.gmail.com>

fixed bundle

2006/3/7, George Moschovitis <george.moschovitis at gmail.com>:
> there where some confilicts in the first patch, can you resend against
> the latest repo please?
>
> -g.
>
> On 3/6/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > thanks, will try this first thing in the morning ;-)
> > most probably we will have 0.29.0 out tommorow, so if you have any
> > more patches, please send them now!
> >
> > -g.
> >
> > On 3/6/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > > hi list!
> > >
> > > * Glue::FileCache improved: escaping bad characters in filenames
> > > * Glue::FileCache: safe locking of files added, making this caching
> > > method safe for multiprocess
> > >
> > > Bryan: this patch makes test case tc_session.rb runs well!
> > >
> > > regards
> > >
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> > >
> > >
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.gz
Type: application/x-gzip
Size: 11108 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060307/8551fe58/attachment.gz 

From george.moschovitis at gmail.com  Tue Mar  7 05:45:59 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 12:45:59 +0200
Subject: [Nitro] Rob's testing improvements.
Message-ID: <afd80b540603070245p3f138c6fj3df68b245443be94@mail.gmail.com>

Can someone please explain the changes Rob made to the testing framework?

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From nusgnaf at gmail.com  Tue Mar  7 07:28:36 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Tue, 7 Mar 2006 12:28:36 +0000
Subject: [Nitro] Best way to pass parameter to Nitro::Server?
In-Reply-To: <bd93db0d0603062306h4527ef1dh9e1d01333384949c@mail.gmail.com>
References: <716700c90603020045u71bd4519v1ce0fb1883ee17c4@mail.gmail.com>
	<bd93db0d0603062306h4527ef1dh9e1d01333384949c@mail.gmail.com>
Message-ID: <716700c90603070428t319bd53r72e3205d764955a4@mail.gmail.com>

Hi, Bryan.
My plan is to create small CLI app to fire up applications. Right now
the shipped scgi_service, scgi_ctl in Nitro use a direct require 'run'
to fire up the proto application, as in the run.rb, we have the
following code:
Nitro.run(Proto)
it will use a runner to setup many parameter under the scene, losing
the parameter from command line.

The attached is a direct port of scgi_ctrl, scgi_service, scgi_cluster
in scgi-rails to nitro. They are CLI tools. I decide to reuse Zed's
work, so it dependes on scgi-rails. The
NitroProcessor#process_request method in nitro_scgi_ctrl was shameless
stolen from nitro/adapter/scgi.rb. I decide not to complicate
server/runner.rb, instead I break up Nitro.run in run.rb to

Nitro::Server.map ={
   blabla
}

leaving the nitro_scgi_service to explicitely pass parameters.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nitro_scgi_cluster
Type: application/octet-stream
Size: 6999 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060307/b7f04d7f/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nitro_scgi_service
Type: application/octet-stream
Size: 1862 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060307/b7f04d7f/attachment-0001.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nitro_scgi_ctrl
Type: application/octet-stream
Size: 4462 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060307/b7f04d7f/attachment-0002.obj 

From nusgnaf at gmail.com  Tue Mar  7 07:34:16 2006
From: nusgnaf at gmail.com (Fang Sun)
Date: Tue, 7 Mar 2006 12:34:16 +0000
Subject: [Nitro] [Bug] Memory based session cache does not release memory
In-Reply-To: <bd93db0d0603062245w51f55061j8747c1012ca23b85@mail.gmail.com>
References: <716700c90603050552u673ab3abqa61f7e6da537627f@mail.gmail.com>
	<bd93db0d0603062245w51f55061j8747c1012ca23b85@mail.gmail.com>
Message-ID: <716700c90603070434me5306f9ud807d4d4dc82c4ff@mail.gmail.com>

Hi, Bryan.
I have tested og, file and memory backed session using proto app. The
leak exists only with memory backend. In fact with og or file session
backed, when I run Nitro::Session.cache.gc!, the session expired is
deleted from database or filesystem, and the nitro program consumed
memory stayed around 13+ MB.



From george.moschovitis at gmail.com  Tue Mar  7 07:37:44 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 14:37:44 +0200
Subject: [Nitro] please test...
Message-ID: <afd80b540603070437x679d88d9ufa1699b869ad833a@mail.gmail.com>

Dear devs,

I am about to release 0.29.0. If you have some time please test the
code in the repo.
In any case this will be a 'snapshot' release. There is so much work
in progress, but this release fixes many important bugs (and adds some
useful features) so I guess it is best to release this 'snapshot'.

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From guillaume.pierronnet at gmail.com  Tue Mar  7 10:02:19 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Tue, 7 Mar 2006 16:02:19 +0100
Subject: [Nitro] Best way to pass parameter to Nitro::Server?
In-Reply-To: <716700c90603070428t319bd53r72e3205d764955a4@mail.gmail.com>
References: <716700c90603020045u71bd4519v1ce0fb1883ee17c4@mail.gmail.com>
	<bd93db0d0603062306h4527ef1dh9e1d01333384949c@mail.gmail.com>
	<716700c90603070428t319bd53r72e3205d764955a4@mail.gmail.com>
Message-ID: <6a7d49ca0603070702j2a002651g@mail.gmail.com>

scgi is launched by specifying --scgi on the command line, no need of
scgi_* scripts.

you can specify --port and --address switches too.

2006/3/7, Fang Sun <nusgnaf at gmail.com>:
> Hi, Bryan.
> My plan is to create small CLI app to fire up applications. Right now
> the shipped scgi_service, scgi_ctl in Nitro use a direct require 'run'
> to fire up the proto application, as in the run.rb, we have the
> following code:
> Nitro.run(Proto)
> it will use a runner to setup many parameter under the scene, losing
> the parameter from command line.
>
> The attached is a direct port of scgi_ctrl, scgi_service, scgi_cluster
> in scgi-rails to nitro. They are CLI tools. I decide to reuse Zed's
> work, so it dependes on scgi-rails. The
> NitroProcessor#process_request method in nitro_scgi_ctrl was shameless
> stolen from nitro/adapter/scgi.rb. I decide not to complicate
> server/runner.rb, instead I break up Nitro.run in run.rb to
>
> Nitro::Server.map ={
>    blabla
> }
>
> leaving the nitro_scgi_service to explicitely pass parameters.
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>



From guillaume.pierronnet at gmail.com  Tue Mar  7 10:33:47 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Tue, 7 Mar 2006 16:33:47 +0100
Subject: [Nitro] please test...
In-Reply-To: <afd80b540603070437x679d88d9ufa1699b869ad833a@mail.gmail.com>
References: <afd80b540603070437x679d88d9ufa1699b869ad833a@mail.gmail.com>
Message-ID: <6a7d49ca0603070733v17bbc565j@mail.gmail.com>

taggable.rb seems to be broken. I'm investigating...

2006/3/7, George Moschovitis <george.moschovitis at gmail.com>:
> Dear devs,
>
> I am about to release 0.29.0. If you have some time please test the
> code in the repo.
> In any case this will be a 'snapshot' release. There is so much work
> in progress, but this release fixes many important bugs (and adds some
> useful features) so I guess it is best to release this 'snapshot'.
>
> regards,
> George.
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From george.moschovitis at gmail.com  Tue Mar  7 10:45:05 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 17:45:05 +0200
Subject: [Nitro] please test...
In-Reply-To: <6a7d49ca0603070733v17bbc565j@mail.gmail.com>
References: <afd80b540603070437x679d88d9ufa1699b869ad833a@mail.gmail.com>
	<6a7d49ca0603070733v17bbc565j@mail.gmail.com>
Message-ID: <afd80b540603070745y294aa3e8jc9b687a02a954352@mail.gmail.com>

too late... but I think taggable works correctly!

-g.

On 3/7/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> taggable.rb seems to be broken. I'm investigating...
>
> 2006/3/7, George Moschovitis <george.moschovitis at gmail.com>:
> > Dear devs,
> >
> > I am about to release 0.29.0. If you have some time please test the
> > code in the repo.
> > In any case this will be a 'snapshot' release. There is so much work
> > in progress, but this release fixes many important bugs (and adds some
> > useful features) so I guess it is best to release this 'snapshot'.
> >
> > regards,
> > George.
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar  7 10:45:23 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 17:45:23 +0200
Subject: [Nitro] please test...
In-Reply-To: <afd80b540603070745y294aa3e8jc9b687a02a954352@mail.gmail.com>
References: <afd80b540603070437x679d88d9ufa1699b869ad833a@mail.gmail.com>
	<6a7d49ca0603070733v17bbc565j@mail.gmail.com>
	<afd80b540603070745y294aa3e8jc9b687a02a954352@mail.gmail.com>
Message-ID: <afd80b540603070745h5887c829t782df0164b656f8b@mail.gmail.com>

after my fixes that is...

-g.

On 3/7/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> too late... but I think taggable works correctly!
>
> -g.
>
> On 3/7/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > taggable.rb seems to be broken. I'm investigating...
> >
> > 2006/3/7, George Moschovitis <george.moschovitis at gmail.com>:
> > > Dear devs,
> > >
> > > I am about to release 0.29.0. If you have some time please test the
> > > code in the repo.
> > > In any case this will be a 'snapshot' release. There is so much work
> > > in progress, but this release fixes many important bugs (and adds some
> > > useful features) so I guess it is best to release this 'snapshot'.
> > >
> > > regards,
> > > George.
> > >
> > >
> > > --
> > > http://www.gmosx.com
> > > http://www.navel.gr
> > > http://www.nitrohq.com
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar  7 10:56:57 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 17:56:57 +0200
Subject: [Nitro] [ANN] Nitro + Og 0.29.0: R operator, Improve error handling,
	routing, template overloafing, memcached store and more
Message-ID: <afd80b540603070756o6b3e5fc0rbceffc1199659a0c@mail.gmail.com>

Dear devs,

The Nitro Development Team is proud to announce new versions of Nitro and Og

homepage: http://www.nitrohq.com
install: gem install nitro
download: http://rubyforge.org/projects/nitro/
trac: http://devlab.oree.ch/trac/nitrohq
irc: irc.freenode.net #nitro
faq: http://oxyliquit.manveru.net
mailing list: http://rubyforge.org/pipermail/nitro-general/


A bold step towards maturity. Great care was taken to
fix reported bugs and fine tune many aspects of Nitro and Og.
As always some great new features where added. Special thanks fly
to Jonas Pfenniger, Bryan Sotto, Rob Pitt and Guillaume
Pierronnet for making this release possible.

Most notable changes:

* Greatly improved error detection and report. Nitro now provides
  line exact error reports at compile or run time, in actions
  or templates. Stay tuned for more improvements in error
  handling.

* Nitro now supports a very useful url endoding/decoding system
  along with a Camping inspired R operator. This new system allows
  you to refer to controllers and actions in your code, without
  bothering about the enforced nice urls and rewrite/routing
  rules. This allows you to write more robust code.

  Here are some examples:

  redirect R(UsersController, :login, :name, 'gmosx')
  <a href="#{R(ForaController, :delete, :id, 3)}">delete</a>

  some more helpers and better integration of this feature
  will be available soon.

* To better support the URL codec, a new router was implemented.
  The new version is both more powerfull and faster. Here are
  some examples:

  r.add_route(%r{rewritten/url/(.*)}, :controller => IdController,
:action => :register, :param => :name)
  r.add_route(%r{another/zelo/(.*)/(.*)}, :controller =>
AdminController, :action => :kick, :params => [:name, :age])
  r.add_route(%r{cool/(.*)_(.*).html}, :controller => AdminController,
:action => :long, :params => [:name, :age])

* Og now supports calculations and aggregations. Here are some
  examples:

  User.min(:age)
  User.average(:age)
  User.maximum(:age)
  User.min(:age, :group => :profession) # => [..] (aggregation)
  User.sum(:age, :group => :role) # => [..]

  and more!

* Improved template_root handling. Now Nitro automatically generates
  a template_root stack for each controller / publishable object.
  Of course the developer can fully customize this stack.
  This is great for reusable modules of functionality (parts). Lets
  say you have implemented a forum part. When you 'mount' the
  forum controller a path relative to the mount point is pushed
  in the template_root stack. When the dispatcher searches for
  a template, it traverses the stack. In the described scenario
  it first looks in the application template root, next in a template
  root the points inside the part template root and then in the
  proto dir template root. The proto dir template root is automatically
  pushed as the last root in the stack. This system facilitates
  'object oriented sites'.

  Use the following method to customize the template_root.

  def self.setup_template_root(path)
    ...
  end

* Added support for JSP like application scoped (global)
  variables.

  def login
    ...
    global[:users] << current_user
    application[:users] << current_user
    ...
  end

  The global/application hash is backed by any of the available
  cache stores.

* Improved Taggable mixin, now provides more helpers and supports
  tag garbage collection through reference counting.

* Added a new store for the generalized caching system that is
  backed by a MemCache server. Useful to extract the last ounch
  of performance in a production environment.

* Fixes in the SCGI adapter.

* Many Og bug fixes and optimizations.

* More flexible static include compiler.

* Improved testing support.

* Integrated latest versions of Scriptaculous and Facets.

* Many, many bug fixes and small improvements throughout the
  code.


Nitro provides everything you need to develop professional Web
applications using Ruby and Javascript. Nitro redefines Rapid
Application Development by providing a clean, yet efficient API,
a layer of domain specific languages implemented on top of
Ruby and the most powerful and elegant object relational
mapping solution available everywhere.


have fun,
George Moschovitis


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar  7 11:03:18 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 7 Mar 2006 18:03:18 +0200
Subject: [Nitro] I will be missing for some days...
Message-ID: <afd80b540603070803u7b399b4wacda97918fd936aa@mail.gmail.com>

Dear devs,

again i will be offline for some days. This time I am now worried though.
I know Nitro is in great hands. I am really looking forward to see
some new community enhancements ;-)

have fun,
George.

PS: This time my absence will be much shorter...


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From james_b at neurogami.com  Tue Mar  7 12:12:42 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 07 Mar 2006 10:12:42 -0700
Subject: [Nitro] [ANN] Nitro + Og 0.29.0: R operator,
 Improve error handling, routing, template overloafing,
 memcached store and more
In-Reply-To: <afd80b540603070756o6b3e5fc0rbceffc1199659a0c@mail.gmail.com>
References: <afd80b540603070756o6b3e5fc0rbceffc1199659a0c@mail.gmail.com>
Message-ID: <440DBF0A.9050401@neurogami.com>

George Moschovitis wrote:
> Dear devs,
> 
> The Nitro Development Team is proud to announce new versions of Nitro and Og
> 
> homepage: http://www.nitrohq.com
> install: gem install nitro
> download: http://rubyforge.org/projects/nitro/
> trac: http://devlab.oree.ch/trac/nitrohq
> irc: irc.freenode.net #nitro
> faq: http://oxyliquit.manveru.net
> mailing list: http://rubyforge.org/pipermail/nitro-general/
> 
> 
> A bold step towards maturity. Great care was taken to
> fix reported bugs and fine tune many aspects of Nitro and Og.
> As always some great new features where added. Special thanks fly
> to Jonas Pfenniger, Bryan Sotto, Rob Pitt and Guillaume
> Pierronnet for making this release possible.
> 

Outstanding.  Thanks to George and the Og/Nitro dev team.


-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://www.30secondrule.com   - Building Better Tools


From rob at motionpath.com  Wed Mar  8 05:08:54 2006
From: rob at motionpath.com (rob)
Date: Wed, 8 Mar 2006 10:08:54 +0000
Subject: [Nitro] Rob's testing improvements.
In-Reply-To: <afd80b540603070245p3f138c6fj3df68b245443be94@mail.gmail.com>
References: <afd80b540603070245p3f138c6fj3df68b245443be94@mail.gmail.com>
Message-ID: <E6EA2B73-B0EB-4101-A4AA-210F47440A96@motionpath.com>

I made it so that it loaded the server map meaning the normal routes  
in your application will work and nothing is needed to begin testing  
other than to load the test library and go:

get(:url => '/whatever/blah')

I also made it so cookies persisted between get calls, i.e. if the  
method above sent some new cookies, the next call to get would act  
like the user had accepted them and was sending them back again.

I added a reset_context method incase you wanted to stop this  
behavior for some reason. If you reset the context, it forgets all  
the cookies and goes back to being a "clean slate".

Can't remember what else off the top of my head, that was probably it :)


On 7 Mar 2006, at 10:45, George Moschovitis wrote:

> Can someone please explain the changes Rob made to the testing  
> framework?
>
> -g.
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From manveru at weez.co.jp  Wed Mar  8 05:37:10 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Wed, 8 Mar 2006 19:37:10 +0900
Subject: [Nitro] Rob's testing improvements.
In-Reply-To: <E6EA2B73-B0EB-4101-A4AA-210F47440A96@motionpath.com>
References: <afd80b540603070245p3f138c6fj3df68b245443be94@mail.gmail.com>
	<E6EA2B73-B0EB-4101-A4AA-210F47440A96@motionpath.com>
Message-ID: <200603081937.10622.manveru@weez.co.jp>

This sounds sweet :)
thanks for your work, i might actually get into TTD with nitro some day...

On Wednesday 08 March 2006 19:08, rob wrote:
> I made it so that it loaded the server map meaning the normal routes
> in your application will work and nothing is needed to begin testing
> other than to load the test library and go:
>
> get(:url => '/whatever/blah')
>
> I also made it so cookies persisted between get calls, i.e. if the
> method above sent some new cookies, the next call to get would act
> like the user had accepted them and was sending them back again.
>
> I added a reset_context method incase you wanted to stop this
> behavior for some reason. If you reset the context, it forgets all
> the cookies and goes back to being a "clean slate".
>
> Can't remember what else off the top of my head, that was probably it :)
>
> On 7 Mar 2006, at 10:45, George Moschovitis wrote:
> > Can someone please explain the changes Rob made to the testing
> > framework?
> >
> > -g.
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general


From rob at motionpath.com  Wed Mar  8 10:30:51 2006
From: rob at motionpath.com (Rob Pitt)
Date: Wed, 8 Mar 2006 15:30:51 +0000
Subject: [Nitro] [PATCH] Stop empty cookies crashing nitro.
Message-ID: <F8C021C6-B37E-4D96-B2B2-24ACDC4B4F08@motionpath.com>

Self-explanatory.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: empty-cookie-fix.patch.bz2
Type: application/octet-stream
Size: 7330 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060308/2f13d1f2/attachment.obj 

From rob at motionpath.com  Wed Mar  8 10:42:33 2006
From: rob at motionpath.com (Rob Pitt)
Date: Wed, 8 Mar 2006 15:42:33 +0000
Subject: [Nitro] Trac
Message-ID: <18A99093-F7CC-465E-ACDC-923B04E96C3E@motionpath.com>

Are we still supposed to throw these patches randomly at the list or  
do we use trac now?


From bryan.a.soto at gmail.com  Wed Mar  8 13:30:59 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 8 Mar 2006 10:30:59 -0800
Subject: [Nitro] Trac
In-Reply-To: <18A99093-F7CC-465E-ACDC-923B04E96C3E@motionpath.com>
References: <18A99093-F7CC-465E-ACDC-923B04E96C3E@motionpath.com>
Message-ID: <bd93db0d0603081030gb6f58fs936b64cb58f1b065@mail.gmail.com>

On 3/8/06, Rob Pitt <rob at motionpath.com> wrote:
> Are we still supposed to throw these patches randomly at the list or
> do we use trac now?

You're on the core team to, so I'd suppose your opinion would carry
some weight. :)

What are your thoughts? Though reading your question makes me think
you favor the latter.



From george.moschovitis at gmail.com  Wed Mar  8 14:51:39 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 8 Mar 2006 21:51:39 +0200
Subject: [Nitro] Trac
In-Reply-To: <bd93db0d0603081030gb6f58fs936b64cb58f1b065@mail.gmail.com>
References: <18A99093-F7CC-465E-ACDC-923B04E96C3E@motionpath.com>
	<bd93db0d0603081030gb6f58fs936b64cb58f1b065@mail.gmail.com>
Message-ID: <afd80b540603081151v27fb7d0fi481c0204083607fe@mail.gmail.com>

Hello,

I suggest you use trac. But, I would like to see the patches posted to
the list with the [PATCH] Subject prefix too.

regards,
George.

On 3/8/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/8/06, Rob Pitt <rob at motionpath.com> wrote:
> > Are we still supposed to throw these patches randomly at the list or
> > do we use trac now?
>
> You're on the core team to, so I'd suppose your opinion would carry
> some weight. :)
>
> What are your thoughts? Though reading your question makes me think
> you favor the latter.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar  8 14:53:29 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 8 Mar 2006 21:53:29 +0200
Subject: [Nitro] Rob's testing improvements.
In-Reply-To: <E6EA2B73-B0EB-4101-A4AA-210F47440A96@motionpath.com>
References: <afd80b540603070245p3f138c6fj3df68b245443be94@mail.gmail.com>
	<E6EA2B73-B0EB-4101-A4AA-210F47440A96@motionpath.com>
Message-ID: <afd80b540603081153w1ca1e62dr8e24ec42c3225f60@mail.gmail.com>

Ok nice ;-)

-g.

On 3/8/06, rob <rob at motionpath.com> wrote:
> I made it so that it loaded the server map meaning the normal routes
> in your application will work and nothing is needed to begin testing
> other than to load the test library and go:
>
> get(:url => '/whatever/blah')
>
> I also made it so cookies persisted between get calls, i.e. if the
> method above sent some new cookies, the next call to get would act
> like the user had accepted them and was sending them back again.
>
> I added a reset_context method incase you wanted to stop this
> behavior for some reason. If you reset the context, it forgets all
> the cookies and goes back to being a "clean slate".
>
> Can't remember what else off the top of my head, that was probably it :)
>
>
> On 7 Mar 2006, at 10:45, George Moschovitis wrote:
>
> > Can someone please explain the changes Rob made to the testing
> > framework?
> >
> > -g.
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Wed Mar  8 15:54:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 8 Mar 2006 12:54:23 -0800
Subject: [Nitro] Kirbybase/inheritance/schema evolution
In-Reply-To: <440D2C1F.3080402@digitalvalence.com>
References: <bd93db0d0603052339t25b8a374u7fb3b7348b70d132@mail.gmail.com>
	<afd80b540603060243s6ca4dbfek33858e2ca9da0e5a@mail.gmail.com>
	<440D2C1F.3080402@digitalvalence.com>
Message-ID: <bd93db0d0603081254r1d8cbf59qd021fcd614307467@mail.gmail.com>

Hi Dylan,

On 3/6/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
> I recently started implementing a desktop application using OG to
> connect to KirbyBase as the back end.  While implementing my initial
> model tests I came across a couple of problems with the schema
> inheritance and schema evolution features.
>
> 1.  Schema inheritance creates extra columns for each subclass
> property.  Say we have a super class, Person, which has two children,
> Manager and Programmer.  If Manager has an employees property and
> Programmer has a cupsOfCoffee property schema evolution will drop the
> column for whichever property came first.  This may be a general
> inheritance and evolution problem, not just specific to Kirbybase.  I
> haven't checked the other stores for more advanced methods of handling
> this, such as using descendents.

I think this might be the problem outlined at:
http://rubyforge.org/pipermail/nitro-general/2006-January/002601.html

It should be fixed in the current release 0.29.0.

<snipped too much to digest />

Wow, I'll have to save this for when I have some free time. ;)

Kirby has a few issues because it doesn't natively support joins in
it's queries. If you're looking at deploying, rather than server based
stuff, you might be better off going with Sqlite. But if your fixes
work, I do hope to see a patch. You sound experienced and we can use
all the help we can get. :)

To get rid of those false changes:

darcs diff -u
* Shows changes in unified diff format.

darcs revert [optional filename]
* Will iterate through changes [entire repo if no filename given].
** y to revert
** n to ignore
** q to quit
** d for done
** a for all (sames as answering yes to each change)

darcs record [optional filename]
* records changes [entire repo if no filename given]
** commands as in revert

darcs send -o bundle_name (Need to be connected to Internet for this)
* Outputs named changes [record] to file
** commands as in revert

Optionally, you can just copy the files you want to save, darcs revert
the entire repo, then move them back to their proper place. ;)

Then email bundle to list or submit on Trac http://devlab.oree.ch/trac/nitrohq.

Thanks,

Bryan



From bryan.a.soto at gmail.com  Wed Mar  8 16:47:03 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 8 Mar 2006 13:47:03 -0800
Subject: [Nitro] Og tests...
In-Reply-To: <afd80b540603070112r372d1d20p7dab942a5ea72ea7@mail.gmail.com>
References: <afd80b540603070112r372d1d20p7dab942a5ea72ea7@mail.gmail.com>
Message-ID: <bd93db0d0603081347p31275c66ud2e50bc6e25706cb@mail.gmail.com>

On 3/7/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Bryan
> I am sorry to dissapoint you but rake test does not run all tests (it
> doesnt run tests in test/mixin for example). Saddly test like
> tc_taggable do not pass (and tags seem a bit f*cked up). Anyway, I am
> working on this...

Hmm... tests in mixin were failing for me on both Gentoo Linux and WinXP.

>From a fresh checkout, I added

+  def test_zzz
+    fail("This is the last file on the command line and should fail
so we know it ran.")
+  end
+

to the last file in the Rake generated command line (in my case,
og/test/glue/tc_revisable.rb).

$ RUBYOPT=-rubygems rake test
(in /home/bryan/checkout/darcs/repo.nitrohq.com/og)
/usr/bin/ruby18
-Ilib:test:/home/bryan/checkout/darcs/repo.nitrohq.com/glue/lib:/home/bryan/checkout/darcs/repo.nitrohq.com/nitro/lib:/home/bryan/checkout/darcs/repo.nitrohq.com/og/lib
"/usr/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb"
"test/og/tc_scoped.rb" "test/og/tc_store.rb" "test/og/tc_cacheable.rb"
"test/og/tc_validation.rb" "test/og/tc_inheritance2.rb"
"test/og/tc_accumulator.rb" "test/og/tc_validation_loop.rb"
"test/og/tc_polymorphic.rb" "test/og/tc_relation.rb"
"test/og/tc_types.rb" "test/og/tc_resolve.rb" "test/og/tc_override.rb"
"test/og/tc_join.rb" "test/og/tc_delete_all.rb"
"test/og/tc_aggregations_calculations.rb" "test/og/tc_validation2.rb"
"test/og/tc_select.rb" "test/og/tc_multiple.rb"
"test/og/tc_multi_validations.rb" "test/og/tc_ez.rb"
"test/og/tc_finder.rb" "test/og/tc_reverse.rb"
"test/og/tc_inheritance.rb" "test/og/mixin/tc_hierarchical.rb"
"test/og/mixin/tc_timestamped.rb" "test/og/mixin/tc_orderable.rb"
"test/og/mixin/tc_optimistic_locking.rb"
"test/og/mixin/tc_taggable.rb" "test/og/store/tc_kirby.rb"
"test/og/store/tc_sti.rb" "test/og/store/tc_filesys.rb"
"test/glue/tc_revisable.rb"
Database "test" dropped
Database "test" dropped
Loaded suite /usr/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader
Started
.........................................E"-- count = 0 -> delete"
"-- count = 0 -> delete"
"-- count = 0 -> delete"
..
Finished in 7.276416 seconds.

  1) Error:
test_zzz(TestOgRevisable):
RuntimeError: This is the last file on the command line and should
fail so we know it ran.
    ./test/glue/tc_revisable.rb:23:in `test_zzz'

44 tests, 221 assertions, 0 failures, 1 errors

Do you get something different?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From kgergely at mcl.hu  Thu Mar  9 07:11:15 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Thu, 9 Mar 2006 13:11:15 +0100
Subject: [Nitro] multilingual model
Message-ID: <20060309121115.GA30712@mlabdial.hit.bme.hu>

Hi!

Is there a support for multilingual model generation in nitro?

Is there a search support for the mailing list?

thx
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/17022921/attachment.bin 

From kgergely at mcl.hu  Thu Mar  9 07:17:26 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Thu, 9 Mar 2006 13:17:26 +0100
Subject: [Nitro] switching from sqlite to mysql
Message-ID: <20060309121726.GB30712@mlabdial.hit.bme.hu>

Hi!

I'm stucked at the point switching from sqlite tomysql.

I get the error below. Can sy. please help to figure it out what's going 
on?

thx

valaki at debian-server:~/sundiszno-nitro$ ruby -rubygems start.rb
D, [2006-03-08T14:10:50.397856 #12268] DEBUG -- : Using Memory sessions.
I, [2006-03-08T14:10:51.536405 #12268]  INFO -- : Og uses the Mysql store.
E, [2006-03-08T14:10:51.601388 #12268] ERROR -- : Og.setup had problems: NameError => undefined method `fetch_row' for class `Mysql::Result'
/usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/scaffolding.rb:325:in `compile_scaffolding_code': undefined method `manageable_classes' for nil:NilClass (NoMethodError)
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/controller.rb:270:in `mounted'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/dispatcher.rb:85:in `mount'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/dispatcher.rb:66:in `each'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/dispatcher.rb:66:in `mount'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/dispatcher.rb:40:in `initialize'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/server.rb:89:in `new'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/server.rb:89:in `start'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro/server.rb:132:in `run'
        from /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/nitro.rb:77:in `start'
        from start.rb:21


-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/fa62c4e0/attachment.bin 

From kashia at vfemail.net  Thu Mar  9 07:42:52 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 09 Mar 2006 13:42:52 +0100
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <20060309121726.GB30712@mlabdial.hit.bme.hu>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
Message-ID: <op.s55b9qoeg3glhz@jo.local>

Hi,

> I get the error below. Can sy. please help to figure it out what's going
> on?
>E, [2006-03-08T14:10:51.601388 #12268] ERROR -- : Og.setup had problems: NameError => undefined method `fetch_row' for class `Mysql::Result'

That sounds like you're using 'fetch_row', which doesn't seem to be supported by the Mysql adapter.
If your goal is, to just get one row, you can use the following:

  obj = FooBar.find_one(:condition......)

This returns a single row.

Hope that helped,

Kash

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kashia at vfemail.net  Thu Mar  9 07:52:16 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 09 Mar 2006 13:52:16 +0100
Subject: [Nitro] multilingual model
In-Reply-To: <20060309121115.GA30712@mlabdial.hit.bme.hu>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
Message-ID: <op.s55cpenug3glhz@jo.local>

On Thu, 09 Mar 2006 13:11:15 +0100, Gergely Kontra <kgergely at mcl.hu> wrote:

> Hi!
>
> Is there a support for multilingual model generation in nitro?

I'm pretty much certain that something like this doesn't exist yet. What do you mean by "model generation"? As far as I know, you still have to create the models yourself.

> Is there a search support for the mailing list?

I looked for it, but couldn't find any search function, sorry to disappoint you.
But, if you have specific questions, you can try to search/post at http://oxyliquit.manveru.net

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kgergely at mcl.hu  Thu Mar  9 07:53:45 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Thu, 9 Mar 2006 13:53:45 +0100
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <op.s55b9qoeg3glhz@jo.local>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
	<op.s55b9qoeg3glhz@jo.local>
Message-ID: <20060309125345.GC30712@mlabdial.hit.bme.hu>

Kashia Buch azt ?rta, hogy: 
> > I get the error below. Can sy. please help to figure it out what's going
> > on?
> >E, [2006-03-08T14:10:51.601388 #12268] ERROR -- : Og.setup had problems: NameError => undefined method `fetch_row' for class `Mysql::Result'
> 
> That sounds like you're using 'fetch_row', which doesn't seem to be supported by the Mysql adapter.
> If your goal is, to just get one row, you can use the following:
> 
>   obj = FooBar.find_one(:condition......)

No, I don't use fetch_row at all!
This happens at startup with a very minimal model and controller. (I can 
attach the source if needed...)
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/bfb2dac2/attachment.bin 

From rob at motionpath.com  Thu Mar  9 07:55:11 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 12:55:11 +0000
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <op.s55b9qoeg3glhz@jo.local>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
	<op.s55b9qoeg3glhz@jo.local>
Message-ID: <FE8D6580-2006-4F4F-B6EA-20715181ACCB@motionpath.com>

It's easily possible I committed a fix to STI that tried to use that  
method.

Bryan... help! :D

(Suggest add this method to the mysql adapter)

On 9 Mar 2006, at 12:42, Kashia Buch wrote:

> Hi,
>
>> I get the error below. Can sy. please help to figure it out what's  
>> going
>> on?
>> E, [2006-03-08T14:10:51.601388 #12268] ERROR -- : Og.setup had  
>> problems: NameError => undefined method `fetch_row' for class  
>> `Mysql::Result'
>
> That sounds like you're using 'fetch_row', which doesn't seem to be  
> supported by the Mysql adapter.
> If your goal is, to just get one row, you can use the following:
>
>   obj = FooBar.find_one(:condition......)
>
> This returns a single row.
>
> Hope that helped,
>
> Kash
>
> -- 
> Feel the love
> http://pinkjuice.com/pics/ruby.png
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From rob at motionpath.com  Thu Mar  9 08:00:33 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 13:00:33 +0000
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <20060309125345.GC30712@mlabdial.hit.bme.hu>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
	<op.s55b9qoeg3glhz@jo.local>
	<20060309125345.GC30712@mlabdial.hit.bme.hu>
Message-ID: <9C1B0995-552C-49FF-9E61-132ADB05EA83@motionpath.com>

This is my fault I know why this happens.

Would be nice if someone else could fix it since I don't generally  
meddle with mysql.

The problem is one of the STI fixes I supplied recently uses  
fetch_row inside of sql.rb.

On 9 Mar 2006, at 12:53, Gergely Kontra wrote:

> Kashia Buch azt ?rta, hogy:
>>> I get the error below. Can sy. please help to figure it out  
>>> what's going
>>> on?
>>> E, [2006-03-08T14:10:51.601388 #12268] ERROR -- : Og.setup had  
>>> problems: NameError => undefined method `fetch_row' for class  
>>> `Mysql::Result'
>>
>> That sounds like you're using 'fetch_row', which doesn't seem to  
>> be supported by the Mysql adapter.
>> If your goal is, to just get one row, you can use the following:
>>
>>   obj = FooBar.find_one(:condition......)
>
> No, I don't use fetch_row at all!
> This happens at startup with a very minimal model and controller.  
> (I can
> attach the source if needed...)
> -- 
> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/ 
> ~kgergely ]------+
> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356  
> 9656 |
> | Budapest University of Technology and Economics        ICQ#  
> 175564914 |
> +------------- "Olyan langesz vagyok, hogy poroltoval kellene  
> jarnom!" -+
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general




From kgergely at mcl.hu  Thu Mar  9 08:03:45 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Thu, 9 Mar 2006 14:03:45 +0100
Subject: [Nitro] multilingual model
In-Reply-To: <op.s55cpenug3glhz@jo.local>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
	<op.s55cpenug3glhz@jo.local>
Message-ID: <20060309130345.GD30712@mlabdial.hit.bme.hu>

Kashia Buch azt ?rta, hogy: 
> On Thu, 09 Mar 2006 13:11:15 +0100, Gergely Kontra <kgergely at mcl.hu> wrote:
> 
> > Hi!
> >
> > Is there a support for multilingual model generation in nitro?
> 
> I'm pretty much certain that something like this doesn't exist yet. 
> What do you mean by "model generation"? As far as I know, you still 
> have to create the models yourself.

Like this (hope ge the idea despite of the unchecked code written here):

class Blog
  property :author, String
  localized :body
end

would automagically create:
# globally one:
class Locale
  property :shortname, Varchar(2)
  property :longname, Varchar(64)
end

class Blog
  property :author
  has_many :body Blog_i18n
end

class Blog_i18n
  property :author, String
  belongs_to :locale, Locale
end

> I looked for it, but couldn't find any search function, sorry to 
> disappoint you.
> But, if you have specific questions, you can try to search/post at 
> http://oxyliquit.manveru.net

Not too much doc in the air, but I'll try to get my firm switch to nitro 
anyway...

thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/41859821/attachment.bin 

From rob at motionpath.com  Thu Mar  9 08:17:29 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 13:17:29 +0000
Subject: [Nitro] multilingual model
In-Reply-To: <20060309130345.GD30712@mlabdial.hit.bme.hu>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
	<op.s55cpenug3glhz@jo.local>
	<20060309130345.GD30712@mlabdial.hit.bme.hu>
Message-ID: <0FF8DB88-DF82-4D4C-BF11-6467AC2C0DB8@motionpath.com>

I am currently working on a multilanguage project.

I have created a relation abstractor where you define properties like:

localized_strings :body

In the model. Then to access the strings for a language you go:

record.body[@language].to_s
record.body[@language] = "some text"

or

record.body[@country].to_s
record.body[@country] = "some text"

It also comes with an extensive test case that is project specific  
but shows you how it works and could easily be modified to use with  
other applications.

If you think you will find this useful I will e-mail it to you.

On 9 Mar 2006, at 13:03, Gergely Kontra wrote:

> Kashia Buch azt ?rta, hogy:
>> On Thu, 09 Mar 2006 13:11:15 +0100, Gergely Kontra  
>> <kgergely at mcl.hu> wrote:
>>
>>> Hi!
>>>
>>> Is there a support for multilingual model generation in nitro?
>>
>> I'm pretty much certain that something like this doesn't exist yet.
>> What do you mean by "model generation"? As far as I know, you still
>> have to create the models yourself.
>
> Like this (hope ge the idea despite of the unchecked code written  
> here):
>
> class Blog
>   property :author, String
>   localized :body
> end
>
> would automagically create:
> # globally one:
> class Locale
>   property :shortname, Varchar(2)
>   property :longname, Varchar(64)
> end
>
> class Blog
>   property :author
>   has_many :body Blog_i18n
> end
>
> class Blog_i18n
>   property :author, String
>   belongs_to :locale, Locale
> end
>
>> I looked for it, but couldn't find any search function, sorry to
>> disappoint you.
>> But, if you have specific questions, you can try to search/post at
>> http://oxyliquit.manveru.net
>
> Not too much doc in the air, but I'll try to get my firm switch to  
> nitro
> anyway...
>
> thx
> Gergo
> -- 
> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/ 
> ~kgergely ]------+
> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356  
> 9656 |
> | Budapest University of Technology and Economics        ICQ#  
> 175564914 |
> +------------- "Olyan langesz vagyok, hogy poroltoval kellene  
> jarnom!" -+
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general




From rob at motionpath.com  Thu Mar  9 08:28:11 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 13:28:11 +0000
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <9C1B0995-552C-49FF-9E61-132ADB05EA83@motionpath.com>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
	<op.s55b9qoeg3glhz@jo.local>
	<20060309125345.GC30712@mlabdial.hit.bme.hu>
	<9C1B0995-552C-49FF-9E61-132ADB05EA83@motionpath.com>
Message-ID: <2743C0BD-76A9-4C46-A788-FA4460761771@motionpath.com>

If nobody does fix it I will fix it at the weekend.

To be very sure that this is the cause though (I am only 90% certain)  
it would be nice if you e-mailed me your complete application to  
reproduce the error.

On 9 Mar 2006, at 13:00, Rob Pitt wrote:

> This is my fault I know why this happens.
>
> Would be nice if someone else could fix it since I don't generally
> meddle with mysql.
>
> The problem is one of the STI fixes I supplied recently uses
> fetch_row inside of sql.rb.
>
> On 9 Mar 2006, at 12:53, Gergely Kontra wrote:
>
>> Kashia Buch azt ?rta, hogy:
>>>> I get the error below. Can sy. please help to figure it out
>>>> what's going
>>>> on?
>>>> E, [2006-03-08T14:10:51.601388 #12268] ERROR -- : Og.setup had
>>>> problems: NameError => undefined method `fetch_row' for class
>>>> `Mysql::Result'
>>>
>>> That sounds like you're using 'fetch_row', which doesn't seem to
>>> be supported by the Mysql adapter.
>>> If your goal is, to just get one row, you can use the following:
>>>
>>>   obj = FooBar.find_one(:condition......)
>>
>> No, I don't use fetch_row at all!
>> This happens at startup with a very minimal model and controller.
>> (I can
>> attach the source if needed...)
>> -- 
>> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/
>> ~kgergely ]------+
>> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356
>> 9656 |
>> | Budapest University of Technology and Economics        ICQ#
>> 175564914 |
>> +------------- "Olyan langesz vagyok, hogy poroltoval kellene
>> jarnom!" -+
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general




From vseguip at gmail.com  Thu Mar  9 08:47:29 2006
From: vseguip at gmail.com (vseguip at gmail.com)
Date: Thu, 9 Mar 2006 14:47:29 +0100
Subject: [Nitro] [PATCH] Entity copying methods
In-Reply-To: <afd80b540602010305o3a0ee6ewd13152118750b608@mail.gmail.com>
References: <1138372566.2932.34.camel@robs-p4>
	<dda3b98e0601270758w1bfd2792v1935be08733031c8@mail.gmail.com>
	<1138380314.2932.40.camel@robs-p4>
	<afd80b540602010116k7032afc5j7095938b298489af@mail.gmail.com>
	<1138789134.4872.22.camel@robs-p4>
	<afd80b540602010305o3a0ee6ewd13152118750b608@mail.gmail.com>
Message-ID: <dda3b98e0603090547q4651643exafaa5f3952851910@mail.gmail.com>

On 2/1/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Rob,
>
> one other important thing. Please try to follow the coding conventiosn
> a bit closer.
>
> thanks in advance,
> -g.
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From vseguip at gmail.com  Thu Mar  9 08:47:42 2006
From: vseguip at gmail.com (vseguip at gmail.com)
Date: Thu, 9 Mar 2006 14:47:42 +0100
Subject: [Nitro] [PATCH] Entity copying methods
In-Reply-To: <1138789134.4872.22.camel@robs-p4>
References: <1138372566.2932.34.camel@robs-p4>
	<dda3b98e0601270758w1bfd2792v1935be08733031c8@mail.gmail.com>
	<1138380314.2932.40.camel@robs-p4>
	<afd80b540602010116k7032afc5j7095938b298489af@mail.gmail.com>
	<1138789134.4872.22.camel@robs-p4>
Message-ID: <dda3b98e0603090547x14d3d2cfp3de4f55d45afd16c@mail.gmail.com>

 Hi Rob,

   I was was just wondering if you finally got to implement the copy
of "join through" data. I'm interested since my join through tables
actually relate to other tables and it's actually a PITA the copy
them.

Thanks for you time,
  V. Segu?

On 2/1/06, Rob Pitt <rob at motionpath.com> wrote:
> Sorry I haven't had chance the office has been a bit of a war zone these
> past couple of days and I've had to focus on our core projects. The
> patch does work as is, the only thing it lacks is the cloning options I
> said I would provide and the ability to copy "join through" information
> (and still will do hopefully sometime this week). You can go ahead and
> integrate this the cloning option can wait until the next version.
>
> I also note you have contacted Chris in regards to us hosting a DARCS
> repository for Glycerin which we would be happy to do on one of our
> servers in Telehouse Docklands. When does this process need completion?
>
> On Wed, 2006-02-01 at 11:16 +0200, George Moschovitis wrote:
> > Rob,
> >
> > if  you have a new version of this patch please resend it, I would
> > like to attempt to integrate this.
> >
> > thanks in advance,
> > George.
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From rob at motionpath.com  Thu Mar  9 08:52:28 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 13:52:28 +0000
Subject: [Nitro] multilingual model
In-Reply-To: <20060309133729.GA912@mlabdial.hit.bme.hu>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
	<op.s55cpenug3glhz@jo.local>
	<20060309130345.GD30712@mlabdial.hit.bme.hu>
	<0FF8DB88-DF82-4D4C-BF11-6467AC2C0DB8@motionpath.com>
	<20060309133729.GA912@mlabdial.hit.bme.hu>
Message-ID: <10DDC978-359E-4A1F-B8F9-33D93147D42E@motionpath.com>

Here you go. I think I included all the things you will need.

You will need to remove the reference to monkeynuts (this is one of  
my libraries for "hacks"). I have included the relevant file from the  
library (It used a post-setup hook on Og) and the relevant models/ 
test case.

You will need to do some editing to them (minor, probably just remove  
the monkeynuts references) , and to the unit test (some of the setup  
stuff will be missing), but it should give you an idea.

You can safely ignore all those finder methods in localizer if you  
don't know what they mean, if you don't want them remove them :)

I cc'd it to the Nitro list too incase anyone wants to develop the  
concept and make it a proper part of Nitro.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: multilanguage.tbz2
Type: application/octet-stream
Size: 3539 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/97d1bf20/attachment.obj 
-------------- next part --------------

On 9 Mar 2006, at 13:37, Gergely Kontra wrote:

> Rob Pitt azt ?rta, hogy:
>> If you think you will find this useful I will e-mail it to you.
>
> Thanks, please mail it to me! (In what stage is it?)
> --  
> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/ 
> ~kgergely ]------+
> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356  
> 9656 |
> | Budapest University of Technology and Economics        ICQ#  
> 175564914 |
> +------------- "Olyan langesz vagyok, hogy poroltoval kellene  
> jarnom!" -+


From rob at motionpath.com  Thu Mar  9 08:56:34 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 13:56:34 +0000
Subject: [Nitro] multilingual model
In-Reply-To: <10DDC978-359E-4A1F-B8F9-33D93147D42E@motionpath.com>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
	<op.s55cpenug3glhz@jo.local>
	<20060309130345.GD30712@mlabdial.hit.bme.hu>
	<0FF8DB88-DF82-4D4C-BF11-6467AC2C0DB8@motionpath.com>
	<20060309133729.GA912@mlabdial.hit.bme.hu>
	<10DDC978-359E-4A1F-B8F9-33D93147D42E@motionpath.com>
Message-ID: <5B52B83C-5102-4404-BBBE-A96546103F89@motionpath.com>

Actually, I looked at the code and you should probably just delete  
og_setup_hooks and the section of country.rb that uses it.

If not you will need to change the bit at the end of country. It is  
supposed to throw an exception if you try and set language to an  
object that hasn't been saved but I used something I forgot to  
include to do this and you may well not want it to behave like that  
anyway.

On 9 Mar 2006, at 13:52, Rob Pitt wrote:

> Here you go. I think I included all the things you will need.
>
> You will need to remove the reference to monkeynuts (this is one of  
> my libraries for "hacks"). I have included the relevant file from  
> the library (It used a post-setup hook on Og) and the relevant  
> models/test case.
>
> You will need to do some editing to them (minor, probably just  
> remove the monkeynuts references) , and to the unit test (some of  
> the setup stuff will be missing), but it should give you an idea.
>
> You can safely ignore all those finder methods in localizer if you  
> don't know what they mean, if you don't want them remove them :)
>
> I cc'd it to the Nitro list too incase anyone wants to develop the  
> concept and make it a proper part of Nitro.
>
> <multilanguage.tbz2>
>
> On 9 Mar 2006, at 13:37, Gergely Kontra wrote:
>
>> Rob Pitt azt ?rta, hogy:
>>> If you think you will find this useful I will e-mail it to you.
>>
>> Thanks, please mail it to me! (In what stage is it?)
>> -- +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/ 
>> ~kgergely ]------+
>> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356  
>> 9656 |
>> | Budapest University of Technology and Economics        ICQ#  
>> 175564914 |
>> +------------- "Olyan langesz vagyok, hogy poroltoval kellene  
>> jarnom!" -+
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general




From rob at motionpath.com  Thu Mar  9 08:56:58 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 13:56:58 +0000
Subject: [Nitro] [PATCH] Entity copying methods
In-Reply-To: <dda3b98e0603090547x14d3d2cfp3de4f55d45afd16c@mail.gmail.com>
References: <1138372566.2932.34.camel@robs-p4>
	<dda3b98e0601270758w1bfd2792v1935be08733031c8@mail.gmail.com>
	<1138380314.2932.40.camel@robs-p4>
	<afd80b540602010116k7032afc5j7095938b298489af@mail.gmail.com>
	<1138789134.4872.22.camel@robs-p4>
	<dda3b98e0603090547x14d3d2cfp3de4f55d45afd16c@mail.gmail.com>
Message-ID: <FB71CBE5-A0C2-4252-B1E6-9A57F3975B45@motionpath.com>

Sorry I haven't. I've been ill and now I'm snowed under with work!

On 9 Mar 2006, at 13:47, vseguip at gmail.com wrote:

>  Hi Rob,
>
>    I was was just wondering if you finally got to implement the copy
> of "join through" data. I'm interested since my join through tables
> actually relate to other tables and it's actually a PITA the copy
> them.
>
> Thanks for you time,
>   V. Segu?
>
> On 2/1/06, Rob Pitt <rob at motionpath.com> wrote:
>> Sorry I haven't had chance the office has been a bit of a war zone  
>> these
>> past couple of days and I've had to focus on our core projects. The
>> patch does work as is, the only thing it lacks is the cloning  
>> options I
>> said I would provide and the ability to copy "join through"  
>> information
>> (and still will do hopefully sometime this week). You can go ahead  
>> and
>> integrate this the cloning option can wait until the next version.
>>
>> I also note you have contacted Chris in regards to us hosting a DARCS
>> repository for Glycerin which we would be happy to do on one of our
>> servers in Telehouse Docklands. When does this process need  
>> completion?
>>
>> On Wed, 2006-02-01 at 11:16 +0200, George Moschovitis wrote:
>>> Rob,
>>>
>>> if  you have a new version of this patch please resend it, I would
>>> like to attempt to integrate this.
>>>
>>> thanks in advance,
>>> George.
>>>
>>>
>>> --
>>> http://www.gmosx.com
>>> http://www.navel.gr
>>> http://www.nitrohq.com
>>>
>>> _______________________________________________
>>> Nitro-general mailing list
>>> Nitro-general at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general




From george.moschovitis at gmail.com  Thu Mar  9 09:06:17 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Mar 2006 16:06:17 +0200
Subject: [Nitro] Og tests...
In-Reply-To: <bd93db0d0603081347p31275c66ud2e50bc6e25706cb@mail.gmail.com>
References: <afd80b540603070112r372d1d20p7dab942a5ea72ea7@mail.gmail.com>
	<bd93db0d0603081347p31275c66ud2e50bc6e25706cb@mail.gmail.com>
Message-ID: <afd80b540603090606g1bee09a4hc17caf8067dd84e0@mail.gmail.com>

Bryan,

if you have some time, please have a look at those... I will probably
have some time to work on nitro later tonight (I also have a lot of
office related stuff).

-g.


On 3/8/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/7/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Bryan
> > I am sorry to dissapoint you but rake test does not run all tests (it
> > doesnt run tests in test/mixin for example). Saddly test like
> > tc_taggable do not pass (and tags seem a bit f*cked up). Anyway, I am
> > working on this...
>
> Hmm... tests in mixin were failing for me on both Gentoo Linux and WinXP.
>
> >From a fresh checkout, I added
>
> +  def test_zzz
> +    fail("This is the last file on the command line and should fail
> so we know it ran.")
> +  end
> +
>
> to the last file in the Rake generated command line (in my case,
> og/test/glue/tc_revisable.rb).
>
> $ RUBYOPT=-rubygems rake test
> (in /home/bryan/checkout/darcs/repo.nitrohq.com/og)
> /usr/bin/ruby18
> -Ilib:test:/home/bryan/checkout/darcs/repo.nitrohq.com/glue/lib:/home/bryan/checkout/darcs/repo.nitrohq.com/nitro/lib:/home/bryan/checkout/darcs/repo.nitrohq.com/og/lib
> "/usr/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb"
> "test/og/tc_scoped.rb" "test/og/tc_store.rb" "test/og/tc_cacheable.rb"
> "test/og/tc_validation.rb" "test/og/tc_inheritance2.rb"
> "test/og/tc_accumulator.rb" "test/og/tc_validation_loop.rb"
> "test/og/tc_polymorphic.rb" "test/og/tc_relation.rb"
> "test/og/tc_types.rb" "test/og/tc_resolve.rb" "test/og/tc_override.rb"
> "test/og/tc_join.rb" "test/og/tc_delete_all.rb"
> "test/og/tc_aggregations_calculations.rb" "test/og/tc_validation2.rb"
> "test/og/tc_select.rb" "test/og/tc_multiple.rb"
> "test/og/tc_multi_validations.rb" "test/og/tc_ez.rb"
> "test/og/tc_finder.rb" "test/og/tc_reverse.rb"
> "test/og/tc_inheritance.rb" "test/og/mixin/tc_hierarchical.rb"
> "test/og/mixin/tc_timestamped.rb" "test/og/mixin/tc_orderable.rb"
> "test/og/mixin/tc_optimistic_locking.rb"
> "test/og/mixin/tc_taggable.rb" "test/og/store/tc_kirby.rb"
> "test/og/store/tc_sti.rb" "test/og/store/tc_filesys.rb"
> "test/glue/tc_revisable.rb"
> Database "test" dropped
> Database "test" dropped
> Loaded suite /usr/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader
> Started
> .........................................E"-- count = 0 -> delete"
> "-- count = 0 -> delete"
> "-- count = 0 -> delete"
> ..
> Finished in 7.276416 seconds.
>
>   1) Error:
> test_zzz(TestOgRevisable):
> RuntimeError: This is the last file on the command line and should
> fail so we know it ran.
>     ./test/glue/tc_revisable.rb:23:in `test_zzz'
>
> 44 tests, 221 assertions, 0 failures, 1 errors
>
> Do you get something different?
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar  9 09:12:00 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Mar 2006 16:12:00 +0200
Subject: [Nitro] multilingual model
In-Reply-To: <10DDC978-359E-4A1F-B8F9-33D93147D42E@motionpath.com>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
	<op.s55cpenug3glhz@jo.local>
	<20060309130345.GD30712@mlabdial.hit.bme.hu>
	<0FF8DB88-DF82-4D4C-BF11-6467AC2C0DB8@motionpath.com>
	<20060309133729.GA912@mlabdial.hit.bme.hu>
	<10DDC978-359E-4A1F-B8F9-33D93147D42E@motionpath.com>
Message-ID: <afd80b540603090612u78ad5b5cv11ec414e0aecfc5d@mail.gmail.com>

Thanks, I will have a look... While you are at it, can you email your
multipart enhancemennts and/or upload progress modifications?

thanks,
George.



On 3/9/06, Rob Pitt <rob at motionpath.com> wrote:
> Here you go. I think I included all the things you will need.
>
> You will need to remove the reference to monkeynuts (this is one of
> my libraries for "hacks"). I have included the relevant file from the
> library (It used a post-setup hook on Og) and the relevant models/
> test case.
>
> You will need to do some editing to them (minor, probably just remove
> the monkeynuts references) , and to the unit test (some of the setup
> stuff will be missing), but it should give you an idea.
>
> You can safely ignore all those finder methods in localizer if you
> don't know what they mean, if you don't want them remove them :)
>
> I cc'd it to the Nitro list too incase anyone wants to develop the
> concept and make it a proper part of Nitro.
>
>
>
>
> On 9 Mar 2006, at 13:37, Gergely Kontra wrote:
>
> > Rob Pitt azt ?rta, hogy:
> >> If you think you will find this useful I will e-mail it to you.
> >
> > Thanks, please mail it to me! (In what stage is it?)
> > --
> > +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/
> > ~kgergely ]------+
> > | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356
> > 9656 |
> > | Budapest University of Technology and Economics        ICQ#
> > 175564914 |
> > +------------- "Olyan langesz vagyok, hogy poroltoval kellene
> > jarnom!" -+
>
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From rob at motionpath.com  Thu Mar  9 09:15:22 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 14:15:22 +0000
Subject: [Nitro] multilingual model
In-Reply-To: <afd80b540603090612u78ad5b5cv11ec414e0aecfc5d@mail.gmail.com>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
	<op.s55cpenug3glhz@jo.local>
	<20060309130345.GD30712@mlabdial.hit.bme.hu>
	<0FF8DB88-DF82-4D4C-BF11-6467AC2C0DB8@motionpath.com>
	<20060309133729.GA912@mlabdial.hit.bme.hu>
	<10DDC978-359E-4A1F-B8F9-33D93147D42E@motionpath.com>
	<afd80b540603090612u78ad5b5cv11ec414e0aecfc5d@mail.gmail.com>
Message-ID: <31884F4A-6FC3-4366-9B87-53A85D46CD79@motionpath.com>

Hehe OK G I will send you them at the weekend but seriously I did it  
all wrong, it could be done better by parsing character by character  
without regex and not in blocks like I did it.

On 9 Mar 2006, at 14:12, George Moschovitis wrote:

> Thanks, I will have a look... While you are at it, can you email your
> multipart enhancemennts and/or upload progress modifications?
>
> thanks,
> George.
>
>
>
> On 3/9/06, Rob Pitt <rob at motionpath.com> wrote:
>> Here you go. I think I included all the things you will need.
>>
>> You will need to remove the reference to monkeynuts (this is one of
>> my libraries for "hacks"). I have included the relevant file from the
>> library (It used a post-setup hook on Og) and the relevant models/
>> test case.
>>
>> You will need to do some editing to them (minor, probably just remove
>> the monkeynuts references) , and to the unit test (some of the setup
>> stuff will be missing), but it should give you an idea.
>>
>> You can safely ignore all those finder methods in localizer if you
>> don't know what they mean, if you don't want them remove them :)
>>
>> I cc'd it to the Nitro list too incase anyone wants to develop the
>> concept and make it a proper part of Nitro.
>>
>>
>>
>>
>> On 9 Mar 2006, at 13:37, Gergely Kontra wrote:
>>
>>> Rob Pitt azt ?rta, hogy:
>>>> If you think you will find this useful I will e-mail it to you.
>>>
>>> Thanks, please mail it to me! (In what stage is it?)
>>> --
>>> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/
>>> ~kgergely ]------+
>>> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356
>>> 9656 |
>>> | Budapest University of Technology and Economics        ICQ#
>>> 175564914 |
>>> +------------- "Olyan langesz vagyok, hogy poroltoval kellene
>>> jarnom!" -+
>>
>>
>>
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>>
>>
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general




From george.moschovitis at gmail.com  Thu Mar  9 09:18:44 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Mar 2006 16:18:44 +0200
Subject: [Nitro] multilingual model
In-Reply-To: <31884F4A-6FC3-4366-9B87-53A85D46CD79@motionpath.com>
References: <20060309121115.GA30712@mlabdial.hit.bme.hu>
	<op.s55cpenug3glhz@jo.local>
	<20060309130345.GD30712@mlabdial.hit.bme.hu>
	<0FF8DB88-DF82-4D4C-BF11-6467AC2C0DB8@motionpath.com>
	<20060309133729.GA912@mlabdial.hit.bme.hu>
	<10DDC978-359E-4A1F-B8F9-33D93147D42E@motionpath.com>
	<afd80b540603090612u78ad5b5cv11ec414e0aecfc5d@mail.gmail.com>
	<31884F4A-6FC3-4366-9B87-53A85D46CD79@motionpath.com>
Message-ID: <afd80b540603090618w6c781b1bj4a2f966b6f9fe5cf@mail.gmail.com>

On 3/9/06, Rob Pitt <rob at motionpath.com> wrote:
> Hehe OK G I will send you them at the weekend but seriously I did it
> all wrong, it could be done better by parsing character by character
> without regex and not in blocks like I did it.

well, perhaps you can send an improved patch ;-)

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From zimba.tm at gmail.com  Thu Mar  9 10:19:39 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 9 Mar 2006 16:19:39 +0100
Subject: [Nitro] Current action
Message-ID: <200603091619.39838.zimba.tm@gmail.com>

Hi list,

do you know how to get the name of the current called action by the 
dispatcher ?

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From rob at motionpath.com  Thu Mar  9 10:37:58 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 15:37:58 +0000
Subject: [Nitro] Current action
In-Reply-To: <200603091619.39838.zimba.tm@gmail.com>
References: <200603091619.39838.zimba.tm@gmail.com>
Message-ID: <3DC6F93C-FE54-4717-8ACE-CAFCD99CEFDD@motionpath.com>

Why not use the caller variable?

On 9 Mar 2006, at 15:19, zimba.tm wrote:

> Hi list,
>
> do you know how to get the name of the current called action by the
> dispatcher ?
>
> -- 
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From zimba.tm at gmail.com  Thu Mar  9 10:58:18 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 9 Mar 2006 16:58:18 +0100
Subject: [Nitro] Current action
In-Reply-To: <3DC6F93C-FE54-4717-8ACE-CAFCD99CEFDD@motionpath.com>
References: <200603091619.39838.zimba.tm@gmail.com>
	<3DC6F93C-FE54-4717-8ACE-CAFCD99CEFDD@motionpath.com>
Message-ID: <200603091658.18099.zimba.tm@gmail.com>

Because I'm lazy and I don't want to parse the array ?

No seriously, caller is not usable in all cases, like if you want to access 
the variable in :pre :post aspect methods

On Thursday 09 March 2006 16:37, Rob Pitt wrote:
> Why not use the caller variable?
>
> On 9 Mar 2006, at 15:19, zimba.tm wrote:
> > Hi list,
> >
> > do you know how to get the name of the current called action by the
> > dispatcher ?
> >
> > --
> > Cheers,
> >   zimba.tm
> >
> > weblog : http://zimba.oree.ch
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From zimba.tm at gmail.com  Thu Mar  9 11:24:03 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 9 Mar 2006 17:24:03 +0100
Subject: [Nitro] [PATCH] Compiler, Action and Element
Message-ID: <200603091724.04020.zimba.tm@gmail.com>

-- Tue Mar  7 20:12:09 CET 2006  Jonas Pfenniger <jonas at oree.ch>
  * All compilers now have text and compiler as argument

Thu Mar  9 17:20:43 CET 2006  Jonas Pfenniger <jonas at oree.ch>
  * Added Context#action that is filled by the dispatcher. It contains the 
action the dispatcher has chosen

Thu Mar  9 17:21:54 CET 2006  Jonas Pfenniger <jonas at oree.ch>
  * Nitro::Element : removed unneeded 

Cheers,
  zimba.tm

weblog : http://zimba.oree.ch
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.tgz
Type: application/x-tgz
Size: 11782 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/1a172c09/attachment.bin 

From rob at motionpath.com  Thu Mar  9 12:04:54 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 9 Mar 2006 17:04:54 +0000
Subject: [Nitro] Current action
In-Reply-To: <200603091658.18099.zimba.tm@gmail.com>
References: <200603091619.39838.zimba.tm@gmail.com>
	<3DC6F93C-FE54-4717-8ACE-CAFCD99CEFDD@motionpath.com>
	<200603091658.18099.zimba.tm@gmail.com>
Message-ID: <8215BE3A-A63C-4D70-9492-9214B264AF9E@motionpath.com>

@action_name ?

p instance_variables

On 9 Mar 2006, at 15:58, zimba.tm wrote:

> Because I'm lazy and I don't want to parse the array ?
>
> No seriously, caller is not usable in all cases, like if you want  
> to access
> the variable in :pre :post aspect methods
>
> On Thursday 09 March 2006 16:37, Rob Pitt wrote:
>> Why not use the caller variable?
>>
>> On 9 Mar 2006, at 15:19, zimba.tm wrote:
>>> Hi list,
>>>
>>> do you know how to get the name of the current called action by the
>>> dispatcher ?
>>>
>>> --
>>> Cheers,
>>>   zimba.tm
>>>
>>> weblog : http://zimba.oree.ch
>>> _______________________________________________
>>> Nitro-general mailing list
>>> Nitro-general at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>
> -- 
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From bryan.a.soto at gmail.com  Thu Mar  9 14:46:39 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 9 Mar 2006 11:46:39 -0800
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <2743C0BD-76A9-4C46-A788-FA4460761771@motionpath.com>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
	<op.s55b9qoeg3glhz@jo.local>
	<20060309125345.GC30712@mlabdial.hit.bme.hu>
	<9C1B0995-552C-49FF-9E61-132ADB05EA83@motionpath.com>
	<2743C0BD-76A9-4C46-A788-FA4460761771@motionpath.com>
Message-ID: <bd93db0d0603091146t6b760738l8b2e4bca61d8d40@mail.gmail.com>

On 3/9/06, Rob Pitt <rob at motionpath.com> wrote:
> If nobody does fix it I will fix it at the weekend.
>

I'll give it a look. I wouldn't want you to throw a banana at me... ;)



From bryan.a.soto at gmail.com  Thu Mar  9 15:45:08 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 9 Mar 2006 12:45:08 -0800
Subject: [Nitro] Current action
In-Reply-To: <200603091619.39838.zimba.tm@gmail.com>
References: <200603091619.39838.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603091245i65c13562j1a92884c727fdfca@mail.gmail.com>

On 3/9/06, zimba.tm <zimba.tm at gmail.com> wrote:
> Hi list,
>
> do you know how to get the name of the current called action by the
> dispatcher ?

As a plug, you could post this on http://oxyliquit.manveru.net
If you haven't seen it, give it a look. It's pretty cool.

Oh, and as Rob said, @action_name holds the name of the current
action. Available in the controller and view.

Bryan



From bryan.a.soto at gmail.com  Thu Mar  9 15:59:08 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 9 Mar 2006 12:59:08 -0800
Subject: [Nitro] PATCH: mongrel-adapter-cookies
Message-ID: <bd93db0d0603091259x5765ff2bxe57222d541bfb268@mail.gmail.com>

* mongrel-adapter-cookies
Now reuses Cgi.response_headers from cgi.rb which has the pleasant
bonus of making mongrel support cookies.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mongrel-adapter-cookies.zip
Type: application/zip
Size: 11630 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/60a4f5ce/attachment.zip 

From george.moschovitis at gmail.com  Thu Mar  9 16:31:28 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Mar 2006 22:31:28 +0100
Subject: [Nitro] Current action
In-Reply-To: <bd93db0d0603091245i65c13562j1a92884c727fdfca@mail.gmail.com>
References: <200603091619.39838.zimba.tm@gmail.com>
	<bd93db0d0603091245i65c13562j1a92884c727fdfca@mail.gmail.com>
Message-ID: <afd80b540603091331o21bc039etf9e71d89e4d8e239@mail.gmail.com>

yeah there is no need for context.action...

action_name does the same...

-g.

On 3/9/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/9/06, zimba.tm <zimba.tm at gmail.com> wrote:
> > Hi list,
> >
> > do you know how to get the name of the current called action by the
> > dispatcher ?
>
> As a plug, you could post this on http://oxyliquit.manveru.net
> If you haven't seen it, give it a look. It's pretty cool.
>
> Oh, and as Rob said, @action_name holds the name of the current
> action. Available in the controller and view.
>
> Bryan
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar  9 16:31:53 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Mar 2006 22:31:53 +0100
Subject: [Nitro] PATCH: mongrel-adapter-cookies
In-Reply-To: <bd93db0d0603091259x5765ff2bxe57222d541bfb268@mail.gmail.com>
References: <bd93db0d0603091259x5765ff2bxe57222d541bfb268@mail.gmail.com>
Message-ID: <afd80b540603091331l5ca5e807k3823425b5bbe7915@mail.gmail.com>

nice ;-)

On 3/9/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> * mongrel-adapter-cookies
> Now reuses Cgi.response_headers from cgi.rb which has the pleasant
> bonus of making mongrel support cookies.
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Thu Mar  9 16:33:22 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 9 Mar 2006 13:33:22 -0800
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <20060309121726.GB30712@mlabdial.hit.bme.hu>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603091333s15f558b9pd2d0c7dfafda5087@mail.gmail.com>

On 3/9/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Hi!
>
> I'm stucked at the point switching from sqlite tomysql.
>
> I get the error below. Can sy. please help to figure it out what's going
> on?

Hmm... the Ruby Mysql interface Mysql::Result#fetch_row. That
shouldn't be the issue.

Standard silly questions:
Are you providing the username, password and server address in Og.setup?
Is the Mysql-Ruby interface on your system up to date?

If yes to both, could you provide the code to duplicate? Privately is
fine, if you like.

Thanks,

Bryan



From bryan.a.soto at gmail.com  Thu Mar  9 16:34:32 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 9 Mar 2006 13:34:32 -0800
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <bd93db0d0603091333s15f558b9pd2d0c7dfafda5087@mail.gmail.com>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
	<bd93db0d0603091333s15f558b9pd2d0c7dfafda5087@mail.gmail.com>
Message-ID: <bd93db0d0603091334h17482712od2f55348521ee9d4@mail.gmail.com>

On 3/9/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Hmm... the Ruby Mysql interface Mysql::Result#fetch_row. That

That should be the Ruby Mysql interface _defines_ etc.



From george.moschovitis at gmail.com  Thu Mar  9 16:43:31 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 9 Mar 2006 22:43:31 +0100
Subject: [Nitro] Bounties
Message-ID: <afd80b540603091343u5f89f2c0xed7c91c43576c386@mail.gmail.com>

Dear devs,

here are some bounties for Nitro:

1. add send_file functionity that streams a file to the browser
2. Add attachment handling for incoming/outgoing emails in glue/mailer

perhaps someone with some free time can help here ;-)

best regards,
George.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From rob at motionpath.com  Thu Mar  9 16:59:04 2006
From: rob at motionpath.com (rob)
Date: Thu, 9 Mar 2006 21:59:04 +0000
Subject: [Nitro] switching from sqlite to mysql
In-Reply-To: <bd93db0d0603091146t6b760738l8b2e4bca61d8d40@mail.gmail.com>
References: <20060309121726.GB30712@mlabdial.hit.bme.hu>
	<op.s55b9qoeg3glhz@jo.local>
	<20060309125345.GC30712@mlabdial.hit.bme.hu>
	<9C1B0995-552C-49FF-9E61-132ADB05EA83@motionpath.com>
	<2743C0BD-76A9-4C46-A788-FA4460761771@motionpath.com>
	<bd93db0d0603091146t6b760738l8b2e4bca61d8d40@mail.gmail.com>
Message-ID: <5C8CD5F2-333F-4B9A-B44A-BF67B8936003@motionpath.com>

LOL you read too many mailing lists :D

On 9 Mar 2006, at 19:46, Bryan Soto wrote:

> On 3/9/06, Rob Pitt <rob at motionpath.com> wrote:
>> If nobody does fix it I will fix it at the weekend.
>>
>
> I'll give it a look. I wouldn't want you to throw a banana at me... ;)
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From guillaume.pierronnet at gmail.com  Thu Mar  9 17:00:05 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Thu, 9 Mar 2006 23:00:05 +0100
Subject: [Nitro] [PATCH] littles fixes
Message-ID: <6a7d49ca0603091400u63fbb495i@mail.gmail.com>

bundle content:

  * require fix in element.rb
  * some tiny spark fixes
  * fixes on Glue::Taggable
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.gz
Type: application/x-gzip
Size: 11750 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060309/f33f6296/attachment.gz 

From dylanb at digitalvalence.com  Thu Mar  9 19:19:43 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Thu, 09 Mar 2006 18:19:43 -0600
Subject: [Nitro] Kirbybase/inheritance/schema evolution
In-Reply-To: <bd93db0d0603081254r1d8cbf59qd021fcd614307467@mail.gmail.com>
References: <bd93db0d0603052339t25b8a374u7fb3b7348b70d132@mail.gmail.com>	
	<afd80b540603060243s6ca4dbfek33858e2ca9da0e5a@mail.gmail.com>	
	<440D2C1F.3080402@digitalvalence.com>
	<bd93db0d0603081254r1d8cbf59qd021fcd614307467@mail.gmail.com>
Message-ID: <4410C61F.10300@digitalvalence.com>

Thanks for the reply Bryan.  I think you're spot on about that thread 
relating to the same issue I had been having.  I have things working on 
my end now, but it is always fun seeing how someone else fixes the same 
problem.

Thanks for the Darcs tips, I will have to try them out.  I'm hoping that 
RDT didn't make a million changes to the file formatting when I wasn't 
looking.  That would muck things up a bit.

As for the needing to digest later part, I tend to run at the mouth... 
or fingers, or whatever the appropriate phrase is.  Especially when 
tired, which is one of the few times I get around to answering my 
email.  Hopefully there is something of value there. 

Thanks for the tips.  Hopefully I can cook up some patches if I am not 
just replicating .29 fixes.

Dylan

Bryan Soto wrote:
> Hi Dylan,
>
> On 3/6/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
>   
>> I recently started implementing a desktop application using OG to
>> connect to KirbyBase as the back end.  While implementing my initial
>> model tests I came across a couple of problems with the schema
>> inheritance and schema evolution features.
>>
>> 1.  Schema inheritance creates extra columns for each subclass
>> property.  Say we have a super class, Person, which has two children,
>> Manager and Programmer.  If Manager has an employees property and
>> Programmer has a cupsOfCoffee property schema evolution will drop the
>> column for whichever property came first.  This may be a general
>> inheritance and evolution problem, not just specific to Kirbybase.  I
>> haven't checked the other stores for more advanced methods of handling
>> this, such as using descendents.
>>     
>
> I think this might be the problem outlined at:
> http://rubyforge.org/pipermail/nitro-general/2006-January/002601.html
>
> It should be fixed in the current release 0.29.0.
>
> <snipped too much to digest />
>
> Wow, I'll have to save this for when I have some free time. ;)
>
> Kirby has a few issues because it doesn't natively support joins in
> it's queries. If you're looking at deploying, rather than server based
> stuff, you might be better off going with Sqlite. But if your fixes
> work, I do hope to see a patch. You sound experienced and we can use
> all the help we can get. :)
>
> To get rid of those false changes:
>
> darcs diff -u
> * Shows changes in unified diff format.
>
> darcs revert [optional filename]
> * Will iterate through changes [entire repo if no filename given].
> ** y to revert
> ** n to ignore
> ** q to quit
> ** d for done
> ** a for all (sames as answering yes to each change)
>
> darcs record [optional filename]
> * records changes [entire repo if no filename given]
> ** commands as in revert
>
> darcs send -o bundle_name (Need to be connected to Internet for this)
> * Outputs named changes [record] to file
> ** commands as in revert
>
> Optionally, you can just copy the files you want to save, darcs revert
> the entire repo, then move them back to their proper place. ;)
>
> Then email bundle to list or submit on Trac http://devlab.oree.ch/trac/nitrohq.
>
> Thanks,
>
> Bryan
>
>   



From bryan.a.soto at gmail.com  Thu Mar  9 21:21:55 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 9 Mar 2006 18:21:55 -0800
Subject: [Nitro] Dot's in URL issue
Message-ID: <bd93db0d0603091821h17d7f10dmbbf591a334b46a6@mail.gmail.com>

I know it has been brought up many times in the past that Nitro
ignores all URLs to actions containing a period.

I've looked through the Webrick and Mongrel adapters and they both
look for a physical file before delegating to Nitro. The Apache
.htaccess file checks whether the requested file exists before
delegating to Nitro, so that covers Apache with Cgi and FastCgi. I'm
not familiar with Lighttpd and Scgi though. I'm positive they'd
function the same way, so if someone can confirm that they look for a
physical file before delegating to Nitro, I'd be happy to send a patch
removing the checks for periods.

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From james_b at neurogami.com  Fri Mar 10 02:06:42 2006
From: james_b at neurogami.com (James Britt)
Date: Fri, 10 Mar 2006 00:06:42 -0700
Subject: [Nitro] Regexp-based URL routing
Message-ID: <44112582.7040408@neurogami.com>

I'm trying to get an app to invoke one or another controller method 
based on some pattern matching, if the URL does not have a default 
mapping to a method.

I recall there being some sort of route or map or ann method that might 
enable this, but cannot find any examples.


Basically, this is a method_missing sort of thing, but is their a Nitro 
way to do this?



-- 
James Britt

"Blanket statements are over-rated"


From aglarond at gmail.com  Fri Mar 10 02:58:31 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Fri, 10 Mar 2006 08:58:31 +0100
Subject: [Nitro] Regexp-based URL routing
In-Reply-To: <44112582.7040408@neurogami.com>
References: <44112582.7040408@neurogami.com>
Message-ID: <55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>

Hi James,

On 3/10/06, James Britt <james_b at neurogami.com> wrote:
> I'm trying to get an app to invoke one or another controller method
> based on some pattern matching, if the URL does not have a default
> mapping to a method.
>

I'll be doing the same thing shortly, so I've been watching out for
this sort of thing.

> I recall there being some sort of route or map or ann method that might
> enable this, but cannot find any examples.
>
>
> Basically, this is a method_missing sort of thing, but is their a Nitro
> way to do this?
>

>From George's 0.29.0 announcement:

* To better support the URL codec, a new router was implemented.
 The new version is both more powerfull and faster. Here are
 some examples:

 r.add_route(%r{rewritten/url/(.*)}, :controller => IdController,
:action => :register, :param => :name)
 r.add_route(%r{another/zelo/(.*)/(.*)}, :controller =>
AdminController, :action => :kick, :params => [:name, :age])
 r.add_route(%r{cool/(.*)_(.*).html}, :controller => AdminController,
:action => :long, :params => [:name, :age])


Does this help you any?

- Dimitri



From guillaume.pierronnet at gmail.com  Fri Mar 10 04:38:29 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Fri, 10 Mar 2006 10:38:29 +0100
Subject: [Nitro] Dot's in URL issue
In-Reply-To: <bd93db0d0603091821h17d7f10dmbbf591a334b46a6@mail.gmail.com>
References: <bd93db0d0603091821h17d7f10dmbbf591a334b46a6@mail.gmail.com>
Message-ID: <6a7d49ca0603100138k1622b3bfs@mail.gmail.com>

hi Bryan,

i got a lighttpd setup and it checked for the existence of the file
before delegating to handler (fastcgi and scgi are almost the same).
Some times ago, i did have a scgi setup on my apache and it looked  up
the dot in the url before delegating to scgi handler. But this can be
changed with some tweaking.

This patch looks safe to me.


2006/3/10, Bryan Soto <bryan.a.soto at gmail.com>:
> I know it has been brought up many times in the past that Nitro
> ignores all URLs to actions containing a period.
>
> I've looked through the Webrick and Mongrel adapters and they both
> look for a physical file before delegating to Nitro. The Apache
> .htaccess file checks whether the requested file exists before
> delegating to Nitro, so that covers Apache with Cgi and FastCgi. I'm
> not familiar with Lighttpd and Scgi though. I'm positive they'd
> function the same way, so if someone can confirm that they look for a
> physical file before delegating to Nitro, I'd be happy to send a patch
> removing the checks for periods.
>
> Thanks,
>
> Bryan
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From james_b at neurogami.com  Fri Mar 10 11:20:22 2006
From: james_b at neurogami.com (James Britt)
Date: Fri, 10 Mar 2006 09:20:22 -0700
Subject: [Nitro] Regexp-based URL routing
In-Reply-To: <55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>
References: <44112582.7040408@neurogami.com>
	<55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>
Message-ID: <4411A746.6020604@neurogami.com>

Dimitri Aivaliotis wrote:

> 
>>From George's 0.29.0 announcement:
> 
> * To better support the URL codec, a new router was implemented.
>  The new version is both more powerfull and faster. Here are
>  some examples:
> 
>  r.add_route(%r{rewritten/url/(.*)}, :controller => IdController,
> :action => :register, :param => :name)
>  r.add_route(%r{another/zelo/(.*)/(.*)}, :controller =>
> AdminController, :action => :kick, :params => [:name, :age])
>  r.add_route(%r{cool/(.*)_(.*).html}, :controller => AdminController,
> :action => :long, :params => [:name, :age])
> 
> 
> Does this help you any?


Only if I can figure out what 'r' is supposed to be in my application.

(My pet peeve: code-snippet examples that are missing the context needed 
to make sense to people who didn't write the code.)

Thanks,

-- 
James Britt

"You harmonize; then you customize."
  - Wilson Pickett


From james_b at neurogami.com  Fri Mar 10 12:52:25 2006
From: james_b at neurogami.com (James Britt)
Date: Fri, 10 Mar 2006 10:52:25 -0700
Subject: [Nitro] OK,  I give up (Re:  Regexp-based URL routing)
In-Reply-To: <4411A746.6020604@neurogami.com>
References: <44112582.7040408@neurogami.com>	<55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>
	<4411A746.6020604@neurogami.com>
Message-ID: <4411BCD9.4090103@neurogami.com>

James Britt wrote:
> Dimitri Aivaliotis wrote:
> 
> 
>>>From George's 0.29.0 announcement:
>>
>>* To better support the URL codec, a new router was implemented.
>> The new version is both more powerfull and faster. Here are
>> some examples:
>>
>> r.add_route(%r{rewritten/url/(.*)}, :controller => IdController,
>>:action => :register, :param => :name)
>> r.add_route(%r{another/zelo/(.*)/(.*)}, :controller =>
>>AdminController, :action => :kick, :params => [:name, :age])
>> r.add_route(%r{cool/(.*)_(.*).html}, :controller => AdminController,
>>:action => :long, :params => [:name, :age])
>>
>>
>>Does this help you any?
>

As it turns out, not one bit.

> 
> 
> Only if I can figure out what 'r' is supposed to be in my application

I give up.  I cannot figure out what 'r' is supposed to be or how I get 
a reference to the router so I can add routes.




James





From bryan.a.soto at gmail.com  Fri Mar 10 14:16:39 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 10 Mar 2006 11:16:39 -0800
Subject: [Nitro] OK, I give up (Re: Regexp-based URL routing)
In-Reply-To: <4411BCD9.4090103@neurogami.com>
References: <44112582.7040408@neurogami.com>
	<55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>
	<4411A746.6020604@neurogami.com> <4411BCD9.4090103@neurogami.com>
Message-ID: <bd93db0d0603101116y3c64fc1bnede8adf2d8c225ec@mail.gmail.com>

Hi James,

On 3/10/06, James Britt <james_b at neurogami.com> wrote:
> James Britt wrote:
> > Dimitri Aivaliotis wrote:
> >
> >
> >>>From George's 0.29.0 announcement:
> >>
> >>* To better support the URL codec, a new router was implemented.
> >> The new version is both more powerfull and faster. Here are
> >> some examples:
> >>
> >> r.add_route(%r{rewritten/url/(.*)}, :controller => IdController,
> >>:action => :register, :param => :name)
> >> r.add_route(%r{another/zelo/(.*)/(.*)}, :controller =>
> >>AdminController, :action => :kick, :params => [:name, :age])
> >> r.add_route(%r{cool/(.*)_(.*).html}, :controller => AdminController,
> >>:action => :long, :params => [:name, :age])
> >>
> >>
> >>Does this help you any?
> >
>
> As it turns out, not one bit.
>
> >
> >
> > Only if I can figure out what 'r' is supposed to be in my application
>
> I give up.  I cannot figure out what 'r' is supposed to be or how I get
> a reference to the router so I can add routes.

>From what I can figure out, you have access to the magic 'r' in your
controller in context.dispatcher:

<snippet from breakpoint in controller action>
context.dispatcher
=> #<Nitro::Dispatcher:0xb75c5a98 @routes=[],
@controllers={"/admin"=>AdminController, "/"=>MyArrayController}>
irb(#<MyArrayController:0xb75671c8>):003:0> context.dispatcher.routes
=> []
irb(#<MyArrayController:0xb75671c8>):004:0> context.dispatcher.add_route()
ArgumentError: wrong number of arguments (0 for 2)
</snippet>

Maybe that means you can dynamically add and remove routes? I don't
know if that's what you're looking for though.

Doing it statically, it seems that you can create your own dispatcher
in your run.rb file:

d= Nitro::Dispatcher.new

# add your controllers to it
d.mount(
   '/' => MainController, # mounts /
   '/users' => UsersController # mounts /users
 )

# add your routes...
d.add_route ...

# Then
Nitro::Server.run(:dispatcher => d) # Rather than Nitro.run or start or ...

I don't know if there's a prefered way to do this. Hopefully George or
someone else will chime in. This is just what I was able to dig up. I
hope it helps you.

Bryan



From george.moschovitis at gmail.com  Fri Mar 10 14:28:48 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 10 Mar 2006 20:28:48 +0100
Subject: [Nitro] OK, I give up (Re: Regexp-based URL routing)
In-Reply-To: <4411BCD9.4090103@neurogami.com>
References: <44112582.7040408@neurogami.com>
	<55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>
	<4411A746.6020604@neurogami.com> <4411BCD9.4090103@neurogami.com>
Message-ID: <afd80b540603101128j13d1ea81x23a9527005373d77@mail.gmail.com>

Hello James,

I apologize for the bad example, and for rushing this feature in the release.
Anyway, the dispatcher is the router, so you can get access to the
magic variable
with

server.dispatcher.

About setting the routing with annotations, saddly this seems broken
in the release gems:(

Attached you can find some quick replacements for dispatcher.rb and
router.rb that fix this, till I have access to my laptop to send a
proper patch. If you replace thos files, here is a better example how
this works:

def view_user
   "params:  #{request[:id]} and #{request[:mode]}"
end
ann :view_user, :route => [ /user_(\d*)_(*?)\.html/, :id, :mode ]

I will clean this up, ASAP. Sorry again for the inconvienience.

regards,
George.






On 3/10/06, James Britt <james_b at neurogami.com> wrote:
> James Britt wrote:
> > Dimitri Aivaliotis wrote:
> >
> >
> >>>From George's 0.29.0 announcement:
> >>
> >>* To better support the URL codec, a new router was implemented.
> >> The new version is both more powerfull and faster. Here are
> >> some examples:
> >>
> >> r.add_route(%r{rewritten/url/(.*)}, :controller => IdController,
> >>:action => :register, :param => :name)
> >> r.add_route(%r{another/zelo/(.*)/(.*)}, :controller =>
> >>AdminController, :action => :kick, :params => [:name, :age])
> >> r.add_route(%r{cool/(.*)_(.*).html}, :controller => AdminController,
> >>:action => :long, :params => [:name, :age])
> >>
> >>
> >>Does this help you any?
> >
>
> As it turns out, not one bit.
>
> >
> >
> > Only if I can figure out what 'r' is supposed to be in my application
>
> I give up.  I cannot figure out what 'r' is supposed to be or how I get
> a reference to the router so I can add routes.
>
>
>
>
> James
>
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: dispatcher.rb
Type: application/octet-stream
Size: 6327 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060310/722152ea/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: router.rb
Type: application/octet-stream
Size: 2343 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060310/722152ea/attachment-0001.obj 

From james_b at neurogami.com  Fri Mar 10 17:02:54 2006
From: james_b at neurogami.com (James Britt)
Date: Fri, 10 Mar 2006 15:02:54 -0700
Subject: [Nitro] OK, I give up (Re: Regexp-based URL routing)
In-Reply-To: <afd80b540603101128j13d1ea81x23a9527005373d77@mail.gmail.com>
References: <44112582.7040408@neurogami.com>	<55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>	<4411A746.6020604@neurogami.com>
	<4411BCD9.4090103@neurogami.com>
	<afd80b540603101128j13d1ea81x23a9527005373d77@mail.gmail.com>
Message-ID: <4411F78E.5070102@neurogami.com>

George Moschovitis wrote:
> Hello James,
> 
> I apologize for the bad example, and for rushing this feature in the release.
> Anyway, the dispatcher is the router, so you can get access to the
> magic variable
> with
> 
> server.dispatcher.
> 

I see.

> About setting the routing with annotations, saddly this seems broken
> in the release gems:(

Sadness.

> 
> Attached you can find some quick replacements for dispatcher.rb and
> router.rb that fix this, till I have access to my laptop to send a
> proper patch. If you replace thos files, here is a better example how
> this works:
> 
> def view_user
>    "params:  #{request[:id]} and #{request[:mode]}"
> end
> ann :view_user, :route => [ /user_(\d*)_(*?)\.html/, :id, :mode ]


Thanks for the fix.  I had tried something similar to this, but with an 
older syntax , which kept giving me "no such method 'first'" errors



I shall give the new files a whirl.

James


From kgergely at mcl.hu  Fri Mar 10 21:40:05 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Sat, 11 Mar 2006 03:40:05 +0100
Subject: [Nitro] graph handling (selfjoin)
Message-ID: <20060311024005.GB26563@mlabdial.hit.bme.hu>

Hi!

I'm stuck with a silly easy problem:

I've a graph, like this:

class Sector
  property :name, String, :unique => true
  many_to_many :parents, Sector
end

My problems:
- how can I collect those nodes, which don't have any parents?

- how can I modify the model, that I can see the other side, so, add 
  somehow a children attribute, which really has the children?

thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060310/0f657452/attachment.bin 

From kashia at vfemail.net  Sat Mar 11 07:05:49 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sat, 11 Mar 2006 13:05:49 +0100
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <20060311024005.GB26563@mlabdial.hit.bme.hu>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
Message-ID: <op.s58zvznbg3glhz@jo.local>

Hi,

read about that problem a bit, after reading what you said on IRC.

On Sat, 11 Mar 2006 03:40:05 +0100, Gergely Kontra <kgergely at mcl.hu> wrote:
> I've a graph, like this:

on IRC you said that you wanted to create a DAG, Directed Acyclic Graph.
http://en.wikipedia.org/wiki/Directed_acyclic_graph.  After reading this
I wondered what you meant with this Sector example, since in a DAG there
are no parents, there are only children (as it seems).

> class Sector
>   property :name, String, :unique => true
>   many_to_many :parents, Sector
> end

Here's how I would implement a DAG:

class Sector
   property :name, String, :unique => true
   many_to_many :children, Sector

   validate_unique, :name
end

Note the :children, and not :parents

> My problems:
> - how can I collect those nodes, which don't have any parents?

"A source is a vertex with no incoming edges" (wikipedia)

I'd probably use SQL for that problem.

Sector.find(:condition => "oid NOT IN (SELECT sector_oid2 FROM ogj_sector_sector)")

I did a quick test with it and it works, please note though, that this
uses a subselect, which may or may not work in mysql or other databases.

this query reads: "Find all sectors, where the oid of the current sector
is NOT on the right hand side of a relation."
The right side of a relations sais, that it is a child of something.

> - how can I modify the model, that I can see the other side, so, add
>   somehow a children attribute, which really has the children?

I may or may not have understood your question correctly, but this is how
you would access/create children then:

a = Sector.new("a")
b = Sector.new("b")
c = Sector.new("c")

b.add_child a

Sector.find(:condition => "oid NOT IN (SELECT sector_oid2 FROM ogj_sector_sector)")

ogtest=# select * from ogsector where oid NOT IN (SELECT sector_oid2 FROM ogj_sector_sector);
  name | oid
------+-----
  b    |   2
  c    |   3

So this returns the b and c Sectors, which are the root nodes.

b.children.each do |child|
   p child.name
end

should print "a"

This is untested code, might not work as expected.

Hope to have helped somehow,

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kgergely at mcl.hu  Sat Mar 11 08:55:13 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Sat, 11 Mar 2006 14:55:13 +0100
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <op.s58zvznbg3glhz@jo.local>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
	<op.s58zvznbg3glhz@jo.local>
Message-ID: <20060311135512.GA28699@mlabdial.hit.bme.hu>

Kashia Buch azt ?rta, hogy: 
> on IRC you said that you wanted to create a DAG, Directed Acyclic Graph.
> http://en.wikipedia.org/wiki/Directed_acyclic_graph.  After reading this
> I wondered what you meant with this Sector example, since in a DAG there
> are no parents, there are only children (as it seems).
Hmm. Well, you specify the children, you also have parents on the other 
side.
If D-->C means "C is the child of D", and I have a graph like>

D--
|  |
v  v
A  C
|  |
v  |
B<-

I can say A and C is the parent of B.

> class Sector
>    property :name, String, :unique => true
>    many_to_many :children, Sector
> 
>    validate_unique, :name
> end
> 
> Note the :children, and not :parents
> > My problems:
> > - how can I collect those nodes, which don't have any parents?
> "A source is a vertex with no incoming edges" (wikipedia)
> I'd probably use SQL for that problem.
> Sector.find(:condition => "oid NOT IN (SELECT sector_oid2 FROM ogj_sector_sector)")
Ok, thanks.
Anyway, I hoped there is some "cheaper" or "prettier" way.
I mean, Og's one goal is to hide the details of SQL, isn't it?

> > - how can I modify the model, that I can see the other side, so, add
> >   somehow a children attribute, which really has the children?
> I may or may not have understood your question correctly, but this is how
> you would access/create children then:

Ok, I'll try to explain a bit more.

My goal:
admin seems to be a good piece of code, I'd like to build a graph with 
that. For this to be comfortable, when editing a Sector object, I'd 
like to see both the parents and both the children of that sector.
The only way to achieve this seems me to write SQL query.
BUT:
If you join 2 distinct tables (eg. in the tc_store.rb testcase), you 
can easily access a_category.comments, and a_comment.category.

thanks for your efforts, and the quick replies.
The community seems very active, friendly and helpful :)

Just need some good docs, and I will ask fewer stupid questions (maybe)

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060311/a3789264/attachment.bin 

From kashia at vfemail.net  Sat Mar 11 10:09:27 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sat, 11 Mar 2006 16:09:27 +0100
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <20060311135512.GA28699@mlabdial.hit.bme.hu>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
	<op.s58zvznbg3glhz@jo.local>
	<20060311135512.GA28699@mlabdial.hit.bme.hu>
Message-ID: <op.s588d1x5g3glhz@jo.local>

On Sat, 11 Mar 2006 14:55:13 +0100, Gergely Kontra <kgergely at mcl.hu> wrote:
> I can say A and C is the parent of B.

Yup, you can say that, but mapping speech into something the computer
understands is something else *g*
As I see it, the arrows are "one way", so traversing is only possible
in one direction, so that's why the "parent" thing confuses me.
And if traversing is only possible in one direction, there should be
no easy way to go "backwards" up the graph again.

For a two-way graph, my mind would say:

class Sector
   many_to_many :parents, Sector # possible ', though => ChildToParent'
   many_to_many :children, Sector # possible ', though => ParentToChild'
end

Then it would be clear that one can traverse in both directions. But
I don't think that is really needed nor really good here.

> Anyway, I hoped there is some "cheaper" or "prettier" way.
> I mean, Og's one goal is to hide the details of SQL, isn't it?

Yup, Og can hide all details it knows about.

For example, there is a "is Hierarchical" for models. It creates a
hierarchical structure for the model, speaking you can access parents
and children of the given node.

You could look at this example, and create a "is DirectedAcyclicGraph" to
abstract everything you will write later.

> My goal:
> admin seems to be a good piece of code, I'd like to build a graph with
> that. For this to be comfortable, when editing a Sector object, I'd
> like to see both the parents and both the children of that sector.

Ah, do I understand correctly, you want to write a "Handler" for a DAG
class?
For creating/adding new nodes to a given node and such? Sounds neat.

> The only way to achieve this seems me to write SQL query.
> BUT:
> If you join 2 distinct tables (eg. in the tc_store.rb testcase), you
> can easily access a_category.comments, and a_comment.category.

I sat down and thought a while about this... The conclusion I came to:
I myself don't know about Og in that case, I can't think that way.

For example (tc_store.rb)
Comment.one(:condition => "body = 'Comment 4'", :include => :user)

How does that include User? Does it join those two?

And then, I'm not really sure what you want to accomplish here with
joins.

But, to not leave you without trying to help you further along ;)

given node from which you want to display children and parents:
node = Sector[oid]

parents:
parents = Sector.find(:condition => "oid IN (SELECT sector_oid FROM ogj_sector_sector WHERE sector_oid2 = #{node.oid})")

children:
node.children

These three things are the ones you needed for your project, right? :)

###

So, having said all that, I'm very sure there is another way to get the
parents, but I can only think in SQL, not in Og magic ;)

> thanks for your efforts, and the quick replies.
> The community seems very active, friendly and helpful :)

Indeed indeed, I love all of 'em :P

Btw, are you ok with adding your question to "Nitro Q&A" if this issue
is resolved later?
http://oxyliquit.manveru.net/ , you'd just have to copy paste your first
post (or whatever you mean that'd be useful for the next Newbie that would
want to know about something similar).


> Just need some good docs, and I will ask fewer stupid questions (maybe)

Yes yes, everyone is complaining about docs, just right now on IRC as well ;)
We're in the process of making everything better :P

And, your questions are by far _NOT_ stupid (or I am) ^^;

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kashia at vfemail.net  Sat Mar 11 14:45:52 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sat, 11 Mar 2006 20:45:52 +0100
Subject: [Nitro] Self-join :through =>
Message-ID: <op.s59k6qy0g3glhz@jo.local>

Hi,
problem:

class Foo
   many_to_many :relations, Foo, :through => FooRelation
   # many_to_many :relations, Foo
end

class FooRelation
   property :type

   #has_one Foo
   #has_one Foo
   # does not work, only creates one column

   #has_one :foo, Foo
   #has_one :foo2, Foo

   #has_one :foo, Foo
   #has_one :foo2, Foo, :field => 'manga_oid2'

end

Whichever I try, nothing works.

has_one :foo2, Foo
PGError: ERROR:  column "manga_oid2" referenced in foreign key constraint does not exist

has_one :foo2, Foo, :field => 'manga_oid2'
PGError: ERROR:  column "manga2_oid" referenced in foreign key constraint does not exist

note the column names ;/

How do I resolve this mess? Hope someone can help.

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From dylanb at digitalvalence.com  Sat Mar 11 14:57:55 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Sat, 11 Mar 2006 13:57:55 -0600
Subject: [Nitro] kirby patches and a test fix
Message-ID: <44132BC3.9080905@digitalvalence.com>

Ok, I've attached my kirbybase changes.  Let me know if I screwed up on 
the darcs thing, or the coding conventions, or whatever. 

Most of the tests now run.  Simple tests like tc_store that rely on sql 
string queries still fail; I haven't gotten around to writing an sql > 
object manipulation class yet.

Some of my comments were meant for me; it was pretty late when I cooked 
up this patch.  So please excuse any generalizations.

Changes:
*kirby.rb*
-The store now uses Structs to access the database.  The structs are 
then converted into objects of the appropriate class, taking the ogtype 
column into account.  This takes care of problems with inheritance as 
Kirby was returning the superclass in all cases.  Also, when a subclass 
row was returned with a different set of properties the kb_create 
function was failing.  No need to use kb_create with this method.
-The ogtype column is now added as it should be.  I do this by adding 
the property to the SchemaInheritanceBase mixin.  Is there some reason 
it is not done this way generally ?
-Create table no longer attempts to drop the required :recno column.
-Create table now uses the list of all properties for an inheritance 
hierarchy, and so doesn't drop columns when adding a subclass
-Simple aggregations like .max, .avg, .min, and .sum now work.  
Combining multiple aggregations in a statement doesn't - yet.  These use 
all records from the table.  Suboptimal, performance wise, so should be 
optimized.
-:group now works
-some excess unused code was cleaned up
-it is now possible to do single column requests using the :column 
option, this returns an array of values for that column
-the join and unjoin functions should now store and remove the table ids 
properly provided there is only 0 or 1 set of '::' in the class name.  
Some regex tweaking would fix this problem.  I'm not a master of 
regex... yet.
-:condition now has more applicable cases, and now supports parameter 
escaping ala prepared statements.    LIMIT, AND, and OR are also 
supported in strings.  :limit condition is supported.  Added support for 
 >= and <=.  :ogtype is added automatically to a list of existing 
conditions on schema_inheritance rather than the two being mutually 
exclusive.  :ogtype conditions specified at a higher level take precedence.
-.all works
-[id] now returns the proper classes on schema_inheritance.
-with_scope now works

remaining problems:
-options are currently overridden by with_scope rather than the other 
way around.  For many cases this will be fine, but a fix should be 
fairly simple. 
-sql strings aren't processed.  This is a big limitation for some cases.
-performance for medium to large datasets is likely to be ungood.

*sql.rb
*-some refactoring; moved the column calculation for inheritance and the 
prepared statement string escaping to its own method
*-*uses schema_inheritance? rather than ancestors_contains for that test.

Some of these changes may be overboard, but they pass most of the 
tests.  Hopefully they will prove to be a useful start for anyone 
considering using this store.

Comments and concerns are welcome.

Dylan Bruzenak
-------------- next part --------------
A non-text attachment was scrubbed...
Name: kirby_fixes.zip
Type: application/octet-stream
Size: 16553 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060311/d93a84d7/attachment.obj 

From kgergely at mcl.hu  Sat Mar 11 16:39:13 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Sat, 11 Mar 2006 22:39:13 +0100
Subject: [Nitro] custom validation
Message-ID: <20060311213913.GA30956@mlabdial.hit.bme.hu>

Hi!

Is there a piece of example of custom validation?
(I don't think it will fit to the standard validateion routines)
I'd like to check, if there is no cycle in my DAG when inserting 
a vertex.

thx
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060311/1f5a19e1/attachment.bin 

From vagabond at cataclysm-software.net  Sat Mar 11 18:39:11 2006
From: vagabond at cataclysm-software.net (Andrew Thompson)
Date: Sat, 11 Mar 2006 23:39:11 +0000
Subject: [Nitro] custom validation
In-Reply-To: <20060311213913.GA30956@mlabdial.hit.bme.hu>
References: <20060311213913.GA30956@mlabdial.hit.bme.hu>
Message-ID: <44135F9F.90907@cataclysm-software.net>

Gergely Kontra wrote:
> Hi!
> 
> Is there a piece of example of custom validation?
> (I don't think it will fit to the standard validateion routines)
> I'd like to check, if there is no cycle in my DAG when inserting 
> a vertex.

I'd also be interested in this, I tried to use the add function on the 
error object, but I couldn't get it to work. I ended up manually 
validating some stuff and combining it with the result of the builtin 
validation - not pretty...

Andrew



From james_b at neurogami.com  Sun Mar 12 00:28:08 2006
From: james_b at neurogami.com (James Britt)
Date: Sat, 11 Mar 2006 22:28:08 -0700
Subject: [Nitro] Inconsistent Kirbybase behavior
Message-ID: <4413B168.5000100@neurogami.com>

I've just updated the gems for  Og/Nitro, as well as for Kirbybase, on 
both a WinXP and a Linux box.

On the XP box I have no problem using the Kirby adapter (well, it didn't 
like Timestamped, so I took that out), but on the Linux box I'm getting 
  a "Cannot insert/update recno field" error when I try to save an object.

The code is this:
Og.setup( :store => :kirby,
           :evolve_schema => false,
            :base_dir  => 'db_tracker')

...

    ct = ClickTrack.new( :url =>  CGI.unescape( args_h[ 'u' ] ), :client 
=> ENV['REMOTE_ADDR']  )
    ct.save
  # error when calling save on linux, but not WinXP :(




I've Ruby 1.8.4 on WinXP, and 1.8.3 on the Linux box (I'm skeptical that 
this is the cause, but maybe).

I've fiddled with the 'evolve_schema' part, but no help.

Thoughts?


Thanks!

-- 
James Britt

?Design depends largely on constraints.?
  ? Charles Eames


From james_b at neurogami.com  Sun Mar 12 00:40:38 2006
From: james_b at neurogami.com (James Britt)
Date: Sat, 11 Mar 2006 22:40:38 -0700
Subject: [Nitro] Inconsistent Kirbybase behavior
In-Reply-To: <4413B168.5000100@neurogami.com>
References: <4413B168.5000100@neurogami.com>
Message-ID: <4413B456.3040004@neurogami.com>

James Britt wrote:
> I've just updated the gems for  Og/Nitro, as well as for Kirbybase, on 
> both a WinXP and a Linux box.
> 
> On the XP box I have no problem using the Kirby adapter (well, it didn't 
> like Timestamped, so I took that out), but on the Linux box I'm getting 
>   a "Cannot insert/update recno field" error when I try to save an object.
> 


I've just noticed that I get the same error on the Linux box even if I 
try using the sqlite adapter.

Hmm....


James


From nitrojesus.5.pistos at geoshell.com  Sun Mar 12 01:05:24 2006
From: nitrojesus.5.pistos at geoshell.com (nitrojesus.5.pistos at geoshell.com)
Date: Sun, 12 Mar 2006 01:05:24 -0500
Subject: [Nitro] Og bug -- conflict with open-uri
Message-ID: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>

$ irb
irb(main):001:0> require 'open-uri'
=> true
irb(main):002:0> open( "http://www.google.com" ) { |html| }
=> nil
irb(main):003:0> require 'og'
=> true
irb(main):004:0> open( "http://www.google.com" ) { |html| }
NoMethodError: undefined method `status=' for #<StringIO:0xb7945eac>
        from /usr/lib/ruby/1.8/open-uri.rb:280:in `open_http'
        from /usr/lib/ruby/1.8/open-uri.rb:629:in `buffer_open'
        from /usr/lib/ruby/1.8/open-uri.rb:167:in `open_loop'
        from /usr/lib/ruby/1.8/open-uri.rb:165:in `open_loop'
        from /usr/lib/ruby/1.8/open-uri.rb:135:in `open_uri'
        from /usr/lib/ruby/1.8/open-uri.rb:531:in `open'
        from /usr/lib/ruby/1.8/open-uri.rb:86:in `open'
        from (irb):4
irb(main):005:0> Og::Version
=> "0.24.0"

------

I'm on Gentoo.  I've actually emerged 0.25.0.

Pistos



From m.fellinger at gmail.com  Sun Mar 12 01:51:49 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Sun, 12 Mar 2006 15:51:49 +0900
Subject: [Nitro] Oxyliquit tip
In-Reply-To: <op.s50mebp8g3glhz@jo.local>
References: <afd80b540603060254l6bfb66a5jb45b69d213b7795b@mail.gmail.com>
	<op.s5zpd8neg3glhz@jo.local>
	<bd93db0d0603061334w65efe1a2he6a526afc4ca987b@mail.gmail.com>
	<op.s50mebp8g3glhz@jo.local>
Message-ID: <9c00d3e00603112251nc698f42n8c92cd604eee4054@mail.gmail.com>

Got Internet at a friend of mine, hopefully i can setup a subnetsomewhere with one of my neighbors, however, i upgraded oxyliquit -should be working... (also restarted because of the model-change)
On 3/7/06, Kashia Buch <kashia at vfemail.net> wrote:> Yeh,>> > On 3/6/06, Kashia Buch <kashia at vfemail.net> wrote:> >> BTW: Oxylquit had a feed already built in /feed , enjoy :)> > I'll see if I can find the source, since g mentioned it's open source.> > Patches welcome, after all :)>> Already patched that, but couldn't upload it due to Michael going home :P> Since he moved he doesn't have internet at home anymore, poor guy ^^;>> I don't know yet if it's open source... I don't have any ideas about> it and I don't know anything about copyright and licensing.>> For now:>> What you can do:>   * look at the code>   * work on the code>   * send me patches>   * use small parts of the code in your own project>   * use ideas of it in your own projects>   * if you redistribute it somehow, tell them where you got it from>> But please don't:>   * redistibute the work as your own>   * claim it is yours>   * use whole parts in your own projects>   * make a new "Nitro FAQ" with it>   * use anything for commercial projects>>> Basically everything I'd expect from nice people anyway.> But, over all:  ask me when you want something, I'm very friendly ^_^>> 'nuff said about unpleasent thingies>> I wanted to, somehow introduce the Nitro FAQ to the list somehow, but> since I didn't have any idea how to do that, and Michael just spoiled> it, it doesn't matter now :D>> There are already some usefully questions and answers in it now, have> fun using it.>> One word about those rating numbers at the end of each answer to your> questions: 0 is unrated, 4 and 5 declare the Question as answered and> will be shown in the "closed" list. If you still feel that there is a> lack of information, you can still answer. The user interface and the> registering/viewing/editing might still be very unstable, please post> anything you get (errors, weird behaviour) right to me, I'll see that> it gets fixed as soon as I can.>> Oh, and about editing:>> Use textile (I don't like it, but people seem to use it all the time)> and for code use:>> <code syntax="ruby">> class Foor>    attr :bar> end> </code>>> It is a pleasure to code in Nitro/Og, thank you, I thank everyone who> contibutes to this wonderful project and see the "Oxyliquit" as a way> to contribute to my favourite Web-Framework.>> I sure like to talk, don't I ;D>> I'll leave it at that, everyone, use the faq, and enjoy it :)>> Kash>> --> Feel the love> http://pinkjuice.com/pics/ruby.png> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From james_b at neurogami.com  Sun Mar 12 02:15:10 2006
From: james_b at neurogami.com (James Britt)
Date: Sun, 12 Mar 2006 00:15:10 -0700
Subject: [Nitro] Inconsistent Kirbybase behavior
In-Reply-To: <4413B456.3040004@neurogami.com>
References: <4413B168.5000100@neurogami.com> <4413B456.3040004@neurogami.com>
Message-ID: <4413CA7E.8050708@neurogami.com>

James Britt wrote:
...

> 
> 
> 
> I've just noticed that I get the same error on the Linux box even if I 
> try using the sqlite adapter.
> 


Clearing  out at directories and files created by the kirbybase adapter 
seems to have straightened this out, though I now get hard-to-reproduce 
errors (something about '2 args where 1 expected' or something like 
that) that may be related to the WEBrick service I'm using.

James


From bryan.a.soto at gmail.com  Sun Mar 12 03:12:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 12 Mar 2006 00:12:23 -0800
Subject: [Nitro] [PATCH] Compiler, Action and Element
In-Reply-To: <200603091724.04020.zimba.tm@gmail.com>
References: <200603091724.04020.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603120012k7253e707kb092046ac4693f94@mail.gmail.com>

On 3/9/06, zimba.tm <zimba.tm at gmail.com> wrote:
> -- Tue Mar  7 20:12:09 CET 2006  Jonas Pfenniger <jonas at oree.ch>
>   * All compilers now have text and compiler as argument
>

Okay, unifies the interfaces. I like that.

> Thu Mar  9 17:20:43 CET 2006  Jonas Pfenniger <jonas at oree.ch>
>   * Added Context#action that is filled by the dispatcher. It contains the
> action the dispatcher has chosen
>

Is that necessary? There is an @action_name variable available to the
controller and view.

> Thu Mar  9 17:21:54 CET 2006  Jonas Pfenniger <jonas at oree.ch>
>   * Nitro::Element : removed unneeded
>

Simple fix. Nothing to object to here. :)

> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From dylanb at digitalvalence.com  Sun Mar 12 03:39:46 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Sun, 12 Mar 2006 02:39:46 -0600
Subject: [Nitro] Inconsistent Kirbybase behavior
In-Reply-To: <4413CA7E.8050708@neurogami.com>
References: <4413B168.5000100@neurogami.com> <4413B456.3040004@neurogami.com>
	<4413CA7E.8050708@neurogami.com>
Message-ID: <4413DE52.5010906@digitalvalence.com>

Are you using schema inheritance ?  Can you send along a stack for the 
argument error ?

 I don't use the nitro side of things(just og so far), can you pass on 
some model objects or a test case so  I can run this against my new 
kirby code(patch submitted to this list earlier today) ?  I'm interested 
to see a more in depth use case.  Might be able to smoke out some of 
these glitches.

thanks,
Dylan Bruzenak



James Britt wrote:
> James Britt wrote:
> ...
>
>   
>>
>> I've just noticed that I get the same error on the Linux box even if I 
>> try using the sqlite adapter.
>>
>>     
>
>
> Clearing  out at directories and files created by the kirbybase adapter 
> seems to have straightened this out, though I now get hard-to-reproduce 
> errors (something about '2 args where 1 expected' or something like 
> that) that may be related to the WEBrick service I'm using.
>
> James
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>   



From fabian at oggu.de  Sun Mar 12 05:35:34 2006
From: fabian at oggu.de (Fabian Buch)
Date: Sun, 12 Mar 2006 11:35:34 +0100
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
Message-ID: <ed0dd90e400f4154f63623a248464941@oggu.de>


Am 12.03.2006 um 07:05 schrieb nitrojesus.5.pistos at geoshell.com:
>
> I'm on Gentoo.  I've actually emerged 0.25.0.

This is seriously outdated. You should deffinitely update to 0.29.0!

Fabian



From kgergely at mcl.hu  Sun Mar 12 07:31:25 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Sun, 12 Mar 2006 13:31:25 +0100
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <op.s588d1x5g3glhz@jo.local>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
	<op.s58zvznbg3glhz@jo.local>
	<20060311135512.GA28699@mlabdial.hit.bme.hu>
	<op.s588d1x5g3glhz@jo.local>
Message-ID: <20060312123125.GA1206@mlabdial.hit.bme.hu>

Kashia Buch azt ?rta, hogy: 
> For a two-way graph, my mind would say:
> 
> class Sector
>    many_to_many :parents, Sector # possible ', though => ChildToParent'
>    many_to_many :children, Sector # possible ', though => ParentToChild'
> end

Not working here :(

What is very annoying, that, if it joins 2 distinct tables, it is 
possible to go into both directions.

So, is it possible?

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060312/7008523a/attachment.bin 

From kashia at vfemail.net  Sun Mar 12 09:34:26 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sun, 12 Mar 2006 15:34:26 +0100
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <20060312123125.GA1206@mlabdial.hit.bme.hu>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
	<op.s58zvznbg3glhz@jo.local>
	<20060311135512.GA28699@mlabdial.hit.bme.hu>
	<op.s588d1x5g3glhz@jo.local>
	<20060312123125.GA1206@mlabdial.hit.bme.hu>
Message-ID: <op.s6a1foqjg3glhz@jo.local>

Hi,

>> class Sector
>>    many_to_many :parents, Sector # possible ', though => ChildToParent'
>>    many_to_many :children, Sector # possible ', though => ParentToChild'
>> end
>
> Not working here :(
>
> What is very annoying, that, if it joins 2 distinct tables, it is
> possible to go into both directions.
>
> So, is it possible?

Yes, as you can see at my other posting about self joins, this doesn't seem
to work...
I hope George or someone else can help here.

Please, I need this!

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kgergely at mcl.hu  Sun Mar 12 14:41:52 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Sun, 12 Mar 2006 20:41:52 +0100
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
Message-ID: <20060312194152.GA2787@mlabdial.hit.bme.hu>

Hi!

I'm trying to run nitro under Apache2 with fcgid.

I'm almost there, the server is running, but I get 500 response, and my 
log only says:

[Sun Mar 12 20:28:16 2006] [error] [client 192.168.1.1] Premature end of script headers: fcgi.rb

With webrick it worked. So where to hunt for errors?

thx
Gergo

-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060312/09713a94/attachment.bin 

From james_b at neurogami.com  Sun Mar 12 15:11:20 2006
From: james_b at neurogami.com (James Britt)
Date: Sun, 12 Mar 2006 13:11:20 -0700
Subject: [Nitro] Inconsistent Kirbybase behavior
In-Reply-To: <4413DE52.5010906@digitalvalence.com>
References: <4413B168.5000100@neurogami.com>
	<4413B456.3040004@neurogami.com>	<4413CA7E.8050708@neurogami.com>
	<4413DE52.5010906@digitalvalence.com>
Message-ID: <44148068.9070101@neurogami.com>

Dylan Bruzenak wrote:
> Are you using schema inheritance ?  Can you send along a stack for the 
> argument error ?

Um, no.  I had tried different usage of :evolve_schema, as I saw it 
referenced in some warning messages, but nothing I did with it helped, 
so I took it out.

Completely deleting the local directory and file created by Kirbybase 
seemed to fixed things.

The app is about as simple an Og app as one can make.  One class, about 
3 properties, all strings.

Create an instance, save it.

(On reflection, using Og is for the moment overkill; the data is so 
simple that directly using Kirbybase would make sense.  But I may evolve 
the app, and using Og to wrap the DB stuff is nice; I could swap out 
another DB if I need to, though the pure-Ruby approach is important 
right now.)



Thanks,


James


From fabian at oggu.de  Sun Mar 12 15:32:40 2006
From: fabian at oggu.de (Fabian Buch)
Date: Sun, 12 Mar 2006 21:32:40 +0100
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <20060312194152.GA2787@mlabdial.hit.bme.hu>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
Message-ID: <66c6ad210d20789d3894c9f724e84705@oggu.de>


Am 12.03.2006 um 20:41 schrieb Gergely Kontra:
> [Sun Mar 12 20:28:16 2006] [error] [client 192.168.1.1] Premature end 
> of script headers: fcgi.rb
>
> With webrick it worked. So where to hunt for errors?

Not very easy to debug. You can try running the dispatcher by hand and 
see what happens. It might or might not help. Fcgi is very difficult to 
debug.

Make absolutely sure there are no errors if you run run.rb, also no Og 
errors.

Once I also had the problem that I had fcgi.rb and cgi.rb in the public 
directory and apache didn't know whichone to choose. From that on I 
always rename them to dispatcher.cgi and dispatcher.fcgi which is 
clearer.

Fabian



From dylanb at digitalvalence.com  Sun Mar 12 15:43:17 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Sun, 12 Mar 2006 14:43:17 -0600
Subject: [Nitro] Inconsistent Kirbybase behavior
In-Reply-To: <44148068.9070101@neurogami.com>
References: <4413B168.5000100@neurogami.com>	<4413B456.3040004@neurogami.com>	<4413CA7E.8050708@neurogami.com>	<4413DE52.5010906@digitalvalence.com>
	<44148068.9070101@neurogami.com>
Message-ID: <441487E5.4010501@digitalvalence.com>

Strange error.  At least it is solved for the moment.  I didn't have 
that particular problem with any of my versions of Kirby; must have been 
something in the db file layout.  You wouldn't happen to have backups of 
the 'bad' files ?

We may never know...

Dylan

James Britt wrote:
> Dylan Bruzenak wrote:
>   
>> Are you using schema inheritance ?  Can you send along a stack for the 
>> argument error ?
>>     
>
> Um, no.  I had tried different usage of :evolve_schema, as I saw it 
> referenced in some warning messages, but nothing I did with it helped, 
> so I took it out.
>
> Completely deleting the local directory and file created by Kirbybase 
> seemed to fixed things.
>
> The app is about as simple an Og app as one can make.  One class, about 
> 3 properties, all strings.
>
> Create an instance, save it.
>
> (On reflection, using Og is for the moment overkill; the data is so 
> simple that directly using Kirbybase would make sense.  But I may evolve 
> the app, and using Og to wrap the DB stuff is nice; I could swap out 
> another DB if I need to, though the pure-Ruby approach is important 
> right now.)
>
>
>
> Thanks,
>
>
> James
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>   



From kgergely at mcl.hu  Sun Mar 12 15:51:45 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Sun, 12 Mar 2006 21:51:45 +0100
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <66c6ad210d20789d3894c9f724e84705@oggu.de>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
	<66c6ad210d20789d3894c9f724e84705@oggu.de>
Message-ID: <20060312205145.GB3152@mlabdial.hit.bme.hu>

Fabian Buch azt ?rta, hogy: 
> 
> Am 12.03.2006 um 20:41 schrieb Gergely Kontra:
> > [Sun Mar 12 20:28:16 2006] [error] [client 192.168.1.1] Premature end 
> > of script headers: fcgi.rb
> >
> > With webrick it worked. So where to hunt for errors?
> 
> Not very easy to debug. You can try running the dispatcher by hand and 
> see what happens. It might or might not help. Fcgi is very difficult to 
> debug.
> 
> Make absolutely sure there are no errors if you run run.rb, also no Og 
> errors.

Hmm, tried, no errors. Now what to do? Stucked!
(I know about the obsolate columns, but they are not an issue)

valaki at debian-server:~/sundiszno-nitro/public$ ruby fcgi.rb
D, [2006-03-12T21:40:34.171587 #5072] DEBUG -- : Using Memory sessions.
I, [2006-03-12T21:40:36.399530 #5072]  INFO -- : Og uses the Psql store.
D, [2006-03-12T21:40:39.886140 #5072] DEBUG -- : Table tevekenyseg already exists
D, [2006-03-12T21:40:40.008964 #5072] DEBUG -- : Join table 'j_sector_tevekenyseg' already exists.
D, [2006-03-12T21:40:40.191796 #5072] DEBUG -- : Table teaor already exists
D, [2006-03-12T21:40:40.235078 #5072] DEBUG -- : Join table 'j_sector_teaor' already exists.
D, [2006-03-12T21:40:40.359088 #5072] DEBUG -- : Table sector already exists
I, [2006-03-12T21:40:40.366866 #5072]  INFO -- : WARNING: You have an obsolete field 'level' on table 'sector' and :evolve_schema is not set or is in cautious mode!
I, [2006-03-12T21:40:40.367878 #5072]  INFO -- : WARNING: You have an obsolete field 'on_site' on table 'sector' and :evolve_schema is not set or is in cautious mode!
D, [2006-03-12T21:40:40.393778 #5072] DEBUG -- : Join table 'j_sector_sector' already exists.
D, [2006-03-12T21:40:40.398504 #5072] DEBUG -- : Join table 'j_sector_teaor' already exists.
D, [2006-03-12T21:40:40.403268 #5072] DEBUG -- : Join table 'j_sector_tevekenyseg' already exists.
D, [2006-03-12T21:40:40.694079 #5072] DEBUG -- : PostgreSQL processing foreign key constraints
D, [2006-03-12T21:40:40.695612 #5072] DEBUG -- : PostgreSQL finished setting constraints. No action was taken in 0.00 seconds.


-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060312/e622fb7b/attachment.bin 

From epiperak at gmail.com  Mon Mar 13 01:13:37 2006
From: epiperak at gmail.com (Emmanouil Piperakis)
Date: Mon, 13 Mar 2006 15:13:37 +0900
Subject: [Nitro] Self-join :through =>
In-Reply-To: <op.s59k6qy0g3glhz@jo.local>
References: <op.s59k6qy0g3glhz@jo.local>
Message-ID: <dea8525a0603122213q28acf05u8bc66f4553d6d0f3@mail.gmail.com>

Easy...

class Foo
  many_to_many :foorelations, Foorelation
end

class Foorelation
  many_to_many :foos, Foo
end

Is that what you want to do?
BTW, I am using postgres 8.1

Emmanouil

On 3/12/06, Kashia Buch <kashia at vfemail.net> wrote:
>
> Hi,
> problem:
>
> class Foo
>    many_to_many :relations, Foo, :through => FooRelation
>    # many_to_many :relations, Foo
> end
>
> class FooRelation
>    property :type
>
>    #has_one Foo
>    #has_one Foo
>    # does not work, only creates one column
>
>    #has_one :foo, Foo
>    #has_one :foo2, Foo
>
>    #has_one :foo, Foo
>    #has_one :foo2, Foo, :field => 'manga_oid2'
>
> end
>
> Whichever I try, nothing works.
>
> has_one :foo2, Foo
> PGError: ERROR:  column "manga_oid2" referenced in foreign key constraint
> does not exist
>
> has_one :foo2, Foo, :field => 'manga_oid2'
> PGError: ERROR:  column "manga2_oid" referenced in foreign key constraint
> does not exist
>
> note the column names ;/
>
> How do I resolve this mess? Hope someone can help.
>
> Kashia
>
> --
> Feel the love
> http://pinkjuice.com/pics/ruby.png
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/nitro-general/attachments/20060313/45e1f65a/attachment.html 

From rob at motionpath.com  Mon Mar 13 04:52:12 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 09:52:12 +0000
Subject: [Nitro] STI (Again =))
Message-ID: <30CF8422-9A3A-46FB-8739-CBBF8BBE05EC@motionpath.com>

There was a bug in my fix to the last STI bug causing it to only  
return the correct STI class in a multiple-find case sometimes (it  
never returned it before).

Now it always returns the right class...


-------------- next part --------------
A non-text attachment was scrubbed...
Name: sti-fix-more.patch.bz2
Type: application/octet-stream
Size: 7593 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060313/2adb8ba8/attachment.obj 

From rob at motionpath.com  Mon Mar 13 05:17:14 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 10:17:14 +0000
Subject: [Nitro] Add :table property to join relationships
Message-ID: <BC2D5047-5212-416C-B721-A89818C66950@motionpath.com>

I noticed that this patch I made never appeared in glycerin (and I'm  
using it so I hope it does!)

It adds a :table property to join relationships so you can have  
multiple joins with the same class by choosing unique table names.  
Has test case too.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: join.patch.bz2
Type: application/octet-stream
Size: 11092 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060313/ebd5e015/attachment.obj 

From zimba.tm at gmail.com  Mon Mar 13 05:26:48 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Mon, 13 Mar 2006 11:26:48 +0100
Subject: [Nitro] Add :table property to join relationships
In-Reply-To: <BC2D5047-5212-416C-B721-A89818C66950@motionpath.com>
References: <BC2D5047-5212-416C-B721-A89818C66950@motionpath.com>
Message-ID: <200603131126.48485.zimba.tm@gmail.com>

On Monday 13 March 2006 11:17, Rob Pitt wrote:
> I noticed that this patch I made never appeared in glycerin (and I'm
> using it so I hope it does!)
>
> It adds a :table property to join relationships so you can have
> multiple joins with the same class by choosing unique table names.
> Has test case too.

Don't forget to add the [PATCH] tag to your mail subject when submitting.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From zimba.tm at gmail.com  Mon Mar 13 05:28:07 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Mon, 13 Mar 2006 11:28:07 +0100
Subject: [Nitro] What's up with the repos ?
Message-ID: <200603131128.07048.zimba.tm@gmail.com>

Hi list,

I don't know what to do. Should I commit all submited patches to devlab's 
repo ? From some days, the nitrohq repo didn't change and devlab's too.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From zimba.tm at gmail.com  Mon Mar 13 05:29:33 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Mon, 13 Mar 2006 11:29:33 +0100
Subject: [Nitro] [PATCH] Compiler, Action and Element
In-Reply-To: <bd93db0d0603120012k7253e707kb092046ac4693f94@mail.gmail.com>
References: <200603091724.04020.zimba.tm@gmail.com>
	<bd93db0d0603120012k7253e707kb092046ac4693f94@mail.gmail.com>
Message-ID: <200603131129.33809.zimba.tm@gmail.com>

On Sunday 12 March 2006 09:12, Bryan Soto wrote:
> > Thu Mar  9 17:20:43 CET 2006  Jonas Pfenniger <jonas at oree.ch>
> >   * Added Context#action that is filled by the dispatcher. It contains
> > the action the dispatcher has chosen
>
> Is that necessary? There is an @action_name variable available to the
> controller and view.

It was a quick reaction because I didn't know @action_name exist. I will let 
this patch out of the repo until it is really needed.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From kashia at vfemail.net  Mon Mar 13 05:36:31 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Mon, 13 Mar 2006 11:36:31 +0100
Subject: [Nitro] Nitro FAQ moved: http://oxyliquit.de/
Message-ID: <op.s6ck25lig3glhz@jo.local>

Sorry for the inconvenience,

but the Nitro Q&A page now has it's own server.

http://oxyliquit.de/
http://www.oxyliquit.de/

Enjoy!

Most notable change:
  * the search works

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From rob at motionpath.com  Mon Mar 13 05:40:52 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 10:40:52 +0000
Subject: [Nitro] Add :table property to join relationships
In-Reply-To: <200603131126.48485.zimba.tm@gmail.com>
References: <BC2D5047-5212-416C-B721-A89818C66950@motionpath.com>
	<200603131126.48485.zimba.tm@gmail.com>
Message-ID: <DED1373C-1A55-433C-AE7C-719686931C3F@motionpath.com>

Oops :)

On 13 Mar 2006, at 10:26, zimba.tm wrote:

> On Monday 13 March 2006 11:17, Rob Pitt wrote:
>> I noticed that this patch I made never appeared in glycerin (and I'm
>> using it so I hope it does!)
>>
>> It adds a :table property to join relationships so you can have
>> multiple joins with the same class by choosing unique table names.
>> Has test case too.
>
> Don't forget to add the [PATCH] tag to your mail subject when  
> submitting.
>
> -- 
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From rob at motionpath.com  Mon Mar 13 05:44:04 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 10:44:04 +0000
Subject: [Nitro] What's up with the repos ?
In-Reply-To: <200603131128.07048.zimba.tm@gmail.com>
References: <200603131128.07048.zimba.tm@gmail.com>
Message-ID: <F23A66C2-6619-4890-9E6E-7153871432FD@motionpath.com>

I think George is on a break?

So I would say you should submit them to devlab and take charge of  
the repo again...

On 13 Mar 2006, at 10:28, zimba.tm wrote:

> Hi list,
>
> I don't know what to do. Should I commit all submited patches to  
> devlab's
> repo ? From some days, the nitrohq repo didn't change and devlab's  
> too.
>
> -- 
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From kashia at vfemail.net  Mon Mar 13 05:50:31 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Mon, 13 Mar 2006 11:50:31 +0100
Subject: [Nitro] Self-join :through =>
In-Reply-To: <dea8525a0603122213q28acf05u8bc66f4553d6d0f3@mail.gmail.com>
References: <op.s59k6qy0g3glhz@jo.local>
	<dea8525a0603122213q28acf05u8bc66f4553d6d0f3@mail.gmail.com>
Message-ID: <op.s6clqhlag3glhz@jo.local>

Hey,

thanks for trying to help, but:

> Easy...

not quite so much :)

> class Foo
>   many_to_many :foorelations, Foorelation
> end
>
> class Foorelation
>   many_to_many :foos, Foo
> end
>
> Is that what you want to do?

Sadly no.

I want to make a self join through a second class.

this is what I came finally up with: (it works)
I think Robs patches could've helped me enormously.

class Foo
   # ...
   many_to_many :relations, Manga, :through => MangaRelation, :control => :none
end

class MangaRelation
   property :type, String

   property :foo, Fixnum, :field => 'foo_oid', :foreign_key => 'oid', :relation => true, :control => :none
   property :foo2, Fixnum, :field => 'foo_oid2', :foreign_key => 'oid', :relation => true, :control => :none
end

There's the problem that I can't use controls for Foo in that case then...
minor detail for now, at least it works :)

for everyone who is curious how I use this:
foo.relations(:select => 'ogfoo.*, type')

And no, every single thing here is not nice, but atm I don't know any other way.

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From rob at motionpath.com  Mon Mar 13 06:04:29 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 11:04:29 +0000
Subject: [Nitro] Line 58 in compiler.rb (someone completely changed the
	compiler)
Message-ID: <F67B131E-7672-4D4D-9084-CD4EC43D9E79@motionpath.com>

     for template_root in @controller.instance_variable_get 
(:@template_root)

^-------- what if @template_root isn't set? This has made my app  
explode... where is the new template_root supposed to come from?


From rob at motionpath.com  Mon Mar 13 06:13:38 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 11:13:38 +0000
Subject: [Nitro] My last post about the compiler
Message-ID: <DA617B48-B2DA-49AF-A0B0-CEE9F09E87B8@motionpath.com>

Just ignore it... I had not fully understood the changes made by  
George to the new router.


From rob at motionpath.com  Mon Mar 13 06:46:13 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 11:46:13 +0000
Subject: [Nitro] OK, I give up (Re: Regexp-based URL routing)
In-Reply-To: <afd80b540603101128j13d1ea81x23a9527005373d77@mail.gmail.com>
References: <44112582.7040408@neurogami.com>
	<55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>
	<4411A746.6020604@neurogami.com> <4411BCD9.4090103@neurogami.com>
	<afd80b540603101128j13d1ea81x23a9527005373d77@mail.gmail.com>
Message-ID: <65620E9B-BD13-461D-ADF0-5757211E939E@motionpath.com>

I have just applied this in the hopes of getting rid of my horrid  
public_instance_method hacks (and my older still method_missing ones)...

This is the third time Nitro has changed the mechanism by which you  
can add dynamic routes. I like this one best :)

However, I should point out after applying the patches from glycerin  
and the patches you e-mailed to James Britt, The example you have  
does not work:

ann :view_user, :route => [ /user_(\d*)_(*?)\.html/, :id, :mode ]

I had to put the params in an array:

ann :view_user, :route => [ /user_(\d*)_(*?)\.html/, [:id, :mode]]

If this wasn't intended behavior it would be easy to clean up but i  
thought people should know...

On 10 Mar 2006, at 19:28, George Moschovitis wrote:

> Hello James,
>
> I apologize for the bad example, and for rushing this feature in  
> the release.
> Anyway, the dispatcher is the router, so you can get access to the
> magic variable
> with
>
> server.dispatcher.
>
> About setting the routing with annotations, saddly this seems broken
> in the release gems:(
>
> Attached you can find some quick replacements for dispatcher.rb and
> router.rb that fix this, till I have access to my laptop to send a
> proper patch. If you replace thos files, here is a better example how
> this works:
>
> def view_user
>    "params:  #{request[:id]} and #{request[:mode]}"
> end
> ann :view_user, :route => [ /user_(\d*)_(*?)\.html/, :id, :mode ]
>
> I will clean this up, ASAP. Sorry again for the inconvienience.
>
> regards,
> George.
>
>
>
>
>
>
> On 3/10/06, James Britt <james_b at neurogami.com> wrote:
>> James Britt wrote:
>>> Dimitri Aivaliotis wrote:
>>>
>>>
>>>>> From George's 0.29.0 announcement:
>>>>
>>>> * To better support the URL codec, a new router was implemented.
>>>> The new version is both more powerfull and faster. Here are
>>>> some examples:
>>>>
>>>> r.add_route(%r{rewritten/url/(.*)}, :controller => IdController,
>>>> :action => :register, :param => :name)
>>>> r.add_route(%r{another/zelo/(.*)/(.*)}, :controller =>
>>>> AdminController, :action => :kick, :params => [:name, :age])
>>>> r.add_route(%r{cool/(.*)_(.*).html}, :controller =>  
>>>> AdminController,
>>>> :action => :long, :params => [:name, :age])
>>>>
>>>>
>>>> Does this help you any?
>>>
>>
>> As it turns out, not one bit.
>>
>>>
>>>
>>> Only if I can figure out what 'r' is supposed to be in my  
>>> application
>>
>> I give up.  I cannot figure out what 'r' is supposed to be or how  
>> I get
>> a reference to the router so I can add routes.
>>
>>
>>
>>
>> James
>>
>>
>>
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
> <dispatcher.rb>
> <router.rb>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From rob at motionpath.com  Mon Mar 13 06:53:24 2006
From: rob at motionpath.com (Rob Pitt)
Date: Mon, 13 Mar 2006 11:53:24 +0000
Subject: [Nitro] [PATCH] George's routes fix as a patch
Message-ID: <8C85238A-A349-4C94-88B7-2F8741E9A74E@motionpath.com>

I converted George's e-mail to James Britt into a patch against  
repo.nitrohq.com glycerin current. I also added a tiny fix in case  
you add routes with the method Bryan Soto suggested but do not supply  
a params hash.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: routes.patch.bz2
Type: application/octet-stream
Size: 10212 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060313/47b6b72b/attachment.obj 

From zimba.tm at gmail.com  Mon Mar 13 08:13:56 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Mon, 13 Mar 2006 14:13:56 +0100
Subject: [Nitro] [PATCH] Moved the transformer pipeline at the controller
Message-ID: <200603131413.56508.zimba.tm@gmail.com>

Hi list,

Mon Mar 13 14:06:03 CET 2006  Jonas Pfenniger <jonas at oree.ch>
  * Moved the transformer pipeline at the Controller

You can now change the template transformer pipeline at the controller.

Nitro::Compiler.transformers #=> [StaticInclude, Elements, Morphing, Markup, 
ScriptCompiler, Cleanup, Template]

It means that you can now have different compiler pipelines for each 
Controller.

The old Nitro::Compiler.setup_transform_template method is also available to 
override this behavior.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch
-------------- next part --------------
A non-text attachment was scrubbed...
Name: compiler_transformers
Type: text/x-java
Size: 36835 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060313/4f9744f1/attachment.bin 

From nitrojesus.5.pistos at geoshell.com  Mon Mar 13 10:06:18 2006
From: nitrojesus.5.pistos at geoshell.com (nitrojesus.5.pistos at geoshell.com)
Date: Mon, 13 Mar 2006 10:06:18 -0500
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <ed0dd90e400f4154f63623a248464941@oggu.de>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
Message-ID: <6c9d9ef0603130706w1c10b33aj@mail.gmail.com>

On 12/03/06, Fabian Buch - fabian at oggu.de
<+nitrojesus+pistos+97fd83588c.fabian#oggu.de at spamgourmet.com> wrote:
> > I'm on Gentoo.  I've actually emerged 0.25.0.
>
> This is seriously outdated. You should deffinitely update to 0.29.0!

Hm, okay.  I'll have to craft my own e-build, but that shouldn't be a
problem.  Too bad I'm already well into my project, using
ActiveRecord... :-\

Pistos



From dylanb at digitalvalence.com  Mon Mar 13 10:49:49 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Mon, 13 Mar 2006 09:49:49 -0600
Subject: [Nitro] [PATCH] Re:  kirby patches and a test fix
In-Reply-To: <44132BC3.9080905@digitalvalence.com>
References: <44132BC3.9080905@digitalvalence.com>
Message-ID: <4415949D.3080803@digitalvalence.com>

forgot patch tag ;)
> Ok, I've attached my kirbybase changes.  Let me know if I screwed up 
> on the darcs thing, or the coding conventions, or whatever.
> Most of the tests now run.  Simple tests like tc_store that rely on 
> sql string queries still fail; I haven't gotten around to writing an 
> sql > object manipulation class yet.
>
> Some of my comments were meant for me; it was pretty late when I 
> cooked up this patch.  So please excuse any generalizations.
>
> Changes:
> *kirby.rb*
> -The store now uses Structs to access the database.  The structs are 
> then converted into objects of the appropriate class, taking the 
> ogtype column into account.  This takes care of problems with 
> inheritance as Kirby was returning the superclass in all cases.  Also, 
> when a subclass row was returned with a different set of properties 
> the kb_create function was failing.  No need to use kb_create with 
> this method.
> -The ogtype column is now added as it should be.  I do this by adding 
> the property to the SchemaInheritanceBase mixin.  Is there some reason 
> it is not done this way generally ?
> -Create table no longer attempts to drop the required :recno column.
> -Create table now uses the list of all properties for an inheritance 
> hierarchy, and so doesn't drop columns when adding a subclass
> -Simple aggregations like .max, .avg, .min, and .sum now work.  
> Combining multiple aggregations in a statement doesn't - yet.  These 
> use all records from the table.  Suboptimal, performance wise, so 
> should be optimized.
> -:group now works
> -some excess unused code was cleaned up
> -it is now possible to do single column requests using the :column 
> option, this returns an array of values for that column
> -the join and unjoin functions should now store and remove the table 
> ids properly provided there is only 0 or 1 set of '::' in the class 
> name.  Some regex tweaking would fix this problem.  I'm not a master 
> of regex... yet.
> -:condition now has more applicable cases, and now supports parameter 
> escaping ala prepared statements.    LIMIT, AND, and OR are also 
> supported in strings.  :limit condition is supported.  Added support 
> for >= and <=.  :ogtype is added automatically to a list of existing 
> conditions on schema_inheritance rather than the two being mutually 
> exclusive.  :ogtype conditions specified at a higher level take 
> precedence.
> -.all works
> -[id] now returns the proper classes on schema_inheritance.
> -with_scope now works
>
> remaining problems:
> -options are currently overridden by with_scope rather than the other 
> way around.  For many cases this will be fine, but a fix should be 
> fairly simple. -sql strings aren't processed.  This is a big 
> limitation for some cases.
> -performance for medium to large datasets is likely to be ungood.
>
> *sql.rb
> *-some refactoring; moved the column calculation for inheritance and 
> the prepared statement string escaping to its own method
> *-*uses schema_inheritance? rather than ancestors_contains for that test.
>
> Some of these changes may be overboard, but they pass most of the 
> tests.  Hopefully they will prove to be a useful start for anyone 
> considering using this store.
>
> Comments and concerns are welcome.
>
> Dylan Bruzenak
> ------------------------------------------------------------------------
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

-------------- next part --------------
A non-text attachment was scrubbed...
Name: kirby_fixes.zip
Type: application/octet-stream
Size: 16553 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060313/c37e7ddc/attachment.obj 

From george.moschovitis at gmail.com  Mon Mar 13 15:05:47 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Mar 2006 21:05:47 +0100
Subject: [Nitro] OK, I give up (Re: Regexp-based URL routing)
In-Reply-To: <65620E9B-BD13-461D-ADF0-5757211E939E@motionpath.com>
References: <44112582.7040408@neurogami.com>
	<55c107bf0603092358x47e27545y2929acab0443de75@mail.gmail.com>
	<4411A746.6020604@neurogami.com> <4411BCD9.4090103@neurogami.com>
	<afd80b540603101128j13d1ea81x23a9527005373d77@mail.gmail.com>
	<65620E9B-BD13-461D-ADF0-5757211E939E@motionpath.com>
Message-ID: <afd80b540603131205y56ada191wd06f851fc682c32@mail.gmail.com>

> If this wasn't intended behavior it would be easy to clean up but i
> thought people should know...

Hmm, if possible, please clean this up.

regards,
George.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar 13 15:08:10 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Mar 2006 21:08:10 +0100
Subject: [Nitro] custom validation
In-Reply-To: <44135F9F.90907@cataclysm-software.net>
References: <20060311213913.GA30956@mlabdial.hit.bme.hu>
	<44135F9F.90907@cataclysm-software.net>
Message-ID: <afd80b540603131208x1e44f682g9b41d8f62625b669@mail.gmail.com>

Hello,

if you have a look at glue/validation.rb you will find out how to do
this. At the moment, I dont have access to my files, (and in fact
limited access to the net) so I cant help you more.

-g.


On 3/12/06, Andrew Thompson <vagabond at cataclysm-software.net> wrote:
> Gergely Kontra wrote:
> > Hi!
> >
> > Is there a piece of example of custom validation?
> > (I don't think it will fit to the standard validateion routines)
> > I'd like to check, if there is no cycle in my DAG when inserting
> > a vertex.
>
> I'd also be interested in this, I tried to use the add function on the
> error object, but I couldn't get it to work. I ended up manually
> validating some stuff and combining it with the result of the builtin
> validation - not pretty...
>
> Andrew
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar 13 15:10:02 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Mar 2006 21:10:02 +0100
Subject: [Nitro] What's up with the repos ?
In-Reply-To: <F23A66C2-6619-4890-9E6E-7153871432FD@motionpath.com>
References: <200603131128.07048.zimba.tm@gmail.com>
	<F23A66C2-6619-4890-9E6E-7153871432FD@motionpath.com>
Message-ID: <afd80b540603131210l7f2f83e0w3aa4dd0aa1a30ec9@mail.gmail.com>

Yeah,

please submit your patches to your devlab repo (but please also point
to the list).
One of these days I will point repo.nitrohq.com to your repo anyway.

-g.

On 3/13/06, Rob Pitt <rob at motionpath.com> wrote:
> I think George is on a break?
>
> So I would say you should submit them to devlab and take charge of
> the repo again...
>
> On 13 Mar 2006, at 10:28, zimba.tm wrote:
>
> > Hi list,
> >
> > I don't know what to do. Should I commit all submited patches to
> > devlab's
> > repo ? From some days, the nitrohq repo didn't change and devlab's
> > too.
> >
> > --
> > Cheers,
> >   zimba.tm
> >
> > weblog : http://zimba.oree.ch
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar 13 15:20:22 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Mar 2006 21:20:22 +0100
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <op.s6a1foqjg3glhz@jo.local>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
	<op.s58zvznbg3glhz@jo.local>
	<20060311135512.GA28699@mlabdial.hit.bme.hu>
	<op.s588d1x5g3glhz@jo.local>
	<20060312123125.GA1206@mlabdial.hit.bme.hu>
	<op.s6a1foqjg3glhz@jo.local>
Message-ID: <afd80b540603131220j6e1b2b0cna96ba8b40dcc700a@mail.gmail.com>

> Yes, as you can see at my other posting about self joins, this doesn't seem
> to work...
> I hope George or someone else can help here.
> Please, I need this!

I will not be able to look into this befor the weekend :(

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar 13 15:44:33 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 13 Mar 2006 21:44:33 +0100
Subject: [Nitro] [PATCH] Re: kirby patches and a test fix
In-Reply-To: <4415949D.3080803@digitalvalence.com>
References: <44132BC3.9080905@digitalvalence.com>
	<4415949D.3080803@digitalvalence.com>
Message-ID: <afd80b540603131244i19a3a56fw5a6d9c0c1efda937@mail.gmail.com>

thanks!

On 3/13/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
> forgot patch tag ;)
> > Ok, I've attached my kirbybase changes.  Let me know if I screwed up
> > on the darcs thing, or the coding conventions, or whatever.
> > Most of the tests now run.  Simple tests like tc_store that rely on
> > sql string queries still fail; I haven't gotten around to writing an
> > sql > object manipulation class yet.
> >
> > Some of my comments were meant for me; it was pretty late when I
> > cooked up this patch.  So please excuse any generalizations.
> >
> > Changes:
> > *kirby.rb*
> > -The store now uses Structs to access the database.  The structs are
> > then converted into objects of the appropriate class, taking the
> > ogtype column into account.  This takes care of problems with
> > inheritance as Kirby was returning the superclass in all cases.  Also,
> > when a subclass row was returned with a different set of properties
> > the kb_create function was failing.  No need to use kb_create with
> > this method.
> > -The ogtype column is now added as it should be.  I do this by adding
> > the property to the SchemaInheritanceBase mixin.  Is there some reason
> > it is not done this way generally ?
> > -Create table no longer attempts to drop the required :recno column.
> > -Create table now uses the list of all properties for an inheritance
> > hierarchy, and so doesn't drop columns when adding a subclass
> > -Simple aggregations like .max, .avg, .min, and .sum now work.
> > Combining multiple aggregations in a statement doesn't - yet.  These
> > use all records from the table.  Suboptimal, performance wise, so
> > should be optimized.
> > -:group now works
> > -some excess unused code was cleaned up
> > -it is now possible to do single column requests using the :column
> > option, this returns an array of values for that column
> > -the join and unjoin functions should now store and remove the table
> > ids properly provided there is only 0 or 1 set of '::' in the class
> > name.  Some regex tweaking would fix this problem.  I'm not a master
> > of regex... yet.
> > -:condition now has more applicable cases, and now supports parameter
> > escaping ala prepared statements.    LIMIT, AND, and OR are also
> > supported in strings.  :limit condition is supported.  Added support
> > for >= and <=.  :ogtype is added automatically to a list of existing
> > conditions on schema_inheritance rather than the two being mutually
> > exclusive.  :ogtype conditions specified at a higher level take
> > precedence.
> > -.all works
> > -[id] now returns the proper classes on schema_inheritance.
> > -with_scope now works
> >
> > remaining problems:
> > -options are currently overridden by with_scope rather than the other
> > way around.  For many cases this will be fine, but a fix should be
> > fairly simple. -sql strings aren't processed.  This is a big
> > limitation for some cases.
> > -performance for medium to large datasets is likely to be ungood.
> >
> > *sql.rb
> > *-some refactoring; moved the column calculation for inheritance and
> > the prepared statement string escaping to its own method
> > *-*uses schema_inheritance? rather than ancestors_contains for that test.
> >
> > Some of these changes may be overboard, but they pass most of the
> > tests.  Hopefully they will prove to be a useful start for anyone
> > considering using this store.
> >
> > Comments and concerns are welcome.
> >
> > Dylan Bruzenak
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bakki.kudva at gmail.com  Mon Mar 13 15:48:55 2006
From: bakki.kudva at gmail.com (Bakki Kudva)
Date: Mon, 13 Mar 2006 15:48:55 -0500
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
Message-ID: <f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>

May be simpler to use rubygems to install/upgrade nitro, rails etc.?
-bakki

On 3/13/06, nitrojesus.5.pistos at geoshell.com
<nitrojesus.5.pistos at geoshell.com> wrote:
> On 12/03/06, Fabian Buch - fabian at oggu.de
> <+nitrojesus+pistos+97fd83588c.fabian#oggu.de at spamgourmet.com> wrote:
> > > I'm on Gentoo.  I've actually emerged 0.25.0.
> >
> > This is seriously outdated. You should deffinitely update to 0.29.0!
>
> Hm, okay.  I'll have to craft my own e-build, but that shouldn't be a
> problem.  Too bad I'm already well into my project, using
> ActiveRecord... :-\
>
> Pistos
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From bryan.a.soto at gmail.com  Mon Mar 13 17:01:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 13 Mar 2006 14:01:23 -0800
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
Message-ID: <bd93db0d0603131401o5365f4l7d70d059fb4ede4b@mail.gmail.com>

On 3/11/06, nitrojesus.5.pistos at geoshell.com
<nitrojesus.5.pistos at geoshell.com> wrote:
> irb(main):005:0> Og::Version
> => "0.24.0"
>
> ------
>
> I'm on Gentoo.  I've actually emerged 0.25.0.
>

Fortunately it appears to have been fixed since then. Your example
works fine in current 0.29.0.

$ irb
irb(main):001:0> require 'open-uri'
=> true
irb(main):002:0> open('http://www.google.com') {|html|}
=> nil
irb(main):003:0> require 'og'
LoadError: no such file to load -- og
        from (irb):3:in `require'
        from (irb):3
        from /usr/lib/ruby/1.8/uri/generic.rb:199
irb(main):004:0> require 'rubygems'
=> true
irb(main):005:0> require 'og'
=> true
irb(main):006:0> open('http://www.google.com') {|html|}
=> nil
irb(main):007:0> puts Og::Version
0.29.0
=> nil

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 13 17:07:07 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 13 Mar 2006 14:07:07 -0800
Subject: [Nitro] kirby patches and a test fix
In-Reply-To: <44132BC3.9080905@digitalvalence.com>
References: <44132BC3.9080905@digitalvalence.com>
Message-ID: <bd93db0d0603131407j20fdd1dbjdf05b412201fa0e1@mail.gmail.com>

On 3/11/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:

Lots of good things here. I'm glad to see someone giving Kirby a
little love. Thanks for helping out. :)

> remaining problems:
> -options are currently overridden by with_scope rather than the other
> way around.  For many cases this will be fine, but a fix should be
> fairly simple.
> -sql strings aren't processed.  This is a big limitation for some cases.
> -performance for medium to large datasets is likely to be ungood.
>
> *sql.rb
> *-some refactoring; moved the column calculation for inheritance and the
> prepared statement string escaping to its own method
> *-*uses schema_inheritance? rather than ancestors_contains for that test.
>
> Some of these changes may be overboard, but they pass most of the
> tests.  Hopefully they will prove to be a useful start for anyone
> considering using this store.
>
> Comments and concerns are welcome.

I'll lend my eyeballs to this as soon as I can.

# self.fell_behind?    # => true :(

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 13 18:30:42 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 13 Mar 2006 15:30:42 -0800
Subject: [Nitro] [PATCH] Moved the transformer pipeline at the controller
In-Reply-To: <200603131413.56508.zimba.tm@gmail.com>
References: <200603131413.56508.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603131530u60971e09q6d680e481b43a4ff@mail.gmail.com>

On 3/13/06, zimba.tm <zimba.tm at gmail.com> wrote:
> Hi list,
>
> Mon Mar 13 14:06:03 CET 2006  Jonas Pfenniger <jonas at oree.ch>
>   * Moved the transformer pipeline at the Controller
>
> You can now change the template transformer pipeline at the controller.
>
> Nitro::Compiler.transformers #=> [StaticInclude, Elements, Morphing, Markup,
> ScriptCompiler, Cleanup, Template]
>
> It means that you can now have different compiler pipelines for each
> Controller.
>
> The old Nitro::Compiler.setup_transform_template method is also available to
> override this behavior.

Interesting idea. I notice you've been making a few changes to the
compilers recently. General cleanup or did you have something in mind?
:)

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 13 18:51:08 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 13 Mar 2006 15:51:08 -0800
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <20060312194152.GA2787@mlabdial.hit.bme.hu>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603131551g1264a627nfc681e11872d70b0@mail.gmail.com>

On 3/12/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Hi!
>
> I'm trying to run nitro under Apache2 with fcgid.
>
> I'm almost there, the server is running, but I get 500 response, and my
> log only says:
>
> [Sun Mar 12 20:28:16 2006] [error] [client 192.168.1.1] Premature end of script headers: fcgi.rb
>
> With webrick it worked. So where to hunt for errors?
>

First, check to ensure the shebang line on fcgi.rb is correct. When
switching between my Windows and Linux box, that always caught me. :)

If the path is okay, are you sure you configured httpd.conf correctly?
You have to specify an intial environment. Of particular importance is
setting up RUBYOPT if you installed nitro via gems or make use of any
other gems. Also, Nitro makes use of mysqladmin, so you should specify
the path to it.

FastCgiConfig -initial-env RUBYOPT="rubygems" /
                        -initial-env PATH="/usr/bin"

You're getting fcgi.rb, so I assume you've already enabled mod_rewrite
and have an .htaccess file...

You might try to su to apache and running fcgi.rb from the command
line. Though you might need to be root to do that.

# su apache
# ./fcgi.rb

Or perhaps Apache2 provides more info that we're not making use of in
the default 500 error page?

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From nitrojesus.5.pistos at geoshell.com  Mon Mar 13 20:16:17 2006
From: nitrojesus.5.pistos at geoshell.com (nitrojesus.5.pistos at geoshell.com)
Date: Mon, 13 Mar 2006 20:16:17 -0500
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
Message-ID: <6c9d9ef0603131716s627e15d8l@mail.gmail.com>

I prefer to respect the package manager of whatever distro I'm on as
much as possible.  I'll just wait until ebuilds for the later versions
come out.  Unfortunately, as it stands, nitro+og is too
underdocumented for me to be comfortable using.

I will definitely keep my eye out, though, and check it out again in a
few months.

Pistos

On 13/03/06, Bakki Kudva - bakki.kudva at gmail.com
<+nitrojesus+pistos+3978b84508.bakki.kudva#gmail.com at spamgourmet.com>
wrote:
> May be simpler to use rubygems to install/upgrade nitro, rails etc.?
> -bakki
>
> On 3/13/06, nitrojesus.5.pistos at geoshell.com
> <nitrojesus.5.pistos at geoshell.com> wrote:
> > On 12/03/06, Fabian Buch - fabian at oggu.de
> > <+nitrojesus+pistos+97fd83588c.fabian#oggu.de at spamgourmet.com> wrote:
> > > > I'm on Gentoo.  I've actually emerged 0.25.0.
> > >
> > > This is seriously outdated. You should deffinitely update to 0.29.0!
> >
> > Hm, okay.  I'll have to craft my own e-build, but that shouldn't be a
> > problem.  Too bad I'm already well into my project, using
> > ActiveRecord... :-\
> >
> > Pistos
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From dylanb at digitalvalence.com  Mon Mar 13 20:23:07 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Mon, 13 Mar 2006 19:23:07 -0600
Subject: [Nitro] kirby patches and a test fix
In-Reply-To: <bd93db0d0603131407j20fdd1dbjdf05b412201fa0e1@mail.gmail.com>
References: <44132BC3.9080905@digitalvalence.com>
	<bd93db0d0603131407j20fdd1dbjdf05b412201fa0e1@mail.gmail.com>
Message-ID: <44161AFB.8030801@digitalvalence.com>

You're quite welcome; I only wish I had more time to contribute to 
things like this.  The story of us all ;)

Dylan

Bryan Soto wrote:
> On 3/11/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
>
> Lots of good things here. I'm glad to see someone giving Kirby a
> little love. Thanks for helping out. :)
>
>   
>> remaining problems:
>> -options are currently overridden by with_scope rather than the other
>> way around.  For many cases this will be fine, but a fix should be
>> fairly simple.
>> -sql strings aren't processed.  This is a big limitation for some cases.
>> -performance for medium to large datasets is likely to be ungood.
>>
>> *sql.rb
>> *-some refactoring; moved the column calculation for inheritance and the
>> prepared statement string escaping to its own method
>> *-*uses schema_inheritance? rather than ancestors_contains for that test.
>>
>> Some of these changes may be overboard, but they pass most of the
>> tests.  Hopefully they will prove to be a useful start for anyone
>> considering using this store.
>>
>> Comments and concerns are welcome.
>>     
>
> I'll lend my eyeballs to this as soon as I can.
>
> # self.fell_behind?    # => true :(
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>   



From kashia at vfemail.net  Tue Mar 14 01:20:17 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Tue, 14 Mar 2006 07:20:17 +0100
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <afd80b540603131220j6e1b2b0cna96ba8b40dcc700a@mail.gmail.com>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
	<op.s58zvznbg3glhz@jo.local>
	<20060311135512.GA28699@mlabdial.hit.bme.hu>
	<op.s588d1x5g3glhz@jo.local>
	<20060312123125.GA1206@mlabdial.hit.bme.hu>
	<op.s6a1foqjg3glhz@jo.local>
	<afd80b540603131220j6e1b2b0cna96ba8b40dcc700a@mail.gmail.com>
Message-ID: <op.s6d3v3g1g3glhz@jo.local>

On Mon, 13 Mar 2006 21:20:22 +0100, George Moschovitis <george.moschovitis at gmail.com> wrote:

>> Yes, as you can see at my other posting about self joins, this doesn't seem
>> to work...
>> I hope George or someone else can help here.
>> Please, I need this!
>
> I will not be able to look into this befor the weekend :(

Don't worry, I found an evil workaround, which does it for now.

Kash

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From manveru at weez.co.jp  Tue Mar 14 02:25:40 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Tue, 14 Mar 2006 16:25:40 +0900
Subject: [Nitro] [PATCH] two small flare-patches
Message-ID: <200603141625.40585.manveru@weez.co.jp>

The first patch corrects a problem with the RSS-feeds where the comments-feed 
was not valid because the object it refered to didn't provide the to_href 
method, added the method and pointing the link to the comment to 
"/article/#@article_oid", which works quit well for now, still it would be 
nice to have a link directly to the comments, not sure how we would do that? 
(in-page-anchors?)

The other patch removes a big annoyance that i have to fix everytime 
installing flare, removing the line:
? validate_length :body, :range => 2..1000
since the author should know him/herself how much belongs into one post and i 
have myself crossed that limit far too often (losing all my text if it was 
only one character too much).

I hope both changes are acceptable for everybody, if you want to restrict 
yourself just copy&modify the :title validation :) - i don't think anybody 
wants to do that.

~~~~manveru
-------------- next part --------------
A non-text attachment was scrubbed...
Name: flare_char_restriction.tar.bz2
Type: application/x-tbz
Size: 10254 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060314/e1c0b527/attachment.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: flare_rss.tar.bz2
Type: application/x-tbz
Size: 10320 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060314/e1c0b527/attachment-0001.bin 

From zimba.tm at gmail.com  Tue Mar 14 02:29:21 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Tue, 14 Mar 2006 08:29:21 +0100
Subject: [Nitro] [PATCH] Moved the transformer pipeline at the controller
In-Reply-To: <bd93db0d0603131530u60971e09q6d680e481b43a4ff@mail.gmail.com>
References: <200603131413.56508.zimba.tm@gmail.com>
	<bd93db0d0603131530u60971e09q6d680e481b43a4ff@mail.gmail.com>
Message-ID: <200603140829.21093.zimba.tm@gmail.com>

On Tuesday 14 March 2006 00:30, Bryan Soto wrote:
> Interesting idea. I notice you've been making a few changes to the
> compilers recently. General cleanup or did you have something in mind?

Well, first I needed to unify all compilers.transform methods, what I did in a 
previous patch. Then I had the idea of putting the transformers pipeline in 
the controller to allow better componentization. George must be working on a 
similar thing with the "part" feature.

The idea is to put any nitro app in a /part sub-folder and mount it using one 
command.

Btw the code really needs some cleanup but it's hard to tell what is legacy or 
not so I don't like to touch.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From rob at motionpath.com  Tue Mar 14 04:36:57 2006
From: rob at motionpath.com (Rob Pitt)
Date: Tue, 14 Mar 2006 09:36:57 +0000
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <6c9d9ef0603131716s627e15d8l@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
Message-ID: <12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>

. I too prefer to respect the package manager of a system but if the  
package is out of date then it makes perfect sense to build a better  
version yourself.

There is no harm at all in rolling your own ebuild with a newer  
version and I have to admit I'm not especially happy the Gentoo user  
decided to put Nitro in the portage tree if they intended to abandon  
it and not maintain it properly.

I would recommend using rubygems over portage in this instance on the  
following rational (this can be applied to all distro/package  
managers with a similar situation):

The reason to use a distro's package manager is to:

1) Have a single tool for package management
2) Get free security updates when your package manager is running in  
it's normal update cycles
3) Other distro packages that rely on the package know it's installed

So considering these points, each of them can be rebutted with the  
following:

1) The single tool is of negligible value when rubygems is of equal  
if not greater simplicity to use than portage
2) The ebuild is clearly not maintained properly and doing "gem  
update" is a miniscule extra maintenance overhead (plus you read this  
list and we will talk about updates)
3) There are no known in the wild projects for a community user that  
use Nitro/Og at present that are not distributed with it. This  
benefit is unnecessary and due to the simplicity of ruby package  
management, the power of a system like portage will never be necessary.

Of course, it's your OS but I've managed a lot of OS/servers in my  
time (including Gentoo/portage) and you should consider these points,  
package managers are just tools and using the wrong tool for the job  
will only cause you pain in the long run...

- rp

On 14 Mar 2006, at 01:16, nitrojesus.5.pistos at geoshell.com wrote:

> I prefer to respect the package manager of whatever distro I'm on as
> much as possible.  I'll just wait until ebuilds for the later versions
> come out.  Unfortunately, as it stands, nitro+og is too
> underdocumented for me to be comfortable using.
>
> I will definitely keep my eye out, though, and check it out again in a
> few months.
>
> Pistos
>
> On 13/03/06, Bakki Kudva - bakki.kudva at gmail.com
> <+nitrojesus+pistos+3978b84508.bakki.kudva#gmail.com at spamgourmet.com>
> wrote:
>> May be simpler to use rubygems to install/upgrade nitro, rails etc.?
>> -bakki
>>
>> On 3/13/06, nitrojesus.5.pistos at geoshell.com
>> <nitrojesus.5.pistos at geoshell.com> wrote:
>>> On 12/03/06, Fabian Buch - fabian at oggu.de
>>> <+nitrojesus+pistos+97fd83588c.fabian#oggu.de at spamgourmet.com>  
>>> wrote:
>>>>> I'm on Gentoo.  I've actually emerged 0.25.0.
>>>>
>>>> This is seriously outdated. You should deffinitely update to  
>>>> 0.29.0!
>>>
>>> Hm, okay.  I'll have to craft my own e-build, but that shouldn't  
>>> be a
>>> problem.  Too bad I'm already well into my project, using
>>> ActiveRecord... :-\
>>>
>>> Pistos
>>>
>>> _______________________________________________
>>> Nitro-general mailing list
>>> Nitro-general at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/nitro-general
>>>
>>
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From zimba.tm at gmail.com  Tue Mar 14 05:30:00 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Tue, 14 Mar 2006 11:30:00 +0100
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
	<12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
Message-ID: <200603141130.01453.zimba.tm@gmail.com>

Rob Pitt's arguments are all valid I think.

I would like to add that ebuilds are mostly useful to avoid system corruptions 
(dependencies, file list, ...).

Considering that it's a ruby library, you won't have any problems because it 
only uses the /usr/lib/ruby/gems/* namespace.

On Tuesday 14 March 2006 10:36, Rob Pitt wrote:
> . I too prefer to respect the package manager of a system but if the
> package is out of date then it makes perfect sense to build a better
> version yourself.
>
> There is no harm at all in rolling your own ebuild with a newer
> version and I have to admit I'm not especially happy the Gentoo user
> decided to put Nitro in the portage tree if they intended to abandon
> it and not maintain it properly.
>
> I would recommend using rubygems over portage in this instance on the
> following rational (this can be applied to all distro/package
> managers with a similar situation):
>
> The reason to use a distro's package manager is to:
>
> 1) Have a single tool for package management
> 2) Get free security updates when your package manager is running in
> it's normal update cycles
> 3) Other distro packages that rely on the package know it's installed
>
> So considering these points, each of them can be rebutted with the
> following:
>
> 1) The single tool is of negligible value when rubygems is of equal
> if not greater simplicity to use than portage
> 2) The ebuild is clearly not maintained properly and doing "gem
> update" is a miniscule extra maintenance overhead (plus you read this
> list and we will talk about updates)
> 3) There are no known in the wild projects for a community user that
> use Nitro/Og at present that are not distributed with it. This
> benefit is unnecessary and due to the simplicity of ruby package
> management, the power of a system like portage will never be necessary.
>
> Of course, it's your OS but I've managed a lot of OS/servers in my
> time (including Gentoo/portage) and you should consider these points,
> package managers are just tools and using the wrong tool for the job
> will only cause you pain in the long run...
>
> - rp
>
> On 14 Mar 2006, at 01:16, nitrojesus.5.pistos at geoshell.com wrote:
> > I prefer to respect the package manager of whatever distro I'm on as
> > much as possible.  I'll just wait until ebuilds for the later versions
> > come out.  Unfortunately, as it stands, nitro+og is too
> > underdocumented for me to be comfortable using.
> >
> > I will definitely keep my eye out, though, and check it out again in a
> > few months.
> >
> > Pistos
> >
> > On 13/03/06, Bakki Kudva - bakki.kudva at gmail.com
> > <+nitrojesus+pistos+3978b84508.bakki.kudva#gmail.com at spamgourmet.com>
> >
> > wrote:
> >> May be simpler to use rubygems to install/upgrade nitro, rails etc.?
> >> -bakki
> >>
> >> On 3/13/06, nitrojesus.5.pistos at geoshell.com
> >>
> >> <nitrojesus.5.pistos at geoshell.com> wrote:
> >>> On 12/03/06, Fabian Buch - fabian at oggu.de
> >>> <+nitrojesus+pistos+97fd83588c.fabian#oggu.de at spamgourmet.com>
> >>>
> >>> wrote:
> >>>>> I'm on Gentoo.  I've actually emerged 0.25.0.
> >>>>
> >>>> This is seriously outdated. You should deffinitely update to
> >>>> 0.29.0!
> >>>
> >>> Hm, okay.  I'll have to craft my own e-build, but that shouldn't
> >>> be a
> >>> problem.  Too bad I'm already well into my project, using
> >>> ActiveRecord... :-\
> >>>
> >>> Pistos
> >>>
> >>> _______________________________________________
> >>> Nitro-general mailing list
> >>> Nitro-general at rubyforge.org
> >>> http://rubyforge.org/mailman/listinfo/nitro-general
> >>
> >> _______________________________________________
> >> Nitro-general mailing list
> >> Nitro-general at rubyforge.org
> >> http://rubyforge.org/mailman/listinfo/nitro-general
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From transfire at gmail.com  Tue Mar 14 08:32:21 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 14 Mar 2006 13:32:21 +0000
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
	<12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
Message-ID: <4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>

On 3/14/06, Rob Pitt <rob at motionpath.com> wrote:
> . I too prefer to respect the package manager of a system but if the
> package is out of date then it makes perfect sense to build a better
> version yourself.
>
> There is no harm at all in rolling your own ebuild with a newer
> version and I have to admit I'm not especially happy the Gentoo user
> decided to put Nitro in the portage tree if they intended to abandon
> it and not maintain it properly.
>
> I would recommend using rubygems over portage in this instance on the
> following rational (this can be applied to all distro/package
> managers with a similar situation):

I have to disagree. It's important that operating system maintainer be
in charge of package management. RubyGems seems great and all, but
could you imagine if everyone followed this path? We have to learn a
dozen interfaces to manages our programs, one for Perl, one for
Python, Lisp, Mono, Java, etc. etc. We'd be better off just doing
install the Windows way at that rate. RubyGems makes sense to us a
Ruby developers, but outside that arena it goes no where. That's not
to say Gems isn't of any worht mind you. Obviously it works great for
a Ruby-based webframework, but again that's b/c it's geared for Ruby
developers.

This leads me to a new idea and I wonder if anyone experienced in
eBuilds and/or .debs or any other ditro's package system, would like
to help out. It shouldn't be too hard to use Reap to automatically
generate these packages.

T.



From rob at motionpath.com  Tue Mar 14 09:55:08 2006
From: rob at motionpath.com (Rob Pitt)
Date: Tue, 14 Mar 2006 14:55:08 +0000
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
	<12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
	<4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>
Message-ID: <49093112-29D9-49CC-955E-4C210B8C7AF8@motionpath.com>

I still disagree about the package manager (I also use CPAN for perl,  
etc). I see these systems just as other package managers. Rubygems  
doesn't take any kind of intelligence to operate it's very simple.

I can see this is a little "emacs vs vi" though :)

I like the idea of generating distro packages with reap, good call.

Btw, vi all the way ;)

On 14 Mar 2006, at 13:32, TRANS wrote:

> On 3/14/06, Rob Pitt <rob at motionpath.com> wrote:
>> . I too prefer to respect the package manager of a system but if the
>> package is out of date then it makes perfect sense to build a better
>> version yourself.
>>
>> There is no harm at all in rolling your own ebuild with a newer
>> version and I have to admit I'm not especially happy the Gentoo user
>> decided to put Nitro in the portage tree if they intended to abandon
>> it and not maintain it properly.
>>
>> I would recommend using rubygems over portage in this instance on the
>> following rational (this can be applied to all distro/package
>> managers with a similar situation):
>
> I have to disagree. It's important that operating system maintainer be
> in charge of package management. RubyGems seems great and all, but
> could you imagine if everyone followed this path? We have to learn a
> dozen interfaces to manages our programs, one for Perl, one for
> Python, Lisp, Mono, Java, etc. etc. We'd be better off just doing
> install the Windows way at that rate. RubyGems makes sense to us a
> Ruby developers, but outside that arena it goes no where. That's not
> to say Gems isn't of any worht mind you. Obviously it works great for
> a Ruby-based webframework, but again that's b/c it's geared for Ruby
> developers.
>
> This leads me to a new idea and I wonder if anyone experienced in
> eBuilds and/or .debs or any other ditro's package system, would like
> to help out. It shouldn't be too hard to use Reap to automatically
> generate these packages.
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From nitrojesus.5.pistos at geoshell.com  Tue Mar 14 09:32:37 2006
From: nitrojesus.5.pistos at geoshell.com (nitrojesus.5.pistos at geoshell.com)
Date: Tue, 14 Mar 2006 09:32:37 -0500
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
	<12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
	<4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>
Message-ID: <6c9d9ef0603140632h3c253c36i@mail.gmail.com>

Thank you, Bryan, Rob, zimba and trans for your responses.  I'm taking
your commentary and trying to learn and absorb what I can.  I think my
decision for the moment would be to make an ebuild myself and post it
on ebuildexchange.org (where I put my other not-yet-adopted ebuilds). 
Making ebuilds for gem-based programs has proven easy in the past, so
I don't see this as being a major challenge.  This would allow me to
still respect and use Portage.

I agree with trans in that having to use a rubygems equivalent for
every other language or subsystem would be quite undesirable for me.

Pistos



From rob at motionpath.com  Tue Mar 14 10:07:51 2006
From: rob at motionpath.com (Rob Pitt)
Date: Tue, 14 Mar 2006 15:07:51 +0000
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <6c9d9ef0603140632h3c253c36i@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
	<12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
	<4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>
	<6c9d9ef0603140632h3c253c36i@mail.gmail.com>
Message-ID: <BE1B6206-A956-4411-BEC4-BD20CEC307C9@motionpath.com>

Why don't you help trans with the reap generation of ebuilds? I am  
quite sure not much will change between ebuilds other than the  
version number...

On 14 Mar 2006, at 14:32, nitrojesus.5.pistos at geoshell.com wrote:

> Thank you, Bryan, Rob, zimba and trans for your responses.  I'm taking
> your commentary and trying to learn and absorb what I can.  I think my
> decision for the moment would be to make an ebuild myself and post it
> on ebuildexchange.org (where I put my other not-yet-adopted ebuilds).
> Making ebuilds for gem-based programs has proven easy in the past, so
> I don't see this as being a major challenge.  This would allow me to
> still respect and use Portage.
>
> I agree with trans in that having to use a rubygems equivalent for
> every other language or subsystem would be quite undesirable for me.
>
> Pistos
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From fabian at oggu.de  Tue Mar 14 13:24:38 2006
From: fabian at oggu.de (Fabian Buch)
Date: Tue, 14 Mar 2006 19:24:38 +0100
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <bd93db0d0603131551g1264a627nfc681e11872d70b0@mail.gmail.com>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
	<bd93db0d0603131551g1264a627nfc681e11872d70b0@mail.gmail.com>
Message-ID: <01013e17fc3eb92bf54199e591f041b1@oggu.de>

could you copy&paste your answer below also as answer to 
http://oxyliquit.de/question/16 ? That would be very cool, since it 
might be helpful for future users.

If you are too busy, allow me to copy&paste it for you as quote.

Am 14.03.2006 um 00:51 schrieb Bryan Soto:
> On 3/12/06, Gergely Kontra <kgergely at mcl.hu> wrote:
>> Hi!
>>
>> I'm trying to run nitro under Apache2 with fcgid.

>> [Sun Mar 12 20:28:16 2006] [error] [client 192.168.1.1] Premature end 
>> of script headers: fcgi.rb
>
> First, check to ensure the shebang line on fcgi.rb is correct. When
> switching between my Windows and Linux box, that always caught me. :)
>
> If the path is okay, are you sure you configured httpd.conf correctly?
> You have to specify an intial environment. Of particular importance is
> setting up RUBYOPT if you installed nitro via gems or make use of any
> other gems. Also, Nitro makes use of mysqladmin, so you should specify
> the path to it.
>
> FastCgiConfig -initial-env RUBYOPT="rubygems" /
>                         -initial-env PATH="/usr/bin"
>
> You're getting fcgi.rb, so I assume you've already enabled mod_rewrite
> and have an .htaccess file...
>
> You might try to su to apache and running fcgi.rb from the command
> line. Though you might need to be root to do that.
>
> # su apache
> # ./fcgi.rb
>
> Or perhaps Apache2 provides more info that we're not making use of in
> the default 500 error page?
>
> Bryan



From bryan.a.soto at gmail.com  Tue Mar 14 13:43:20 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 10:43:20 -0800
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <01013e17fc3eb92bf54199e591f041b1@oggu.de>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
	<bd93db0d0603131551g1264a627nfc681e11872d70b0@mail.gmail.com>
	<01013e17fc3eb92bf54199e591f041b1@oggu.de>
Message-ID: <bd93db0d0603141043r5860ce1dj36a29eb8a17cb243@mail.gmail.com>

On 3/14/06, Fabian Buch <fabian at oggu.de> wrote:
> could you copy&paste your answer below also as answer to
> http://oxyliquit.de/question/16 ? That would be very cool, since it
> might be helpful for future users.
>
> If you are too busy, allow me to copy&paste it for you as quote.
>

I'd be happy to do it. It'd bump up my points. :)

First off though, let's make sure it actually helped. ;)

> Am 14.03.2006 um 00:51 schrieb Bryan Soto:
> > On 3/12/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> >> Hi!
> >>
> >> I'm trying to run nitro under Apache2 with fcgid.
>
> >> [Sun Mar 12 20:28:16 2006] [error] [client 192.168.1.1] Premature end
> >> of script headers: fcgi.rb
> >
> > First, check to ensure the shebang line on fcgi.rb is correct. When
> > switching between my Windows and Linux box, that always caught me. :)
> >
> > If the path is okay, are you sure you configured httpd.conf correctly?
> > You have to specify an intial environment. Of particular importance is
> > setting up RUBYOPT if you installed nitro via gems or make use of any
> > other gems. Also, Nitro makes use of mysqladmin, so you should specify
> > the path to it.
> >
> > FastCgiConfig -initial-env RUBYOPT="rubygems" /
> >                         -initial-env PATH="/usr/bin"
> >
> > You're getting fcgi.rb, so I assume you've already enabled mod_rewrite
> > and have an .htaccess file...
> >
> > You might try to su to apache and running fcgi.rb from the command
> > line. Though you might need to be root to do that.
> >
> > # su apache
> > # ./fcgi.rb
> >
> > Or perhaps Apache2 provides more info that we're not making use of in
> > the default 500 error page?
> >
> > Bryan
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 14 16:21:35 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 13:21:35 -0800
Subject: [Nitro] PATCH: nitro-fix-dot-in-url
Message-ID: <bd93db0d0603141320i456f91can6bd40ae487fb5530@mail.gmail.com>

* nitro-fix-dot-in-urls
Should make Nitro process URLs with a '.' (dot/period) in them.

Changes Webrick, Mongrel and cgi.rb. Cgi.rb is what all the other adapters use.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton
-------------- next part --------------
A non-text attachment was scrubbed...
Name: nitro-fix-dot-in-urls.zip
Type: application/zip
Size: 13030 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060314/121b2cfe/attachment.zip 

From bryan.a.soto at gmail.com  Tue Mar 14 17:02:35 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 14:02:35 -0800
Subject: [Nitro] [PATCH] George's routes fix as a patch
In-Reply-To: <8C85238A-A349-4C94-88B7-2F8741E9A74E@motionpath.com>
References: <8C85238A-A349-4C94-88B7-2F8741E9A74E@motionpath.com>
Message-ID: <bd93db0d0603141402rbd2b2a0vc43df1bf94bece11@mail.gmail.com>

On 3/13/06, Rob Pitt <rob at motionpath.com> wrote:
> I converted George's e-mail to James Britt into a patch against
> repo.nitrohq.com glycerin current. I also added a tiny fix in case
> you add routes with the method Bryan Soto suggested but do not supply
> a params hash.

Rob, thanks for submitting this as a patch. I applied to devalb.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 14 18:20:11 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 15:20:11 -0800
Subject: [Nitro] PATCH: fixes-zimbas-compiler-patch
Message-ID: <bd93db0d0603141520t32420390ua0d2e390def34266@mail.gmail.com>

* fixes-zimbas-compiler-patch
Adds a require for zimba's new Nitro::Template wrapper to
Glue::Template. Changes references from Glue::Template to
Nitro::Template and adds a require for 'glue/template' do to a use of
Glue::FileTemplate.

Zimba, could you check this one out? nitro/test/nitro/tc_controller.rb
was failing because it wasn't correctly evaluating a template.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fixes-zimbas-compiler-patch.zip
Type: application/zip
Size: 11979 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060314/37d16e47/attachment.zip 

From bryan.a.soto at gmail.com  Tue Mar 14 19:15:40 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 16:15:40 -0800
Subject: [Nitro] PATCH: fixes-zimbas-compiler-patch
In-Reply-To: <bd93db0d0603141520t32420390ua0d2e390def34266@mail.gmail.com>
References: <bd93db0d0603141520t32420390ua0d2e390def34266@mail.gmail.com>
Message-ID: <bd93db0d0603141615u1dac05d9ja6d5ad070e2d8f7f@mail.gmail.com>

Please disregard. This was for the prior patch. Not the latest.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 14 19:41:39 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 16:41:39 -0800
Subject: [Nitro] [PATCH] Moved the transformer pipeline at the controller
In-Reply-To: <200603140829.21093.zimba.tm@gmail.com>
References: <200603131413.56508.zimba.tm@gmail.com>
	<bd93db0d0603131530u60971e09q6d680e481b43a4ff@mail.gmail.com>
	<200603140829.21093.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603141641t3c6ae8e6o66219d568a9a233a@mail.gmail.com>

On 3/13/06, zimba.tm <zimba.tm at gmail.com> wrote:
> On Tuesday 14 March 2006 00:30, Bryan Soto wrote:
> > Interesting idea. I notice you've been making a few changes to the
> > compilers recently. General cleanup or did you have something in mind?
>
> Well, first I needed to unify all compilers.transform methods, what I did in a
> previous patch. Then I had the idea of putting the transformers pipeline in
> the controller to allow better componentization. George must be working on a
> similar thing with the "part" feature.
>
> The idea is to put any nitro app in a /part sub-folder and mount it using one
> command.
>
> Btw the code really needs some cleanup but it's hard to tell what is legacy or
> not so I don't like to touch.
>

Hmm... I've tried current repo with and without this patch against the
blog example, which does a Compiler.setup_template_transformation.
Against gems 0.29.0, adding a comment works correctly, but with these
compiler patches it doesn't. I see a render /error in the log, but the
error page doesn't actually render.

I'll see if I can figure something out, but I thought I'd post this
for some more eyes.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 14 19:59:15 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 16:59:15 -0800
Subject: [Nitro] [PATCH] two small flare-patches
In-Reply-To: <200603141625.40585.manveru@weez.co.jp>
References: <200603141625.40585.manveru@weez.co.jp>
Message-ID: <bd93db0d0603141659o5cbed8d8y9b4dbc6f1454ac08@mail.gmail.com>

On 3/13/06, Michael Fellinger <manveru at weez.co.jp> wrote:
> The first patch corrects a problem with the RSS-feeds where the comments-feed
> was not valid because the object it refered to didn't provide the to_href
> method, added the method and pointing the link to the comment to
> "/article/#@article_oid", which works quit well for now, still it would be
> nice to have a link directly to the comments, not sure how we would do that?
> (in-page-anchors?)
>

Hmm... Is your blog running with this patch? It'd be nice to see it in
action. By the way, nice redesign. I like the pictures too. :)

> The other patch removes a big annoyance that i have to fix everytime
> installing flare, removing the line:
>  validate_length :body, :range => 2..1000
> since the author should know him/herself how much belongs into one post and i
> have myself crossed that limit far too often (losing all my text if it was
> only one character too much).
>

That doesn't seem very user-friendly. It seems reasonable not to limit
the body, but not that you lose all your text if you go over the
limit... This one I don't think anyone will object to, though I wonder
if there's something we can do about the fact that you lose it all
after the validation error.

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 14 20:24:54 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 17:24:54 -0800
Subject: [Nitro] Patches applied
Message-ID: <bd93db0d0603141724g79edf0a0i49d8aa90404a13f5@mail.gmail.com>

I've applied a couple of patches. Run

$ darcs pull http://devlab.oree.ch/darcs/nitrohq

if you've been pulling from http://repo.nitrohq.com. darcs pull always
uses the latest repo specified, so after you specify devlab once, you
won't have to again.

I'm going to try and get through some more submissions (thanks for
submitting and my apologies that they weren't responded to sooner). If
you've reviewed them, tested them, etc. please reply to the
submission. It'd be a big help. :)

Thanks,

Bryan (whose sorry he hasn't been able to spend more time with Nitro of late)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 14 20:37:26 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 17:37:26 -0800
Subject: [Nitro] [PATCH] two small flare-patches
In-Reply-To: <bd93db0d0603141659o5cbed8d8y9b4dbc6f1454ac08@mail.gmail.com>
References: <200603141625.40585.manveru@weez.co.jp>
	<bd93db0d0603141659o5cbed8d8y9b4dbc6f1454ac08@mail.gmail.com>
Message-ID: <bd93db0d0603141737od20e6b7i990cd1eb8e2616bf@mail.gmail.com>

On 3/14/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Hmm... Is your blog running with this patch? It'd be nice to see it in
> action. By the way, nice redesign. I like the pictures too. :)

Yes, you are. Standard flare just gives an access denied error
(verifiable from gmosx.com).

Any other opinions on how to handle this? I'm not too familiar with
blogs... Do they typically link back to the article or to a separate
comment page?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 14 21:19:32 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 18:19:32 -0800
Subject: [Nitro] PATCH: tests-camel-case
Message-ID: <bd93db0d0603141819o6ac84e55rb4821831cc3f72dd@mail.gmail.com>

* tests-camel-case
Adds test cases submitted on Trac.

tc_camel_case.rb: credited to vsgeuip AT gmail DOT com
tc_camel_case_joins.rb: credited to kashia AT vfemail DOT net

See ticket #13 at http://devlab.oree.ch/trac/nitrohq/ticket/13
--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton
-------------- next part --------------
A non-text attachment was scrubbed...
Name: tests-camel-case.zip
Type: application/zip
Size: 12261 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060314/cd67ea1f/attachment.zip 

From nitrojesus.5.pistos at geoshell.com  Tue Mar 14 22:04:31 2006
From: nitrojesus.5.pistos at geoshell.com (nitrojesus.5.pistos at geoshell.com)
Date: Tue, 14 Mar 2006 22:04:31 -0500
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <49093112-29D9-49CC-955E-4C210B8C7AF8@motionpath.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
	<12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
	<4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>
	<49093112-29D9-49CC-955E-4C210B8C7AF8@motionpath.com>
Message-ID: <6c9d9ef0603141904x42376a7dh@mail.gmail.com>

On 14/03/06, Rob Pitt - rob at motionpath.com wrote:
> I still disagree about the package manager (I also use CPAN for perl,
> etc). I see these systems just as other package managers. Rubygems
> doesn't take any kind of intelligence to operate it's very simple.

Yes, it's not a question of knowing how to use it, it's more of ...
well, yeah, let's just let this topic die.  :)

> I can see this is a little "emacs vs vi" though :)
> Btw, vi all the way ;)

Oh don't start.  ;)  I'm a bit shielded from such debates, though, as
you can imagine (I don't use either one).

FWIW, I sent an e-mail to the nitro ebuild maintainer.  No reply
[yet].  You guys are right in that I don't imagine it to change much
from version to version, since it uses rubygems.

Pistos



From transfire at gmail.com  Tue Mar 14 22:56:02 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 15 Mar 2006 03:56:02 +0000
Subject: [Nitro] Og bug -- conflict with open-uri
In-Reply-To: <6c9d9ef0603141904x42376a7dh@mail.gmail.com>
References: <6c9d9ef0603112205h3dbf68e5k@mail.gmail.com>
	<ed0dd90e400f4154f63623a248464941@oggu.de>
	<6c9d9ef0603130706w1c10b33aj@mail.gmail.com>
	<f7f43c920603131248x721bfcfep869ed381e25eadd1@mail.gmail.com>
	<6c9d9ef0603131716s627e15d8l@mail.gmail.com>
	<12A76AB5-97E4-40CA-90D8-01D837931190@motionpath.com>
	<4b6f054f0603140532p59478127l5322ed690c8dde02@mail.gmail.com>
	<49093112-29D9-49CC-955E-4C210B8C7AF8@motionpath.com>
	<6c9d9ef0603141904x42376a7dh@mail.gmail.com>
Message-ID: <4b6f054f0603141956h44c0d030sf6460d9cf6b75237@mail.gmail.com>

FYI, I got pretty far in a .deb builder for Reap today. Ran into a
problem with setup.rb's --prefix option.

It's interesting how this works. I have to actually pretend install
the package using setup.rb but into a special directory and package
THAT up to create the .deb.

T.



From kashia at vfemail.net  Wed Mar 15 01:20:29 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Wed, 15 Mar 2006 07:20:29 +0100
Subject: [Nitro] PATCH: nitro-fix-dot-in-url
In-Reply-To: <bd93db0d0603141320i456f91can6bd40ae487fb5530@mail.gmail.com>
References: <bd93db0d0603141320i456f91can6bd40ae487fb5530@mail.gmail.com>
Message-ID: <op.s6fykfxlg3glhz@jo.local>

> * nitro-fix-dot-in-urls
> Should make Nitro process URLs with a '.' (dot/period) in them.

Thank you very much! :D

(I hate fiddling around with already installed 0.29 ;), now lets
see, when a .30 comes out :P)

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From bryan.a.soto at gmail.com  Wed Mar 15 02:15:24 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 14 Mar 2006 23:15:24 -0800
Subject: [Nitro] PATCH: nitro-fix-dot-in-url
In-Reply-To: <op.s6fykfxlg3glhz@jo.local>
References: <bd93db0d0603141320i456f91can6bd40ae487fb5530@mail.gmail.com>
	<op.s6fykfxlg3glhz@jo.local>
Message-ID: <bd93db0d0603142315u689ba3d9ja6cdca6cddcda34b@mail.gmail.com>

On 3/14/06, Kashia Buch <kashia at vfemail.net> wrote:
> > * nitro-fix-dot-in-urls
> > Should make Nitro process URLs with a '.' (dot/period) in them.
>
> Thank you very much! :D
>

Thank me if it doesn't break anything in new and interesting ways. ;)

> (I hate fiddling around with already installed 0.29 ;), now lets
> see, when a .30 comes out :P)
>

Actually, beta gems might not be too bad an idea. Does anyone know how
to set that up?

Thanks for the nudge on the upcoming release. We definitely have some
things to figure out...

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Wed Mar 15 03:25:33 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 15 Mar 2006 00:25:33 -0800
Subject: [Nitro] STI (Again =))
In-Reply-To: <30CF8422-9A3A-46FB-8739-CBBF8BBE05EC@motionpath.com>
References: <30CF8422-9A3A-46FB-8739-CBBF8BBE05EC@motionpath.com>
Message-ID: <bd93db0d0603150025o523e730ev5336e07c0d2405ca@mail.gmail.com>

On 3/13/06, Rob Pitt <rob at motionpath.com> wrote:
> There was a bug in my fix to the last STI bug causing it to only
> return the correct STI class in a multiple-find case sometimes (it
> never returned it before).
>
> Now it always returns the right class...

Hi Rob,

I was trying to add a testcase and can't seem to trigger this...
Perhaps you can help?

<testcase>
require File.join(File.dirname(__FILE__), '..', 'CONFIG.rb')

require 'test/unit'

require 'og'

class TC_Sti_OgType < Test::Unit::TestCase
  class Human
    property :name

    schema_inheritance

    def initialize(name)
      @name = name
    end
  end

  class Parent < Human
    property :job
  end

  class Child < Human
    property :toys
  end

  $og1.manage_classes(Human, Parent, Child)

  def setup
    Parent.create('mom')
    Parent.create('dad')
    Child.create('son')
  end

  def test_sti_all
    all = Human.find(:condition => 'name IS NOT NULL')
    assert_equal(3, all.size)
    parents = all.select {|h| h.class == Parent }
    children = all.select {|h| h.class == Child }
    assert_equal(2, parents.size)
    assert_equal(1, children.size)
    assert_equal(%w[dad mom], parents.sort {|x,y| x.name <=>
y.name}.map {|p| p.name } )
    assert_equal(%w[son], children.map {|c| c.name } )
  end
end
</testcase>

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From zimba.tm at gmail.com  Wed Mar 15 04:46:56 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Wed, 15 Mar 2006 10:46:56 +0100
Subject: [Nitro] Split the repos ?
Message-ID: <200603151046.56783.zimba.tm@gmail.com>

What do you think if we would split the different libraries in different 
repos ?

I would like everybody to give it's input.

= Pros
* Less records in darcs. Darcs doesn't grow well with lots of commits.
* Would maybe push to remove the IMHO unnecessary Glue lib.
* Every library is on the same level, thus allowing Facet/Calibre to join the 
party.
* Make bugs more distinct between the different libraries. Eg. It's easier to 
look at all Og changes if they are all listed in the same repo.

= Cons
* Work for me. I would have to setup multiple tracs :-p

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From guillaume.pierronnet at gmail.com  Wed Mar 15 10:15:19 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Wed, 15 Mar 2006 16:15:19 +0100
Subject: [Nitro] nested templates calls and shared data
Message-ID: <6a7d49ca0603150715o7a13c22br@mail.gmail.com>

hi list,

i got a problem. Imagine you have a template with some
"nitro-generated ajax magic" in it but who doesn't generate a full
html page in itself and that is aimed to be called from another
template.

This template looks like this:
album_list.xhtml:

<a href="#" client='check_album'><img src="images/check.gif"/></a>

If you call it from the "main" template who generates the complete
page, you will not see the javascript appending before '</body>',
because the js_buffer is not shared across differents render() method
calls.

main.xhtml:

<html>
  <head>
    <title>blah blah</title>
  </head>just
<body>
  <render href="album_list" />
</body>
</html>


the controller:

class MyController
  class Client
    def check_album
      ajax_update 'tools_block', {
        :action => 'checked_albums',
        :params => "some parameters"
      }
    end
  end
end

i spent all my week-end on trying to implement a sort of 'global
compiling pipeline', with no success.

does anybody have a idea on how to fix it ?



From rob at motionpath.com  Wed Mar 15 10:53:09 2006
From: rob at motionpath.com (Rob Pitt)
Date: Wed, 15 Mar 2006 15:53:09 +0000
Subject: [Nitro] STI (Again =))
In-Reply-To: <bd93db0d0603150025o523e730ev5336e07c0d2405ca@mail.gmail.com>
References: <30CF8422-9A3A-46FB-8739-CBBF8BBE05EC@motionpath.com>
	<bd93db0d0603150025o523e730ev5336e07c0d2405ca@mail.gmail.com>
Message-ID: <BE1245B4-6280-4E32-973A-5BC0AD289F23@motionpath.com>

Hrm,

Don't know why you can't trigger it :/

I can't remember how I triggered it either :/

I think it was with:

   def parents
     sql = %!SELECT * FROM ogpage WHERE oid IN (SELECT page_oid as  
oid FROM ogj_page_page where page_oid2=#{pk} UNION SELECT p.oid FROM  
ogpage p WHERE #{pk} IN ("child_1_oid","child_2_oid","child_3_oid"))!
     self.class.find(:sql => sql)
   end

But I don't see why this is any different to what you did for this  
purpose.

All I can say is it definitely was broken and when I looked at the  
code IIRC the first patch I submitted set the klass variable in the  
read_all method (used for klass.og_allocate) for the first row  
returned and before I submitted any patches, klass was always just  
equal to ThisClass in ThisClass.find(:sql => foo)

ChrisFarms also found it broken in some project he did with a file  
system browser.

I'll try and take a better look at it but it's often hard to find  
time to work on Og/Nitro other than to fix bugs when they stop me  
doing work...

On 15 Mar 2006, at 08:25, Bryan Soto wrote:

> On 3/13/06, Rob Pitt <rob at motionpath.com> wrote:
>> There was a bug in my fix to the last STI bug causing it to only
>> return the correct STI class in a multiple-find case sometimes (it
>> never returned it before).
>>
>> Now it always returns the right class...
>
> Hi Rob,
>
> I was trying to add a testcase and can't seem to trigger this...
> Perhaps you can help?
>
> <testcase>
> require File.join(File.dirname(__FILE__), '..', 'CONFIG.rb')
>
> require 'test/unit'
>
> require 'og'
>
> class TC_Sti_OgType < Test::Unit::TestCase
>   class Human
>     property :name
>
>     schema_inheritance
>
>     def initialize(name)
>       @name = name
>     end
>   end
>
>   class Parent < Human
>     property :job
>   end
>
>   class Child < Human
>     property :toys
>   end
>
>   $og1.manage_classes(Human, Parent, Child)
>
>   def setup
>     Parent.create('mom')
>     Parent.create('dad')
>     Child.create('son')
>   end
>
>   def test_sti_all
>     all = Human.find(:condition => 'name IS NOT NULL')
>     assert_equal(3, all.size)
>     parents = all.select {|h| h.class == Parent }
>     children = all.select {|h| h.class == Child }
>     assert_equal(2, parents.size)
>     assert_equal(1, children.size)
>     assert_equal(%w[dad mom], parents.sort {|x,y| x.name <=>
> y.name}.map {|p| p.name } )
>     assert_equal(%w[son], children.map {|c| c.name } )
>   end
> end
> </testcase>
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general




From transfire at gmail.com  Wed Mar 15 10:53:26 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 15 Mar 2006 15:53:26 +0000
Subject: [Nitro] Split the repos ?
In-Reply-To: <200603151046.56783.zimba.tm@gmail.com>
References: <200603151046.56783.zimba.tm@gmail.com>
Message-ID: <4b6f054f0603150753l2bc54de1s3e5f8e4564e18c9c@mail.gmail.com>

On 3/15/06, zimba.tm <zimba.tm at gmail.com> wrote:
> What do you think if we would split the different libraries in different
> repos ?

Probably a good idea. It's nice that all the libs are sycn'd by
version from an outside sort of viewpoint. But its not very realistic
from a development view point.

T.



From manveru at weez.co.jp  Thu Mar 16 01:13:03 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Thu, 16 Mar 2006 15:13:03 +0900
Subject: [Nitro] [PATCH] two small flare-patches
In-Reply-To: <bd93db0d0603141659o5cbed8d8y9b4dbc6f1454ac08@mail.gmail.com>
References: <200603141625.40585.manveru@weez.co.jp>
	<bd93db0d0603141659o5cbed8d8y9b4dbc6f1454ac08@mail.gmail.com>
Message-ID: <200603161513.03098.manveru@weez.co.jp>

On Wednesday 15 March 2006 09:59, Bryan Soto wrote:
> On 3/13/06, Michael Fellinger <manveru at weez.co.jp> wrote:
> > The first patch corrects a problem with the RSS-feeds where the
> > comments-feed was not valid because the object it refered to didn't
> > provide the to_href method, added the method and pointing the link to the
> > comment to "/article/#@article_oid", which works quit well for now, still
> > it would be nice to have a link directly to the comments, not sure how we
> > would do that? (in-page-anchors?)
>
> Hmm... Is your blog running with this patch? It'd be nice to see it in
> action. By the way, nice redesign. I like the pictures too. :)
>
> > The other patch removes a big annoyance that i have to fix everytime
> > installing flare, removing the line:
> >  validate_length :body, :range => 2..1000
> > since the author should know him/herself how much belongs into one post
> > and i have myself crossed that limit far too often (losing all my text if
> > it was only one character too much).
>
> That doesn't seem very user-friendly. It seems reasonable not to limit
> the body, but not that you lose all your text if you go over the
> limit... This one I don't think anyone will object to, though I wonder
> if there's something we can do about the fact that you lose it all
> after the validation error.
>

Well, it might not seem user-friendly, but that is the way part/admin works. 
And since part/admin is what is used for managing the blog, there is lilttle 
we can do right now apart from writing a seperate administration-interface 
for the flare, which would not serve us well for the future imho, since it 
will be outdated pretty soon.
So, let's talk about further improvments in part/admin, like 
javascript-checking of validation-constraints? or a nice extension for the 
javascript-helper to allow that?
something like:
firstform_name, :length => (2..25)
firstform_name, :valid_chars => ["a".."z", "A".."Z", 0..9]
firstform_pass, :length => (6..255)

not sure how we could translate that to neat JS, but i am _sure_ it can be 
done :)

> Bryan
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From bryan.a.soto at gmail.com  Thu Mar 16 03:25:53 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 16 Mar 2006 00:25:53 -0800
Subject: [Nitro] kirby patches and a test fix
In-Reply-To: <44161AFB.8030801@digitalvalence.com>
References: <44132BC3.9080905@digitalvalence.com>
	<bd93db0d0603131407j20fdd1dbjdf05b412201fa0e1@mail.gmail.com>
	<44161AFB.8030801@digitalvalence.com>
Message-ID: <bd93db0d0603160025t2d26a88bi82ff24de022c69bd@mail.gmail.com>

On 3/13/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
> You're quite welcome; I only wish I had more time to contribute to
> things like this.  The story of us all ;)
>
> Dylan
>

I certainly hope you find some more time. :)

Before patch:
43 tests,  78 assertions, 6 failures, 14 errors

After patch:
43 tests, 126 assertions, 5 failures, 6 errors

Some conflicts with the recently applied og_properties fix in the GTK2 patch.

I like your idea of storing ogtype in the SchemaInheritanceBase
module. Perhaps that should be applied to the other stores. Makes
things simpler if it's just another Og managed property.

One style tip, though you seem to know it. Ruby programs typically
don't use camel case for variables, i.e. resultSet would be result_set
and testCondition would be test_condition. Not a big deal at all, of
course. Just thought I'd mention it not knowing your familiarity with
Ruby. :)

I'll review this a bit more as it's a fairly big change, but I'm
liking what I see so far. :)

Bryan



From bryan.a.soto at gmail.com  Thu Mar 16 03:28:08 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 16 Mar 2006 00:28:08 -0800
Subject: [Nitro] [PATCH] two small flare-patches
In-Reply-To: <200603161513.03098.manveru@weez.co.jp>
References: <200603141625.40585.manveru@weez.co.jp>
	<bd93db0d0603141659o5cbed8d8y9b4dbc6f1454ac08@mail.gmail.com>
	<200603161513.03098.manveru@weez.co.jp>
Message-ID: <bd93db0d0603160028r70844d49kc4422dc6d78e71c8@mail.gmail.com>

On 3/15/06, Michael Fellinger <manveru at weez.co.jp> wrote:
> > That doesn't seem very user-friendly. It seems reasonable not to limit
> > the body, but not that you lose all your text if you go over the
> > limit... This one I don't think anyone will object to, though I wonder
> > if there's something we can do about the fact that you lose it all
> > after the validation error.
> >
>
> Well, it might not seem user-friendly, but that is the way part/admin works.
> And since part/admin is what is used for managing the blog, there is lilttle
> we can do right now apart from writing a seperate administration-interface
> for the flare, which would not serve us well for the future imho, since it
> will be outdated pretty soon.
> So, let's talk about further improvments in part/admin, like
> javascript-checking of validation-constraints? or a nice extension for the
> javascript-helper to allow that?
> something like:
> firstform_name, :length => (2..25)
> firstform_name, :valid_chars => ["a".."z", "A".."Z", 0..9]
> firstform_pass, :length => (6..255)
>
> not sure how we could translate that to neat JS, but i am _sure_ it can be
> done :)

You beat me. I was just thinking it should throw you back to the entry
form with an error message. ;)



From zimba.tm at gmail.com  Thu Mar 16 04:36:09 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 16 Mar 2006 10:36:09 +0100
Subject: [Nitro] [facet] Methodize, Pathize and Modulize
Message-ID: <200603161036.09635.zimba.tm@gmail.com>

Hi list,

I'd like to advertise three methods that were added in facets-1.1.0 I've asked 
TRANS to integrate.

== Methods

String#methodize
String#pathize
String#modulize

== Usage

string = MyModule::MyClass.name
string.methodize #=> "my_module__my_class"
string.pathize #=> "my_module/my_class"
string.modulize #=> "MyModule::MyClass"
# All transformations are revertible
string.methodize.pathize.modulize #=> "MyModule::MyClass"

== Explanation

The goal is to provide methods for ruby magic. Various components of Nitro use 
similar methods. For example, Scaffolding,  that builds Controller methods 
from manages class's names. There are other parts also that I don't remember 
by heart.

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From transfire at gmail.com  Thu Mar 16 08:14:21 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 16 Mar 2006 13:14:21 +0000
Subject: [Nitro] [facet] Methodize, Pathize and Modulize
In-Reply-To: <200603161036.09635.zimba.tm@gmail.com>
References: <200603161036.09635.zimba.tm@gmail.com>
Message-ID: <4b6f054f0603160514j61c15f04y6fe16b07b5fd5681@mail.gmail.com>

Cool. How about posting to ruby-talk?

T.


On 3/16/06, zimba.tm <zimba.tm at gmail.com> wrote:
> Hi list,
>
> I'd like to advertise three methods that were added in facets-1.1.0 I've asked
> TRANS to integrate.
>
> == Methods
>
> String#methodize
> String#pathize
> String#modulize
>
> == Usage
>
> string = MyModule::MyClass.name
> string.methodize #=> "my_module__my_class"
> string.pathize #=> "my_module/my_class"
> string.modulize #=> "MyModule::MyClass"
> # All transformations are revertible
> string.methodize.pathize.modulize #=> "MyModule::MyClass"
>
> == Explanation
>
> The goal is to provide methods for ruby magic. Various components of Nitro use
> similar methods. For example, Scaffolding,  that builds Controller methods
> from manages class's names. There are other parts also that I don't remember
> by heart.
>
> --
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
( o-  ??? went BOOM
 //    trans.
/ /    transfire at gmail.com


From transfire at gmail.com  Thu Mar 16 10:32:23 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 16 Mar 2006 15:32:23 +0000
Subject: [Nitro] Split the repos ?
In-Reply-To: <4b6f054f0603150753l2bc54de1s3e5f8e4564e18c9c@mail.gmail.com>
References: <200603151046.56783.zimba.tm@gmail.com>
	<4b6f054f0603150753l2bc54de1s3e5f8e4564e18c9c@mail.gmail.com>
Message-ID: <4b6f054f0603160732l57dbc09bv77455d233202f12a@mail.gmail.com>

So is split repos a go?



From zimba.tm at gmail.com  Thu Mar 16 13:21:07 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Thu, 16 Mar 2006 19:21:07 +0100
Subject: [Nitro] Split the repos ?
In-Reply-To: <4b6f054f0603160732l57dbc09bv77455d233202f12a@mail.gmail.com>
References: <200603151046.56783.zimba.tm@gmail.com>
	<4b6f054f0603150753l2bc54de1s3e5f8e4564e18c9c@mail.gmail.com>
	<4b6f054f0603160732l57dbc09bv77455d233202f12a@mail.gmail.com>
Message-ID: <200603161921.08066.zimba.tm@gmail.com>

On Thursday 16 March 2006 16:32, TRANS wrote:
> So is split repos a go?

I'm waiting for George to give the Go

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From dylanb at digitalvalence.com  Thu Mar 16 20:31:14 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Thu, 16 Mar 2006 19:31:14 -0600
Subject: [Nitro] kirby patches and a test fix
In-Reply-To: <bd93db0d0603160025t2d26a88bi82ff24de022c69bd@mail.gmail.com>
References: <44132BC3.9080905@digitalvalence.com>	<bd93db0d0603131407j20fdd1dbjdf05b412201fa0e1@mail.gmail.com>	<44161AFB.8030801@digitalvalence.com>
	<bd93db0d0603160025t2d26a88bi82ff24de022c69bd@mail.gmail.com>
Message-ID: <441A1162.1070707@digitalvalence.com>

Let me know if there are any bugs you don't want/have time to just patch 
up; as for the style tip, I could swear that I read that method names 
were underscored and variable names were camel case; but I see now that 
the latter is totally incorrect.  I must have been in a code induced 
haze when I hallucinated that.  I read so much Java code during my day 
job that I didn't notice the disparity.  Thanks for bringing it to my 
attention.

I'll have to earmark some time to reread the unofficial style guidelines 
at caliban. 

Bryan Soto wrote:
> On 3/13/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
>   
>> You're quite welcome; I only wish I had more time to contribute to
>> things like this.  The story of us all ;)
>>
>> Dylan
>>
>>     
>
> I certainly hope you find some more time. :)
>
> Before patch:
> 43 tests,  78 assertions, 6 failures, 14 errors
>
> After patch:
> 43 tests, 126 assertions, 5 failures, 6 errors
>
> Some conflicts with the recently applied og_properties fix in the GTK2 patch.
>
> I like your idea of storing ogtype in the SchemaInheritanceBase
> module. Perhaps that should be applied to the other stores. Makes
> things simpler if it's just another Og managed property.
>
> One style tip, though you seem to know it. Ruby programs typically
> don't use camel case for variables, i.e. resultSet would be result_set
> and testCondition would be test_condition. Not a big deal at all, of
> course. Just thought I'd mention it not knowing your familiarity with
> Ruby. :)
>
> I'll review this a bit more as it's a fairly big change, but I'm
> liking what I see so far. :)
>
> Bryan
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>   



From bryan.a.soto at gmail.com  Fri Mar 17 01:54:29 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 16 Mar 2006 22:54:29 -0800
Subject: [Nitro] Repo update.
Message-ID: <bd93db0d0603162254v516d9932xcd3caeb63a1da053@mail.gmail.com>

I've added a few patches to my repo:

darcs pull http://devlab.oree.ch/~bryan/repo

Zimba's compiler pipeline move.
Every controller can have their own pipeline. It's seems to have a
problem running the blog example, but I like the idea.

My fix for periods in urls.

A config.rb file for the nitro test suite. Sets $NITRO_NO_ENVIRONMENT
to true to prevent the chdir call.

I was hesitant to apply them to the main repo without some feedback.
Hopefully we'll get some this way.

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Sat Mar 18 02:17:30 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 17 Mar 2006 23:17:30 -0800
Subject: [Nitro] [PATCH] littles fixes
In-Reply-To: <6a7d49ca0603091400u63fbb495i@mail.gmail.com>
References: <6a7d49ca0603091400u63fbb495i@mail.gmail.com>
Message-ID: <bd93db0d0603172317k3fc7d358v6c78e9a06fe6ce09@mail.gmail.com>

On 3/9/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> bundle content:
>
>   * require fix in element.rb

Both of the requires are valid and both seem to do different things.
I'm not sure whether to apply this one or not.

>   * some tiny spark fixes

Applied.

>   * fixes on Glue::Taggable

Applied.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From george.moschovitis at gmail.com  Sat Mar 18 05:05:23 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Mar 2006 12:05:23 +0200
Subject: [Nitro] Where is the up to date repo?
Message-ID: <afd80b540603180205ue6827e7wa6f9d6beab6996d1@mail.gmail.com>

are allpatches applied to the devlab repo?

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Sat Mar 18 05:09:41 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Mar 2006 12:09:41 +0200
Subject: [Nitro] Split the repos ?
In-Reply-To: <200603161921.08066.zimba.tm@gmail.com>
References: <200603151046.56783.zimba.tm@gmail.com>
	<4b6f054f0603150753l2bc54de1s3e5f8e4564e18c9c@mail.gmail.com>
	<4b6f054f0603160732l57dbc09bv77455d233202f12a@mail.gmail.com>
	<200603161921.08066.zimba.tm@gmail.com>
Message-ID: <afd80b540603180209p170d06b8ue71d90f97b015335@mail.gmail.com>

I think it is easier to maintain one repo. You have all files
organized in one directory, and you just have to do

darcs pull
or
darcs send -o bundle

and stuff like that...

I would prefer to stay with one repo for the moment. The only benefit
real of multiple repos is that darcs will keep separate CHANGELOGS.

anw, if more people think that multiple repos are better, then lets go
for it. My opinion is that one repo is much easier to manage though.


-g.

On 3/16/06, zimba.tm <zimba.tm at gmail.com> wrote:
> On Thursday 16 March 2006 16:32, TRANS wrote:
> > So is split repos a go?
>
> I'm waiting for George to give the Go
>
> --
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Sat Mar 18 05:15:36 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Mar 2006 12:15:36 +0200
Subject: [Nitro] [facet] Methodize, Pathize and Modulize
In-Reply-To: <4b6f054f0603160514j61c15f04y6fe16b07b5fd5681@mail.gmail.com>
References: <200603161036.09635.zimba.tm@gmail.com>
	<4b6f054f0603160514j61c15f04y6fe16b07b5fd5681@mail.gmail.com>
Message-ID: <afd80b540603180215h6eda7936lc71935b467384904@mail.gmail.com>

nice ;-)

-g.

On 3/16/06, TRANS <transfire at gmail.com> wrote:
> Cool. How about posting to ruby-talk?
>
> T.
>
>
> On 3/16/06, zimba.tm <zimba.tm at gmail.com> wrote:
> > Hi list,
> >
> > I'd like to advertise three methods that were added in facets-1.1.0 I've asked
> > TRANS to integrate.
> >
> > == Methods
> >
> > String#methodize
> > String#pathize
> > String#modulize
> >
> > == Usage
> >
> > string = MyModule::MyClass.name
> > string.methodize #=> "my_module__my_class"
> > string.pathize #=> "my_module/my_class"
> > string.modulize #=> "MyModule::MyClass"
> > # All transformations are revertible
> > string.methodize.pathize.modulize #=> "MyModule::MyClass"
> >
> > == Explanation
> >
> > The goal is to provide methods for ruby magic. Various components of Nitro use
> > similar methods. For example, Scaffolding,  that builds Controller methods
> > from manages class's names. There are other parts also that I don't remember
> > by heart.
> >
> > --
> > Cheers,
> >   zimba.tm
> >
> > weblog : http://zimba.oree.ch
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> ( o-  ??? went BOOM
>  //    trans.
> / /    transfire at gmail.com
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com


From transfire at gmail.com  Sat Mar 18 08:41:51 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 18 Mar 2006 13:41:51 +0000
Subject: [Nitro] Split the repos ?
In-Reply-To: <afd80b540603180209p170d06b8ue71d90f97b015335@mail.gmail.com>
References: <200603151046.56783.zimba.tm@gmail.com>
	<4b6f054f0603150753l2bc54de1s3e5f8e4564e18c9c@mail.gmail.com>
	<4b6f054f0603160732l57dbc09bv77455d233202f12a@mail.gmail.com>
	<200603161921.08066.zimba.tm@gmail.com>
	<afd80b540603180209p170d06b8ue71d90f97b015335@mail.gmail.com>
Message-ID: <4b6f054f0603180541gb04af0fo403a3069887ceef8@mail.gmail.com>

On 3/18/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> I think it is easier to maintain one repo. You have all files
> organized in one directory, and you just have to do
>
> darcs pull
> or
> darcs send -o bundle
>
> and stuff like that...
>
> I would prefer to stay with one repo for the moment. The only benefit
> real of multiple repos is that darcs will keep separate CHANGELOGS.
>
> anw, if more people think that multiple repos are better, then lets go
> for it. My opinion is that one repo is much easier to manage though.

It will be a problem for Facets/Calibre since they currently have much
higher version numbers. Myabe for now we should set up a separate Repo
for these?

Oh my G! It that the solution to the one package for both libs I've
been looking for this whole frig'n time? Two projects under one repo?
Doh!

T.



From george.moschovitis at gmail.com  Sat Mar 18 09:32:36 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sat, 18 Mar 2006 16:32:36 +0200
Subject: [Nitro] Split the repos ?
In-Reply-To: <4b6f054f0603180541gb04af0fo403a3069887ceef8@mail.gmail.com>
References: <200603151046.56783.zimba.tm@gmail.com>
	<4b6f054f0603150753l2bc54de1s3e5f8e4564e18c9c@mail.gmail.com>
	<4b6f054f0603160732l57dbc09bv77455d233202f12a@mail.gmail.com>
	<200603161921.08066.zimba.tm@gmail.com>
	<afd80b540603180209p170d06b8ue71d90f97b015335@mail.gmail.com>
	<4b6f054f0603180541gb04af0fo403a3069887ceef8@mail.gmail.com>
Message-ID: <afd80b540603180632p4c74c13bod6c5fb96f8ab1959@mail.gmail.com>

> It will be a problem for Facets/Calibre since they currently have much
> higher version numbers. Myabe for now we should set up a separate Repo
> for these?

Yeah, Facets/Calibre should be in a separate repo.

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Sun Mar 19 01:57:19 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 18 Mar 2006 22:57:19 -0800
Subject: [Nitro] Where is the up to date repo?
In-Reply-To: <afd80b540603180205ue6827e7wa6f9d6beab6996d1@mail.gmail.com>
References: <afd80b540603180205ue6827e7wa6f9d6beab6996d1@mail.gmail.com>
Message-ID: <bd93db0d0603182257k53b86c27ja2335ba74ec6f04f@mail.gmail.com>

On 3/18/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> are allpatches applied to the devlab repo?
>

Hey George,

Not all. Some of the bigger changes are stored at
http://devlab.oree.ch/~bryan/repo
http://devlab.oree.ch/darcs/nitrohq has the more stable. I have a few
outstanding patches. Hopefully by the time you read this, I'll have a
few more additions.

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From george.moschovitis at gmail.com  Sun Mar 19 05:03:36 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Sun, 19 Mar 2006 12:03:36 +0200
Subject: [Nitro] Where is the up to date repo?
In-Reply-To: <bd93db0d0603182257k53b86c27ja2335ba74ec6f04f@mail.gmail.com>
References: <afd80b540603180205ue6827e7wa6f9d6beab6996d1@mail.gmail.com>
	<bd93db0d0603182257k53b86c27ja2335ba74ec6f04f@mail.gmail.com>
Message-ID: <afd80b540603190203g539935ch3124bb0f1aae4b2b@mail.gmail.com>

Saddly I get errors using both repositories :(
I will probably manually integrate the patches to my repo...

-g.


On 3/19/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/18/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > are allpatches applied to the devlab repo?
> >
>
> Hey George,
>
> Not all. Some of the bigger changes are stored at
> http://devlab.oree.ch/~bryan/repo
> http://devlab.oree.ch/darcs/nitrohq has the more stable. I have a few
> outstanding patches. Hopefully by the time you read this, I'll have a
> few more additions.
>
> Bryan
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From kashia at vfemail.net  Sun Mar 19 07:47:53 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sun, 19 Mar 2006 13:47:53 +0100
Subject: [Nitro] [PATCH] make Nitro work with facets 1.1.0
Message-ID: <op.s6nu53cwg3glhz@jo.local>

Hi, since zimba advertised his new facets I absolutely wanted those fuctions.
Here's a patch to make Nitro work with the new facets and calibre.

I can't verify the errors come from this change though, I'd like other people
to test it first before integrating.

Tests:

Nitro:

Finished in 94.812502 seconds.

   1) Failure:
test_all(TC_Element) [./test/nitro/tc_element.rb:80]:
<"<x:box color=\"#ff0\">\n    xhtml mode\n  </x:box>  \n"> expected to be =~
</div style/>.

34 tests, 117 assertions, 1 failures, 0 errors

Glue:

Finished in 0.408187 seconds.

   1) Failure:
test_manage_entities(TC_MixinsTest) [./test/glue/tc_property_mixins.rb:55]:
<Pure>
of type <Class>
expected to respond_to?<:properties>.

   2) Failure:
test_managing_mixins(TC_MixinsTest) [./test/glue/tc_property_mixins.rb:64]:
<Mixin>
of type <Module>
expected to respond_to?<:properties>.

   3) Error:
test_macro_params(Test::TC_N_Properties):
NoMethodError: undefined method `properties' for Test::Msg:Class
     /Volumes/Data/build/cvs/nitro/og/lib/og/entity.rb:477:in `method_missing'
     ./test/glue/tc_property.rb:100:in `test_macro_params'

   4) Error:
test_props(Test::TC_N_Properties):
NoMethodError: undefined method `properties' for Test::SubMsg:Class
     /Volumes/Data/build/cvs/nitro/og/lib/og/entity.rb:477:in `method_missing'
     ./test/glue/tc_property.rb:78:in `test_props'

36 tests, 160 assertions, 2 failures, 2 errors

Sadly can't run Og test atm.


-- 
Feel the love
http://pinkjuice.com/pics/ruby.png
-------------- next part --------------
A non-text attachment was scrubbed...
Name: facets_one_one_oh.bz2
Type: application/bzip2
Size: 10736 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060319/24920be3/attachment.bin 

From bryan.a.soto at gmail.com  Sun Mar 19 12:25:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sun, 19 Mar 2006 09:25:23 -0800
Subject: [Nitro] [PATCH] make Nitro work with facets 1.1.0
In-Reply-To: <op.s6nu53cwg3glhz@jo.local>
References: <op.s6nu53cwg3glhz@jo.local>
Message-ID: <bd93db0d0603190925nc86564dl2d6db2583fd4bae4@mail.gmail.com>

On 3/19/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hi, since zimba advertised his new facets I absolutely wanted those fuctions.
> Here's a patch to make Nitro work with the new facets and calibre.
>
> I can't verify the errors come from this change though, I'd like other people
> to test it first before integrating.
>
> Tests:
>
> Nitro:
>
> Finished in 94.812502 seconds.
>
>    1) Failure:
> test_all(TC_Element) [./test/nitro/tc_element.rb:80]:
> <"<x:box color=\"#ff0\">\n    xhtml mode\n  </x:box>  \n"> expected to be =~
> </div style/>.
>
> 34 tests, 117 assertions, 1 failures, 0 errors
>

That one was there prior. I believe it's been failing since 0.29.0.

> Glue:
>
> Finished in 0.408187 seconds.
>
>    1) Failure:
> test_manage_entities(TC_MixinsTest) [./test/glue/tc_property_mixins.rb:55]:
> <Pure>
> of type <Class>
> expected to respond_to?<:properties>.
>
>    2) Failure:
> test_managing_mixins(TC_MixinsTest) [./test/glue/tc_property_mixins.rb:64]:
> <Mixin>
> of type <Module>
> expected to respond_to?<:properties>.
>
>    3) Error:
> test_macro_params(Test::TC_N_Properties):
> NoMethodError: undefined method `properties' for Test::Msg:Class
>      /Volumes/Data/build/cvs/nitro/og/lib/og/entity.rb:477:in `method_missing'
>      ./test/glue/tc_property.rb:100:in `test_macro_params'
>
>    4) Error:
> test_props(Test::TC_N_Properties):
> NoMethodError: undefined method `properties' for Test::SubMsg:Class
>      /Volumes/Data/build/cvs/nitro/og/lib/og/entity.rb:477:in `method_missing'
>      ./test/glue/tc_property.rb:78:in `test_props'
>
> 36 tests, 160 assertions, 2 failures, 2 errors
>

Results from the properties to og_propertes patch. They've also been there.

> Sadly can't run Og test atm.
>

Just unpull the patch from me that added the two tc_camel_case* tests.
Or rename the files to xtc* or something like that. That's the only
reason I'm aware of the tests aren't running. Good tests by the way.
:)

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Sun Mar 19 12:45:53 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 19 Mar 2006 17:45:53 +0000
Subject: [Nitro] [PATCH] make Nitro work with facets 1.1.0
In-Reply-To: <bd93db0d0603190925nc86564dl2d6db2583fd4bae4@mail.gmail.com>
References: <op.s6nu53cwg3glhz@jo.local>
	<bd93db0d0603190925nc86564dl2d6db2583fd4bae4@mail.gmail.com>
Message-ID: <4b6f054f0603190945j6e8ba879r1fd77f5c2655f83a@mail.gmail.com>

Nitro will need to be updated to use calibre now too. Basically change
nitro to depend on facets 1.1.0 and calibre 1.2.0. Then install
Calibre and run tests. Wherever a lib can't be found that was
'facet/...' replace with 'calibre/...'

T.



From kashia at vfemail.net  Sun Mar 19 13:37:43 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sun, 19 Mar 2006 19:37:43 +0100
Subject: [Nitro] [PATCH] make Nitro work with facets 1.1.0
In-Reply-To: <4b6f054f0603190945j6e8ba879r1fd77f5c2655f83a@mail.gmail.com>
References: <op.s6nu53cwg3glhz@jo.local>
	<bd93db0d0603190925nc86564dl2d6db2583fd4bae4@mail.gmail.com>
	<4b6f054f0603190945j6e8ba879r1fd77f5c2655f83a@mail.gmail.com>
Message-ID: <op.s6obc5v2g3glhz@jo.local>

Hi,

> Nitro will need to be updated to use calibre now too. Basically change
> nitro to depend on facets 1.1.0 and calibre 1.2.0. Then install
> Calibre and run tests. Wherever a lib can't be found that was
> 'facet/...' replace with 'calibre/...'

Um.... did you read what I wrote earlier?

"....patch to make Nitro work with the new facets and calibre." :)

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From transfire at gmail.com  Sun Mar 19 13:43:51 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 19 Mar 2006 18:43:51 +0000
Subject: [Nitro] [PATCH] make Nitro work with facets 1.1.0
In-Reply-To: <op.s6obc5v2g3glhz@jo.local>
References: <op.s6nu53cwg3glhz@jo.local>
	<bd93db0d0603190925nc86564dl2d6db2583fd4bae4@mail.gmail.com>
	<4b6f054f0603190945j6e8ba879r1fd77f5c2655f83a@mail.gmail.com>
	<op.s6obc5v2g3glhz@jo.local>
Message-ID: <4b6f054f0603191043k7cdd5699wfecb2985fd9dfaee@mail.gmail.com>

> Um.... did you read what I wrote earlier?
>
> "....patch to make Nitro work with the new facets and calibre." :)

Oops! Missed the calibre there.

Thanks for doing this, btw.

T.



From kashia at vfemail.net  Sun Mar 19 13:44:11 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Sun, 19 Mar 2006 19:44:11 +0100
Subject: [Nitro] [PATCH] make Nitro work with facets 1.1.0
In-Reply-To: <bd93db0d0603190925nc86564dl2d6db2583fd4bae4@mail.gmail.com>
References: <op.s6nu53cwg3glhz@jo.local>
	<bd93db0d0603190925nc86564dl2d6db2583fd4bae4@mail.gmail.com>
Message-ID: <op.s6obnxbig3glhz@jo.local>

Hi,

> Just unpull the patch from me that added the two tc_camel_case* tests.
> Or rename the files to xtc* or something like that. That's the only
> reason I'm aware of the tests aren't running. Good tests by the way.

nope, it's not Ogs fault right now, just my CONFIG.rb. Some tests error
out with sqlite as second store and I don't want to change it right now.

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From transfire at gmail.com  Mon Mar 20 16:00:47 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 20 Mar 2006 21:00:47 +0000
Subject: [Nitro] Using Nitro
Message-ID: <4b6f054f0603201300n73095ae7r1a4bed6c8b735af1@mail.gmail.com>

I am ready to build my web project. And obviously, being a contributor
to Nitro, I want to use Nitro. But I have three serious questions. 1)
Is it robust enough? 2) Does it scale? and 3) Where the heck is the
documentation?

On the first point, what's the general opinion here. I know there are
working sites out there. How well are they holding up? And how hard
was it to get them up? I sometimes get the feeling that people become
interested in Nitro/Og but end up going else where b/c of reasons #1
and #3. What do you think?

The application I plan will need to scale big time. So that's a big
consideration. There's been a lot of talk about this with regards to
Rails with mixed opinions. I'm not sure how to appraoch this.

On point #3 I think it's about time for Developers to become 80% Documentors.

T.



From transfire at gmail.com  Mon Mar 20 16:10:28 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 20 Mar 2006 21:10:28 +0000
Subject: [Nitro] What sets Nitro apart from Rails?
Message-ID: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>

Figure while I'm asking this is also a very good question: What sets
Nitro apart from Rails? I think we all know that Og allows one to
build the database structure in Ruy rather then SQL. That's one good
point. But what about beyond this? How is Nitro better than Rails
rather then just equal to it?

Thanks,
T.



From james_b at neurogami.com  Mon Mar 20 17:10:10 2006
From: james_b at neurogami.com (James Britt)
Date: Mon, 20 Mar 2006 15:10:10 -0700
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
Message-ID: <441F2842.8000305@neurogami.com>

TRANS wrote:
> Figure while I'm asking this is also a very good question: What sets
> Nitro apart from Rails? I think we all know that Og allows one to
> build the database structure in Ruy rather then SQL. That's one good
> point. But what about beyond this? How is Nitro better than Rails
> rather then just equal to it?

Better?  That's a loaded question.

They're different.  I've done a few Rails apps, I've done a few Nitro 
apps.   When the application is focused on data, and the data models are 
pretty well-defined in advance, Rails seems to work better for me. (Or 
if there is a ready-made library or something that I can just drop in to 
get the job done.)

If the application ins more about behavior, or I'm migrating a static 
site to a dynamic site, or the data model is evolving as I work out the 
application, Og/Nitro works better.

Og/Nitro works better for me, too, when the data model is simple.

My knowledge of Nitro templating is skimpy, and I've had to hack my own 
variations of skins and such to get what I wanted, so Rails may have an 
upper hand there in some regards, though Rails partials to be somewhat 
arbitrary and poorly thought out.  But that may just be a mismatch 
between how I think of templating and what The Rails team prefers.

As a a third option, I'm increasingly drawn to the idea of growing my 
own framework.  In the time it takes me to figure out the quirks of one 
framework or another I can likely write my own, one that maps to my way 
of thinking.  Towards that end, Og/Nitro may be a bigger help because it 
is less of a monolithic tool set; it seems to  have fewer presumptions 
built in, so it may be easier to pick and choose pieces for one's own 
custom framework.  I may end up using Of stuff for persistence and hack 
my own dispatcher/templating stuff.

Related side note: I poked around with SwitchTower, now called 
Capistrano, for deploying a Rails app, and it *almost* worked (it did 
something weird with log file permissions, making them unwritable).

I suppose for deploying Rails applications it would make sense to get to 
understand the tool, and leverage the community effort in writing 
add-ins and such.

But then  I tried to use it to deploy a Nitro app, and got nowhere.  I 
know (well, suspect) that it can be done, but good luck finding docs for 
this.  Do I invest time in figuring this out, or do I spend that same 
time just assembling my own set of Rake tasks and domain-specific slang 
to get a tool I readily understand, does exactly what I want, and can 
use across frameworks?   I'm going for the latter.



-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://web2.0validator.com    - We're the Dot in Web 2.0


From james_b at neurogami.com  Mon Mar 20 17:23:02 2006
From: james_b at neurogami.com (James Britt)
Date: Mon, 20 Mar 2006 15:23:02 -0700
Subject: [Nitro] Using Nitro
In-Reply-To: <4b6f054f0603201300n73095ae7r1a4bed6c8b735af1@mail.gmail.com>
References: <4b6f054f0603201300n73095ae7r1a4bed6c8b735af1@mail.gmail.com>
Message-ID: <441F2B46.7070009@neurogami.com>

TRANS wrote:
> I am ready to build my web project. And obviously, being a contributor
> to Nitro, I want to use Nitro. But I have three serious questions. 1)
> Is it robust enough? 2) Does it scale? and 3) Where the heck is the
> documentation?
> 
> On the first point, what's the general opinion here. I know there are
> working sites out there. How well are they holding up? And how hard
> was it to get them up? I sometimes get the feeling that people become
> interested in Nitro/Og but end up going else where b/c of reasons #1
> and #3. What do you think?

Good questions.  I picked Nitro for certain apps because it made it 
easier to grow the code from a set of rough ideas.  So I could sketch 
out code, test, change stuff, and work out the behavior.  I wasn't 
bouncing around with SQL or PhpMyAdmin or some funky not-quite-sql Ruby 
DSL thing.  I wrote application code without the tools getting in my way.

But I confess that there were times (as evidenced on this list) when I 
just could not figure out how to do something The Nitro Way.  Routing 
and templating being the main items.

Rails has assorted niceties, such as the bundled testing.  I imagine 
these sorts of things will find their way into Nitro (or be available as 
add-ons; I like that Nitro is more lightweight than Rails).  But not 
yet.  And the available Rails documentation, while grossly inadequate 
once you move  away from basic tasks, is still better that Nitro's.

> 
> The application I plan will need to scale big time. So that's a big
> consideration. There's been a lot of talk about this with regards to
> Rails with mixed opinions. I'm not sure how to appraoch this.

Well, how different, under the hood, are Rails and Nitro?  Assuming you 
use fcgi or scgi, they should scale about the same, unless one is doing 
something funky with resources.

I've not seen any differences in resource consumption between my Rails 
and Nitro apps, but they're not all running under the same 
circumstances, so it's hard to tell.


> 
> On point #3 I think it's about time for Developers to become 80% Documentors.

Oh yes, but it's a tough demand.


-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.jamesbritt.com     - Playing with Better Toys
http://www.30secondrule.com   - Building Better Tools


From kgergely at mcl.hu  Mon Mar 20 18:23:08 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Tue, 21 Mar 2006 00:23:08 +0100
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <bd93db0d0603141043r5860ce1dj36a29eb8a17cb243@mail.gmail.com>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
	<bd93db0d0603131551g1264a627nfc681e11872d70b0@mail.gmail.com>
	<01013e17fc3eb92bf54199e591f041b1@oggu.de>
	<bd93db0d0603141043r5860ce1dj36a29eb8a17cb243@mail.gmail.com>
Message-ID: <20060320232308.GA10859@mlabdial.hit.bme.hu>

Bryan Soto azt ?rta, hogy: 
> First off though, let's make sure it actually helped. ;)

Errr, sorry for the delay, from my next post you can figure it out what 
I did in the past few days :-/

Yes, your comment was useful, the shebang line of the distro was wrong 
(it was hardcoded to /usr/local/bin/ruby).

However, the last error was: I had to open up the port, which should go 
to a separate file in apache2 (the ListenPort directive).

And thanks everyone for the active help

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060320/60f4a932/attachment.bin 

From kgergely at mcl.hu  Mon Mar 20 18:29:00 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Tue, 21 Mar 2006 00:29:00 +0100
Subject: [Nitro] how to tell my boss to use nitro?
Message-ID: <20060320232900.GB10859@mlabdial.hit.bme.hu>

Hi!

Switching to nitro at my company is not easy, because I'm not a boss.
Despite of having very serious problems and stucks with PHP, it is very 
hard to persuade him to begin developing in Nitro.
The facts, and things he said:
1) Now pure php is used for web developement
2) We can use CakePHP, which is a Rails clone!
3) He said: everyone knows PHP, but who knows ruby? If you can't 
   continue developing, what will I do.
   However, he knows, that yes, ruby is a cool language and has a very 
   good learning curve.
4) "Show me a *COMPLEX* site, which uses nitro!"
5) "php+apache+mysql is VERY fast"
6) "cgi is slow"

thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060320/9b247a48/attachment.bin 

From bryan.a.soto at gmail.com  Mon Mar 20 21:40:20 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 20 Mar 2006 18:40:20 -0800
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <01013e17fc3eb92bf54199e591f041b1@oggu.de>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
	<bd93db0d0603131551g1264a627nfc681e11872d70b0@mail.gmail.com>
	<01013e17fc3eb92bf54199e591f041b1@oggu.de>
Message-ID: <bd93db0d0603201840p149c2349vd2ab1efcbd01a6a8@mail.gmail.com>

On 3/14/06, Fabian Buch <fabian at oggu.de> wrote:
> could you copy&paste your answer below also as answer to
> http://oxyliquit.de/question/16 ? That would be very cool, since it
> might be helpful for future users.
>

Hi Fabian,

I was going to submit my answer, but I don't think it quite fits the
question. Perhaps something like:

<paste author="gergely">
I'm trying to run nitro under Apache2 with fcgid.

I'm almost there, the server is running, but I get 500 response, and my
log only says:

[Sun Mar 12 20:28:16 2006] [error] [client 192.168.1.1] Premature end
of script headers: fcgi.rb

With webrick it worked. So where to hunt for errors?
</paste>

would be a better question?

Thanks,

Bryan

PS: I still want to submit it. I need the points... :(

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Mon Mar 20 21:41:20 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 21 Mar 2006 02:41:20 +0000
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <441F2842.8000305@neurogami.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
	<441F2842.8000305@neurogami.com>
Message-ID: <4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>

Thanks James. That helps a great deal.

Well, if I dig in to Nitro I will document a whole bunch --maybe even
enough to be the bases of a book, if others are willing to help edit
and flesh it out. I don't find it too difficult to document myself.
But I admit I'm hesitant b/c as I dig into this I know I will want to
improve/fix things. Which makes me wonder, like you, that maybe I'm
better off with a DIY approach. But I'm willing to dig in if I can
take on a sort of quality control position and get some support for
moving forward. I'm just a perfectionist like that, and if I'm going
to take the trouble of leaning this think pretty much thru-and-thru
and document the bugger, then I want to make it worth my efforts. Know
what I mean?

IN fact, to give a taste of what I mean, right off the bat we have this:

  Attempting local installation of 'nitro'
  Local gem file not found: nitro*.gem
  Attempting remote installation of 'nitro'
  Updating Gem source index for: http://gems.rubyforge.org
  Install required dependency og? [Yn]  Y
  Install required dependency glue? [Yn]  Y
  Install required dependency cmdparse? [Yn]  Y
  Install required dependency gen? [Yn]  Y
  Install required dependency glue? [Yn]  Y
  Install required dependency RedCloth? [Yn]  Y
  Install required dependency ruby-breakpoint? [Yn]  Y
  Install required dependency daemons? [Yn]  Y
  Install required dependency facets? [Yn]  Y
  Successfully installed nitro-0.29.0
  Successfully installed og-0.29.0
  Successfully installed glue-0.29.0
  Successfully installed cmdparse-2.0.0
  Successfully installed gen-0.29.0
  Successfully installed glue-0.29.0
  Successfully installed RedCloth-3.0.3
  Successfully installed ruby-breakpoint-0.5.0
  Successfully installed daemons-0.4.2
  Successfully installed facets-1.0.3
  Installing RDoc documentation for nitro-0.29.0...

  lib/nitro/part.rb:13:40: Skipping require of dynamic string:
"part/#{name}/run.rb"
  Installing RDoc documentation for og-0.29.0...
  Installing RDoc documentation for glue-0.29.0...
  Installing RDoc documentation for cmdparse-2.0.0...
  Installing RDoc documentation for gen-0.29.0...
  Installing RDoc documentation for glue-0.29.0...
  Installing RDoc documentation for ruby-breakpoint-0.5.0...
  Installing RDoc documentation for daemons-0.4.2...
  Installing RDoc documentation for facets-1.0.3...

Let start with glue. Now besides the fact that it's being asked for
twice, the more important fact is that we don;t even need it. Please,
let's get rid of glue altogether. If a glue lib is generic it can go
into calibre. Otherwise put it in nitro itself and be done with it.

Secondly, 'gen' doesn't need to be a seprate project. Maybe one day it
will be worth having as such, but not today. Put it in nitro too.

We don't need cmdparse, calibre has consoleapp.rb which works very
well (and if it lacks anything then fix it or tell me and I will).

Are Redhat, ruby-breakpoint and daemons really _required_? Or can they
be optional? If optional make them optional. It would be nice if we
could print a  list of optional recommended libraries to this process.

Then there is that dynamic require warning. We need to do something
about that but I'm not sure what at the moment. Can we add a :nodoc:
option to that line?

Finally lets compare this too Rails:

  Attempting local installation of 'rails'
  Local gem file not found: rails*.gem
  Attempting remote installation of 'rails'
  Updating Gem source index for: http://gems.rubyforge.org
  Install required dependency rake? [Yn]  Y
  Install required dependency activesupport? [Yn]  Y
  Install required dependency activerecord? [Yn]  Y
  Install required dependency actionpack? [Yn]  Y
  Install required dependency actionmailer? [Yn]  Y
  Install required dependency actionwebservice? [Yn]  Y
  Successfully installed rails-1.0.0
  Successfully installed rake-0.7.0
  Successfully installed activesupport-1.2.5
  Successfully installed activerecord-1.13.2
  Successfully installed actionpack-1.11.2
  Successfully installed actionmailer-1.1.5
  Successfully installed actionwebservice-1.0.0
  Installing RDoc documentation for rake-0.7.0...
  Installing RDoc documentation for activesupport-1.2.5...
  Installing RDoc documentation for activerecord-1.13.2...
  Installing RDoc documentation for actionpack-1.11.2...
  Installing RDoc documentation for actionmailer-1.1.5...
  Installing RDoc documentation for actionwebservice-1.0.0...

Notice how nicely this comes across? It may seem silly but this is the
kind of little thing that helps sell rails. I think nitro can do even
better. Taking the advice I propose, and assuming I'm correct about
the optional libs, we get (I've added calibre as of this release):

  Attempting local installation of 'nitro'
  Local gem file not found: nitro*.gem
  Attempting remote installation of 'nitro'
  Updating Gem source index for: http://gems.rubyforge.org
  Install required dependency og? [Yn]  Y
  Install required dependency facets? [Yn]  Y
  Install required dependency calibre? [Yn]  Y
  Successfully installed nitro-0.29.0
  Successfully installed og-0.29.0
  Successfully installed facets-1.1.0
  Successfully installed calibre-1.2.0
  Installing RDoc documentation for nitro-0.29.0...
  Installing RDoc documentation for og-0.29.0...
  Installing RDoc documentation for facets-1.1.0...
  Installing RDoc documentation for calibre-1.2.0...

Which is much nicer.

T.



From kgergely at mcl.hu  Mon Mar 20 22:06:19 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Tue, 21 Mar 2006 04:06:19 +0100
Subject: [Nitro] easy setup? (no mod_rewrite)
Message-ID: <20060321030619.GC11800@mlabdial.hit.bme.hu>

Hi!

Is it at all planned, that I can set up a nitro app in say:

www.my-shared-host.com/some/directories

?

thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060320/6fd1c4f1/attachment.bin 

From james_b at neurogami.com  Tue Mar 21 01:56:53 2006
From: james_b at neurogami.com (James Britt)
Date: Mon, 20 Mar 2006 23:56:53 -0700
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>	<441F2842.8000305@neurogami.com>
	<4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>
Message-ID: <441FA3B5.7010702@neurogami.com>

TRANS wrote:
> Thanks James. That helps a great deal.

Good, I'm glad.

> 
> Well, if I dig in to Nitro I will document a whole bunch --maybe even
> enough to be the bases of a book, if others are willing to help edit
> and flesh it out. I don't find it too difficult to document myself.

Same here.  I started on some documentation, but the code base kept 
changing, and it got too frustrating.  But I'm taking another stab at it 
now.

The frequent lib name and API changes made it hard to document, and hard 
for people to pick up on Og/Nitro.  Stability is important for growth. 
(But I also appreciate that it is better to evolve and break APIs as 
fast as possible, before you have too large of a user base and scads of 
legacy code to worry about.)


> But I admit I'm hesitant b/c as I dig into this I know I will want to
> improve/fix things. Which makes me wonder, like you, that maybe I'm
> better off with a DIY approach. But I'm willing to dig in if I can
> take on a sort of quality control position and get some support for
> moving forward. I'm just a perfectionist like that, and if I'm going
> to take the trouble of leaning this think pretty much thru-and-thru
> and document the bugger, then I want to make it worth my efforts. Know
> what I mean?

Yes, but something to consider is whether you can influence the 
direction of Nitro, and if Nitro will be open/malleable enough that one 
can easily add on or alter it for your purposes.

For example, I have no ambitions to recreated Og.  And the main parts of 
Nitro are probably fine for me.  If I can a) easily find out how to do 
Nitro stuff, and b) adapt the code to do things a  particular way, then 
building from scratch is not worthwhile.  I also believe George is open 
to reasonable suggestions.



> 
> IN fact, to give a taste of what I mean, right off the bat we have this:
...

> 
> Let start with glue. Now besides the fact that it's being asked for
> twice, the more important fact is that we don;t even need it. Please,
> let's get rid of glue altogether. If a glue lib is generic it can go
> into calibre. Otherwise put it in nitro itself and be done with it.
> 
> Secondly, 'gen' doesn't need to be a seprate project. Maybe one day it
> will be worth having as such, but not today. Put it in nitro too.
> 

What's goofy is that I hacked gen to allow me to create basic controller 
code, then a little later I  upgraded Nitro.  I got a new version of 
gen, though as far as I could tell it was the same code as the previous 
version, but now my hacks didn't run because rubygems only looked in the 
new gen directory.

> We don't need cmdparse, calibre has consoleapp.rb which works very
> well (and if it lacks anything then fix it or tell me and I will).
> 
> Are Redhat, ruby-breakpoint and daemons really _required_? Or can they
> be optional? If optional make them optional. It would be nice if we
> could print a  list of optional recommended libraries to this process.

One of the things that irked me about Rails was the laundry list of 
required gems, things I did not plan on using.  ActionMailer?  Why must 
I install that if I'm not sending mail?  ActionWebStuff? Please let me 
install it when I need it.

I'd like to see a Nitro base install that is pretty minimal, and docs 
that explain what add-ons are available for various tasks.

> 
> Then there is that dynamic require warning. We need to do something
> about that but I'm not sure what at the moment. Can we add a :nodoc:
> option to that line?
> 
> Finally lets compare this too Rails:
> 
>   Attempting local installation of 'rails'
>   Local gem file not found: rails*.gem
>   Attempting remote installation of 'rails'
>   Updating Gem source index for: http://gems.rubyforge.org
>   Install required dependency rake? [Yn]  Y
>   Install required dependency activesupport? [Yn]  Y
>   Install required dependency activerecord? [Yn]  Y
>   Install required dependency actionpack? [Yn]  Y
>   Install required dependency actionmailer? [Yn]  Y
>   Install required dependency actionwebservice? [Yn]  Y
>   Successfully installed rails-1.0.0
>   Successfully installed rake-0.7.0
>   Successfully installed activesupport-1.2.5
>   Successfully installed activerecord-1.13.2
>   Successfully installed actionpack-1.11.2
>   Successfully installed actionmailer-1.1.5
>   Successfully installed actionwebservice-1.0.0
>   Installing RDoc documentation for rake-0.7.0...
>   Installing RDoc documentation for activesupport-1.2.5...
>   Installing RDoc documentation for activerecord-1.13.2...
>   Installing RDoc documentation for actionpack-1.11.2...
>   Installing RDoc documentation for actionmailer-1.1.5...
>   Installing RDoc documentation for actionwebservice-1.0.0...
> 
> Notice how nicely this comes across? It may seem silly but this is the
> kind of little thing that helps sell rails. 

Well, I used to have issues with installations and upgrades.  But one 
thing the Rails team have nailed is packaging and presentation.  (There 
is still a bit too much alchemic knowledge required for certain tasks, 
but overall they've made basic things easy.)


> I think nitro can do even
> better. Taking the advice I propose, and assuming I'm correct about
> the optional libs, we get (I've added calibre as of this release):
> 
>   Attempting local installation of 'nitro'
>   Local gem file not found: nitro*.gem
>   Attempting remote installation of 'nitro'
>   Updating Gem source index for: http://gems.rubyforge.org
>   Install required dependency og? [Yn]  Y
>   Install required dependency facets? [Yn]  Y
>   Install required dependency calibre? [Yn]  Y
>   Successfully installed nitro-0.29.0
>   Successfully installed og-0.29.0
>   Successfully installed facets-1.1.0
>   Successfully installed calibre-1.2.0
>   Installing RDoc documentation for nitro-0.29.0...
>   Installing RDoc documentation for og-0.29.0...
>   Installing RDoc documentation for facets-1.1.0...
>   Installing RDoc documentation for calibre-1.2.0...
> 
> Which is much nicer.

Very.


-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://yourelevatorpitch.com  - Finding Business Focus


From timh at dirtymonday.net  Tue Mar 21 02:46:51 2006
From: timh at dirtymonday.net (TimH)
Date: Mon, 20 Mar 2006 23:46:51 -0800
Subject: [Nitro] Dots in nitro page names
Message-ID: <20060320234651.53af2697.timh@dirtymonday.net>

I named a new Nitro page something like "My run.rb" and it broke that page.  I wasn't even able to delete it.  I tried a few more pages with dots and they all failed (displayed blank page, no html output).  Is this a known thing?  Would there be an easy fix?

--TimH


From m.fellinger at gmail.com  Tue Mar 21 03:15:02 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Tue, 21 Mar 2006 17:15:02 +0900
Subject: [Nitro] how to tell my boss to use nitro?
In-Reply-To: <20060320232900.GB10859@mlabdial.hit.bme.hu>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
Message-ID: <9c00d3e00603210015n647a10eag38e3419bd678588a@mail.gmail.com>

On 3/21/06, Gergely Kontra <kgergely at mcl.hu> wrote:> Hi!>> Switching to nitro at my company is not easy, because I'm not a boss.> Despite of having very serious problems and stucks with PHP, it is very> hard to persuade him to begin developing in Nitro.> The facts, and things he said:> 1) Now pure php is used for web developementPHP... how glad I am to be rid of it (tho i don't know for how long)
> 2) We can use CakePHP, which is a Rails clone!Never heard about it... but it definitly cannot become as good asrails since rails is rails because of ruby - and the same is valid fornitro
> 3) He said: everyone knows PHP, but who knows ruby? If you can't>    continue developing, what will I do.>    However, he knows, that yes, ruby is a cool language and has a very>    good learning curve.Guess you don't have to argue with him then, maybe show him how niceOOP works with ruby...
> 4) "Show me a *COMPLEX* site, which uses nitro!"could you define complex? however, one page that is cited most timesis joy.gr and joyful.gr - there are new pages in the club, but most ofthe are not collected in a list afaik... :|
> 5) "php+apache+mysql is VERY fast"> 6) "cgi is slow"ruby+fcgi+lighttpd+psql is OMFGTFTKYEO (omfg - too fast to keep youreyes open :)>> thx> Gergo> --> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |> | Budapest University of Technology and Economics        ICQ# 175564914 |> +------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+>>> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>>>


From m.fellinger at gmail.com  Tue Mar 21 03:17:05 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Tue, 21 Mar 2006 17:17:05 +0900
Subject: [Nitro] easy setup? (no mod_rewrite)
In-Reply-To: <20060321030619.GC11800@mlabdial.hit.bme.hu>
References: <20060321030619.GC11800@mlabdial.hit.bme.hu>
Message-ID: <9c00d3e00603210017x39a43278j84c5e633c99356b5@mail.gmail.com>

Could you specify that a bit?What kind of deploying do you want? webrick/mongrel/cgi/fcgi/scgi -apache/lighttpd ?
On 3/21/06, Gergely Kontra <kgergely at mcl.hu> wrote:> Hi!>> Is it at all planned, that I can set up a nitro app in say:>> www.my-shared-host.com/some/directories>> ?>> thx> Gergo> --> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |> | Budapest University of Technology and Economics        ICQ# 175564914 |> +------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+>>> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>>>


From m.fellinger at gmail.com  Tue Mar 21 03:26:11 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Tue, 21 Mar 2006 17:26:11 +0900
Subject: [Nitro] Dots in nitro page names
In-Reply-To: <20060320234651.53af2697.timh@dirtymonday.net>
References: <20060320234651.53af2697.timh@dirtymonday.net>
Message-ID: <9c00d3e00603210026p29a7713cx5ae70159f0c58c45@mail.gmail.com>

you probably should add that you tried that in spark :)
On 3/21/06, TimH <timh at dirtymonday.net> wrote:> I named a new Nitro page something like "My run.rb" and it broke that page.  I wasn't even able to delete it.  I tried a few more pages with dots and they all failed (displayed blank page, no html output).  Is this a known thing?  Would there be an easy fix?>> --TimH> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From kgergely at mcl.hu  Tue Mar 21 09:14:36 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Tue, 21 Mar 2006 15:14:36 +0100
Subject: [Nitro] easy setup? (no mod_rewrite)
In-Reply-To: <9c00d3e00603210017x39a43278j84c5e633c99356b5@mail.gmail.com>
References: <20060321030619.GC11800@mlabdial.hit.bme.hu>
	<9c00d3e00603210017x39a43278j84c5e633c99356b5@mail.gmail.com>
Message-ID: <20060321141436.GB16859@mlabdial.hit.bme.hu>

Michael Fellinger azt ?rta, hogy: 
> Could you specify that a bit?What kind of deploying do you want? 
> webrick/mongrel/cgi/fcgi/scgi -apache/lighttpd ?

Sorry. So: if I have apache2 webserver, and can run ruby scripts?
I dunno if I have apache and cannot remove it, is it worth to fire up 
another web server, or stick to apache with nitro...

Eg. cake can operate like this:

www.site.com/my/directory/index.php/cake/path/begins/here

ps: your mail client suxx: it reformats the lines in a weird way.

thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060321/30f84a55/attachment.bin 

From timh at dirtymonday.net  Tue Mar 21 10:38:31 2006
From: timh at dirtymonday.net (TimH)
Date: Tue, 21 Mar 2006 07:38:31 -0800
Subject: [Nitro] Dots in nitro page names
In-Reply-To: <9c00d3e00603210026p29a7713cx5ae70159f0c58c45@mail.gmail.com>
References: <20060320234651.53af2697.timh@dirtymonday.net>
	<9c00d3e00603210026p29a7713cx5ae70159f0c58c45@mail.gmail.com>
Message-ID: <20060321073831.55c3dabe.timh@dirtymonday.net>

On Tue, 21 Mar 2006 17:26:11 +0900
"Michael Fellinger" <m.fellinger at gmail.com> wrote:

> you probably should add that you tried that in spark :)

	Yes, you are right.  It was late :P

	I was told that this issue was fixed in one of the repositories, but that that repository is unstable at the moment...  I guess I'll wait a bit.

--TimH


From transfire at gmail.com  Tue Mar 21 11:46:16 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 21 Mar 2006 16:46:16 +0000
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <441FA3B5.7010702@neurogami.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
	<441F2842.8000305@neurogami.com>
	<4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>
	<441FA3B5.7010702@neurogami.com>
Message-ID: <4b6f054f0603210846h6b635d30k42816d9e401b6b55@mail.gmail.com>

On 3/21/06, James Britt <james_b at neurogami.com> wrote:
> TRANS wrote:
> > Thanks James. That helps a great deal.
>
> Good, I'm glad.
>
> >
> > Well, if I dig in to Nitro I will document a whole bunch --maybe even
> > enough to be the bases of a book, if others are willing to help edit
> > and flesh it out. I don't find it too difficult to document myself.
>
> Same here.  I started on some documentation, but the code base kept
> changing, and it got too frustrating.  But I'm taking another stab at it
> now.
>
> The frequent lib name and API changes made it hard to document, and hard
> for people to pick up on Og/Nitro.  Stability is important for growth.
> (But I also appreciate that it is better to evolve and break APIs as
> fast as possible, before you have too large of a user base and scads of
> legacy code to worry about.)



>
> > But I admit I'm hesitant b/c as I dig into this I know I will want to
> > improve/fix things. Which makes me wonder, like you, that maybe I'm
> > better off with a DIY approach. But I'm willing to dig in if I can
> > take on a sort of quality control position and get some support for
> > moving forward. I'm just a perfectionist like that, and if I'm going
> > to take the trouble of leaning this think pretty much thru-and-thru
> > and document the bugger, then I want to make it worth my efforts. Know
> > what I mean?
>
> Yes, but something to consider is whether you can influence the
> direction of Nitro, and if Nitro will be open/malleable enough that one
> can easily add on or alter it for your purposes.
>
> For example, I have no ambitions to recreated Og.  And the main parts of
> Nitro are probably fine for me.  If I can a) easily find out how to do
> Nitro stuff, and b) adapt the code to do things a  particular way, then
> building from scratch is not worthwhile.  I also believe George is open
> to reasonable suggestions.

Thanks James. You've helped me a great deal. I've come to a conclusion.

Thanks,
T.



From kashia at vfemail.net  Tue Mar 21 12:07:55 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Tue, 21 Mar 2006 18:07:55 +0100
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
Message-ID: <op.s6rwjhi4g3glhz@jo.local>

Hmm.. a pity you already decided, I wanted to think a bit more about
this.

The first thing that came to my mind, after reading the question which
is better, Rails or Nitro, was: "Nitro smells better".

No really, after all the thinking, I'm still for that answer :)

> Figure while I'm asking this is also a very good question: What sets
> Nitro apart from Rails? I think we all know that Og allows one to
> build the database structure in Ruy rather then SQL. That's one good
> point. But what about beyond this? How is Nitro better than Rails
> rather then just equal to it?

If you want Nitro to scale, I think caching with "memcached" would be
the way to go, that means, it'd use the Og store less, which is maybe
a tad slow because of the database.

I just tested my Nitro FAQ with ab (apache bench) using 400 connects,
40 concurrent connections, with a result of 12 requests per second.
This is on a crappy virtual server.

(I actually killed Postgres with that, "too many open files"), lesson
learned, when I get more page views on Oxy I better get a better
server ;)

For what sets Nitro apart from Rails, I don't really know, my reason
would be "freedom" :)

Kashia


-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From fabian at oggu.de  Tue Mar 21 12:46:36 2006
From: fabian at oggu.de (Fabian Buch)
Date: Tue, 21 Mar 2006 18:46:36 +0100
Subject: [Nitro] apache2 + fcgid on debian - almost there (how to debug)
In-Reply-To: <bd93db0d0603201840p149c2349vd2ab1efcbd01a6a8@mail.gmail.com>
References: <20060312194152.GA2787@mlabdial.hit.bme.hu>
	<bd93db0d0603131551g1264a627nfc681e11872d70b0@mail.gmail.com>
	<01013e17fc3eb92bf54199e591f041b1@oggu.de>
	<bd93db0d0603201840p149c2349vd2ab1efcbd01a6a8@mail.gmail.com>
Message-ID: <ebf3167aa0401381b66e29e5d37f06d0@oggu.de>


Am 21.03.2006 um 03:40 schrieb Bryan Soto:
> <paste author="gergely">
...
> </paste>
>
> would be a better question?

Good. Though it's <quote by="gergely">...</quote>  instead of <paste>. 
Should there be a <paste /> alias? There's a syntax help for Oxyliquit 
though (if you ask a question, just lick on "show help").

> PS: I still want to submit it. I need the points... :(

You applied it as your own question and answered it yourself. Now you 
can also rate it yourself (do so to set it as "answered"). Self-rating 
won't higher your stats though, sorry.

There's another rating mechanism in works for tips (dunno when it'll 
appear on the production site yet).

Fabian



From james_b at neurogami.com  Tue Mar 21 12:55:59 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 21 Mar 2006 10:55:59 -0700
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <4b6f054f0603210846h6b635d30k42816d9e401b6b55@mail.gmail.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>	<441F2842.8000305@neurogami.com>	<4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>	<441FA3B5.7010702@neurogami.com>
	<4b6f054f0603210846h6b635d30k42816d9e401b6b55@mail.gmail.com>
Message-ID: <44203E2F.7010002@neurogami.com>

TRANS wrote:

> 
> 
> Thanks James. You've helped me a great deal. I've come to a conclusion.


And .... ?




-- 
James Britt

"A language that doesn't affect the way you think about programming is 
not worth knowing."
   - A. Perlis


From thoran at fastmail.fm  Tue Mar 21 13:33:20 2006
From: thoran at fastmail.fm (thoran)
Date: Wed, 22 Mar 2006 04:33:20 +1000
Subject: [Nitro] Rails+Og, Og & MySQL, and Norm
Message-ID: <1142966000.5453.257175030@webmail.messagingengine.com>

Hi,

I've been playing with Og and trying to shoehorn it onto Rails because I
like Og better overall, and AR better particularly by the way in which
the schema diffs are afforded by Migrations.  

Am I missing something or do we need say, forward and back methods in
Og?  I like safe and I know I can be stupid enough to be production
nightmare if I'm not careful, so safeguards like that are good.  Or, is
my thinking still a little too database-centric, a little too Rails?  Is
it sufficient to version the whole app and deploy and undeploy the whole
thing?  That sort of makes more sense, since presumably one isn't simply
changing the persistence layer, but the relevant methods as well.  Does
anyone with a better idea wish to comment?  James?...  

Also, in fiddling with Og and a non-real database (MySQL) I've found Og
to be quite temperamental.  The output from irb is long.  If no-one
would mind most of 200 lines in here I can do that else I can send it
privately.  However the gist of it is that Og will sometimes load
properly and sometimes it won't.  And sometimes it will recreate the
tables, and sometimes it won't.  I have no explanation for this.  Is it
a known bug or is one of them non-obvious features?  I haven't tested
the same code on Postgres or sqlite or any other store.  Some of the
code I am using is from
http://www.infurious.com/blogs/index.php/aidan/2006/02/14/og_tutorial_part_one.
(See comment above about being dangerous.)  

With the integration effort of Og and Rails, I've been way too
adventurous and tried to implement a layer over both AR and Og, as they
are interchangeable both ways: Nitro+AR anyone?  So far I've managed to
get agnostic database connections working with both AR and Og (via
module Norm (Norm = Neutral ORM)), but I'm returning the persistence
layer specific object and not a more abstracted object.  How do I get a
subclass of whichever ORM I'm trying to use?  

I've also tried a more manual approach which roughly goes like:

module Norm
  def self.connect(...)
  end
end

class ARSnmpDevice < ActiveRecord::Base
end

class OgSnmpDevice
  property :id, Integer
  ...
  set_table :snmp_devices
  set_primary_key :id
end

class SnmpDevice < OgSnmpDevice
  include Norm
end

...

db = SnmpDevice.connect(...)
db.manage(SnmpDevice)
devices = SnmpDevice.all

Or something like that.  I realise its a piss-poor effort so far.  The
main problem is that I have to change the method calls manually where
they are different.  Ideally I would be able to use one set of calls.
Any ideas how to map these?  

And of course in the really funkified version AR would have oid's.  

Am I going about this completely the wrong way or is this just overall a
really bad idea?  



thoran


From james_b at neurogami.com  Tue Mar 21 14:10:37 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 21 Mar 2006 12:10:37 -0700
Subject: [Nitro] Rails+Og, Og & MySQL, and Norm
In-Reply-To: <1142966000.5453.257175030@webmail.messagingengine.com>
References: <1142966000.5453.257175030@webmail.messagingengine.com>
Message-ID: <44204FAD.2070602@neurogami.com>

thoran wrote:
> Hi,
> 
> I've been playing with Og and trying to shoehorn it onto Rails because I
> like Og better overall, and AR better particularly by the way in which
> the schema diffs are afforded by Migrations.  

Interesting.

> 
> Am I missing something or do we need say, forward and back methods in
> Og?  I like safe and I know I can be stupid enough to be production
> nightmare if I'm not careful, so safeguards like that are good.  Or, is
> my thinking still a little too database-centric, a little too Rails?  Is
> it sufficient to version the whole app and deploy and undeploy the whole
> thing?  That sort of makes more sense, since presumably one isn't simply
> changing the persistence layer, but the relevant methods as well.  Does
> anyone with a better idea wish to comment?  James?...  
>

Arguments I've heard regarding the Og approach is that you risk trashing 
data if you change the schema.  My understanding of AR+Migrations is 
that is helps reduce or prevent these problems.

My experience is that schema changes just don't cause problems for me, 
but working the AR way (where my model definitions are split across 
multiple files) does create day-to-day annoyances.

I've also run into problems getting Rails migration code to express my 
desired SQL, could find no docs to explain what I needed to do, and gave 
up.

So, as a practical matter, the Og approach makes my daily coding easier, 
and I'm unconvinced I need to carry an extra burden in my code on the 
off-chance that one day I'll need to migrate a schema and something will 
break.

In and when the day comes that I need a DB change that risks losing 
critical data, I believe that the effort I need to do at that time to 
assure correctness will be less than the cumulative effort in working 
with models driven by table defs.

> Also, in fiddling with Og and a non-real database (MySQL) I've found Og
> to be quite temperamental.  The output from irb is long.  If no-one
> would mind most of 200 lines in here I can do that else I can send it
> privately.  However the gist of it is that Og will sometimes load
> properly and sometimes it won't.  And sometimes it will recreate the
> tables, and sometimes it won't.  I have no explanation for this.  Is it
> a known bug or is one of them non-obvious features?  I haven't tested
> the same code on Postgres or sqlite or any other store.  Some of the
> code I am using is from
> http://www.infurious.com/blogs/index.php/aidan/2006/02/14/og_tutorial_part_one.
> (See comment above about being dangerous.)  
> 

I've not seen this myself.


...
(much stuff omitted)
> 
> And of course in the really funkified version AR would have oid's.  
> 
> Am I going about this completely the wrong way or is this just overall a
> really bad idea?  

It's not something I would do; I would rather see about using the Rails 
VC part with Og filling in for AR.  But even with that I'm not sure what 
the gain would be over simply using Nitro.


-- 
James Britt

"A language that doesn't affect the way you think about programming is 
not worth knowing."
   - A. Perlis


From bryan.a.soto at gmail.com  Tue Mar 21 14:41:08 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 21 Mar 2006 11:41:08 -0800
Subject: [Nitro] Repo update.
Message-ID: <bd93db0d0603211141u63fbb425ne1082c8e1711e138@mail.gmail.com>

I've reverted a couple of patches that were contributing to the
instability to the repo. Hopefully things will be a bit nicer know.
That doesn't mean they're rejected. Just that they're incomplete
because things weren't working properly with them applied.

Just:

$ darcs pull

from devlab. It'll undo them.

This puts us back to our previous test state. All tests pass with Og
and Glue, 1 failure with Nitro. If you get different results, there's
a patching opportunity knocking for you. :)

Here's a list of outstanding patches. If you've tried them out, please
mention it. It saves some time, shows knowledge of the code base, etc.
Basically, it helps them get applied faster.

-
[Nitro] [PATCH] make Nitro work with facets 1.1.0   Kashia Buch
This one is on the top of my list. I should check it out as every day
I run 'gem update -y' and every day I have to unistall the
facets-1.1.0 because I don't have time to check it.

http://rubyforge.org/pipermail/nitro-general/2006-March/003399.html
-
[Nitro] PATCH: tests-camel-case   Bryan Soto
I don't want to apply this one without a fix. But I didn't want people
to miss it. Try it out. It's an interesting one... I spent a few hours
trying to figure out a fix instead of reviewing patches. ;)

http://rubyforge.org/pipermail/nitro-general/2006-March/003372.html
-
[Nitro] PATCH: nitro-fix-dot-in-url   Bryan Soto
I've checked Webrick and Mongrel. If someone could verify SCgi,
FastCgi or even Cgi, I'd feel comfortable applying this.

http://rubyforge.org/pipermail/nitro-general/2006-March/003364.html
-
[Nitro] [PATCH] two small flare-patches   Michael Fellinger
Michael made two changes. I've applied his removal of the size limit
of blog entries. He also changed how the RSS feed works for comments.
It know links back to the article itself. Previously, it would try to
load the admin page and generate an error message. As an example, try
the comments feed for gmosx.com. If there are no other opinions on
what this should do, I'm applying it.

http://rubyforge.org/pipermail/nitro-general/2006-March/003354.html
-
[Nitro] [PATCH] Moved the transformer pipeline at the controller   zimba.tm
I like where this is going, but it's unfinished. It has at least one
problem reported in the thread.

http://rubyforge.org/pipermail/nitro-general/2006-March/003338.html
-
[Nitro] Add :table property to join relationships   Rob Pitt
I missed that this was a patch. Needs to be reviewed.

http://rubyforge.org/pipermail/nitro-general/2006-March/003326.html
-
[Nitro] STI (Again =))   Rob Pitt
I'm going to apply the test case, but this one doesn't seem necessary.

http://rubyforge.org/pipermail/nitro-general/2006-March/003325.html
-
[Nitro] kirby patches and a test fix   Dylan Bruzenak
I like this one. There is one thing that it does which I'd like to
port to the other stores. It makes ogtype a property and Og managed
rather than inserting a field via sql. Seems like a no brainer. The
only thing holding this up is a comparison of the tests. I just want
to make sure it doesn't break anything that wasn't broken before.

http://rubyforge.org/pipermail/nitro-general/2006-March/003306.html
-
[Nitro] PATCH: mongrel-adapter-cookies   Bryan Soto
This works. It just uses code already used by most of the other
adapters. I'm going to apply unless someone finds a reason not to.

http://rubyforge.org/pipermail/nitro-general/2006-March/003283.html
-
[Nitro] [PATCH] Compiler, Action and Element   zimba.tm
I like that this makes all compilers have an identical signature. It
makes dynamic dispatching easier. It did break a test, so it needs a
little bit of work.

http://rubyforge.org/pipermail/nitro-general/2006-March/003279.html
-
[Nitro] [PATCH] Stop empty cookies crashing nitro.   Rob Pitt
This one seems like a good fix. I'll apply it unless there is a good
reason not to.

http://rubyforge.org/pipermail/nitro-general/2006-March/003250.html
-

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From george.moschovitis at gmail.com  Tue Mar 21 15:27:02 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 21 Mar 2006 21:27:02 +0100
Subject: [Nitro] Using Nitro
In-Reply-To: <441F2B46.7070009@neurogami.com>
References: <4b6f054f0603201300n73095ae7r1a4bed6c8b735af1@mail.gmail.com>
	<441F2B46.7070009@neurogami.com>
Message-ID: <afd80b540603211227q4c6fea81o4a6e20e820c96b39@mail.gmail.com>

> Rails has assorted niceties, such as the bundled testing.  I imagine

Nitro supports Rails style testing. Have a look at

lib/nitro/test
lib/og/test

About scalling, if you use the output caching mechanism nitro provides
*excellent* scaleability. In general I havent seen any problems with
performance.

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Tue Mar 21 15:36:02 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 21 Mar 2006 21:36:02 +0100
Subject: [Nitro] Repo update.
In-Reply-To: <bd93db0d0603211141u63fbb425ne1082c8e1711e138@mail.gmail.com>
References: <bd93db0d0603211141u63fbb425ne1082c8e1711e138@mail.gmail.com>
Message-ID: <afd80b540603211236m6ced8b84s5c6d0a6d7f81cdc3@mail.gmail.com>

thanks for the informative email!

-g.


On 3/21/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> I've reverted a couple of patches that were contributing to the
> instability to the repo. Hopefully things will be a bit nicer know.
> That doesn't mean they're rejected. Just that they're incomplete
> because things weren't working properly with them applied.
>
> Just:
>
> $ darcs pull
>
> from devlab. It'll undo them.
>
> This puts us back to our previous test state. All tests pass with Og
> and Glue, 1 failure with Nitro. If you get different results, there's
> a patching opportunity knocking for you. :)
>
> Here's a list of outstanding patches. If you've tried them out, please
> mention it. It saves some time, shows knowledge of the code base, etc.
> Basically, it helps them get applied faster.
>
> -
> [Nitro] [PATCH] make Nitro work with facets 1.1.0   Kashia Buch
> This one is on the top of my list. I should check it out as every day
> I run 'gem update -y' and every day I have to unistall the
> facets-1.1.0 because I don't have time to check it.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003399.html
> -
> [Nitro] PATCH: tests-camel-case   Bryan Soto
> I don't want to apply this one without a fix. But I didn't want people
> to miss it. Try it out. It's an interesting one... I spent a few hours
> trying to figure out a fix instead of reviewing patches. ;)
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003372.html
> -
> [Nitro] PATCH: nitro-fix-dot-in-url   Bryan Soto
> I've checked Webrick and Mongrel. If someone could verify SCgi,
> FastCgi or even Cgi, I'd feel comfortable applying this.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003364.html
> -
> [Nitro] [PATCH] two small flare-patches   Michael Fellinger
> Michael made two changes. I've applied his removal of the size limit
> of blog entries. He also changed how the RSS feed works for comments.
> It know links back to the article itself. Previously, it would try to
> load the admin page and generate an error message. As an example, try
> the comments feed for gmosx.com. If there are no other opinions on
> what this should do, I'm applying it.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003354.html
> -
> [Nitro] [PATCH] Moved the transformer pipeline at the controller   zimba.tm
> I like where this is going, but it's unfinished. It has at least one
> problem reported in the thread.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003338.html
> -
> [Nitro] Add :table property to join relationships   Rob Pitt
> I missed that this was a patch. Needs to be reviewed.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003326.html
> -
> [Nitro] STI (Again =))   Rob Pitt
> I'm going to apply the test case, but this one doesn't seem necessary.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003325.html
> -
> [Nitro] kirby patches and a test fix   Dylan Bruzenak
> I like this one. There is one thing that it does which I'd like to
> port to the other stores. It makes ogtype a property and Og managed
> rather than inserting a field via sql. Seems like a no brainer. The
> only thing holding this up is a comparison of the tests. I just want
> to make sure it doesn't break anything that wasn't broken before.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003306.html
> -
> [Nitro] PATCH: mongrel-adapter-cookies   Bryan Soto
> This works. It just uses code already used by most of the other
> adapters. I'm going to apply unless someone finds a reason not to.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003283.html
> -
> [Nitro] [PATCH] Compiler, Action and Element   zimba.tm
> I like that this makes all compilers have an identical signature. It
> makes dynamic dispatching easier. It did break a test, so it needs a
> little bit of work.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003279.html
> -
> [Nitro] [PATCH] Stop empty cookies crashing nitro.   Rob Pitt
> This one seems like a good fix. I'll apply it unless there is a good
> reason not to.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003250.html
> -
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From lasso at lassoweb.nu  Tue Mar 21 15:41:07 2006
From: lasso at lassoweb.nu (Lars Olsson)
Date: Tue, 21 Mar 2006 21:41:07 +0100
Subject: [Nitro] Strange behavior when using Og and aggregating functions in
	SQLite
Message-ID: <442064E3.8070600@lassoweb.nu>

Hello!

The addition of aggregating functions in Og is great, but unfortunatly 
it seems a bit brokenin places. When working with some timestamp columns 
in SQLite i ran in to trouble. The following code illustrates the problem:

<snip>
require 'rubygems'
require 'og'

# Turn on Og debugging
$DBG = true

# Sample class
class Foo
	
	property :timestamp, Time
	
	def initialize(timestamp = Time.now())
		self.timestamp = timestamp
	end
	
end

# Initialize Og
Og.setup(:destroy => true,
	:name => 'foo',
	:store => 'sqlite')
	
# Create some Foo objects
foo1 = Foo.new
foo2 = Foo.new(Time.now() - 86400)
foo3 = Foo.new(Time.now() - 172800)

# Save object to SQLite
foo1.save
foo2.save
foo3.save

# Fetch maximum, minimum and average values
max = Foo.max(:timestamp)
min = Foo.min(:timestamp)
avg = Foo.avg(:timestamp)

# Print values
puts max, min, avg
</snip>

The above code yields:
2006.0
2006.0
2006.0

...which obviously isn't correct. When using the same SQL queries that 
Og generated directly into SQLite everything works fine.

SELECT MAX(timestamp) FROM ogfoo; ==> 2006-03-21 21:27:07
SELECT MIN(timestamp) FROM ogfoo; ==> 2006-03-19 21:27:07
# With AVG Og is consistent with the standard SQLite client
SELECT AVG(timestamp) FROM ogfoo; ==> 2006.0


Is this a known bug?


Kindly

/Lasso

-- 
________________________________________
Lars Olsson
lasso at lassoweb.nu
http://www.lassoweb.nu/


From james_b at neurogami.com  Tue Mar 21 15:50:27 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 21 Mar 2006 13:50:27 -0700
Subject: [Nitro] Using Nitro
In-Reply-To: <afd80b540603211227q4c6fea81o4a6e20e820c96b39@mail.gmail.com>
References: <4b6f054f0603201300n73095ae7r1a4bed6c8b735af1@mail.gmail.com>	<441F2B46.7070009@neurogami.com>
	<afd80b540603211227q4c6fea81o4a6e20e820c96b39@mail.gmail.com>
Message-ID: <44206713.3050104@neurogami.com>

George Moschovitis wrote:
>>Rails has assorted niceties, such as the bundled testing.  I imagine
> 
> 
> Nitro supports Rails style testing. Have a look at
> 
> lib/nitro/test
> lib/og/test
> 

Ah.

> About scalling, if you use the output caching mechanism nitro provides
> *excellent* scaleability. In general I havent seen any problems with
> performance.

I confess to having resorted to assorted hacks to accomplish caching. 
I'll have to look at this closer.

I also just recently got SCGI running, so I may switch from fcgi to 
that, as fcgi + Apache2 seems to be misbehaving.  (I've started using 
SCGI with Rails, with mostly good results. )


-- 
James Britt


From george.moschovitis at gmail.com  Tue Mar 21 15:52:30 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 21 Mar 2006 21:52:30 +0100
Subject: [Nitro] Strange behavior when using Og and aggregating
	functions in SQLite
In-Reply-To: <442064E3.8070600@lassoweb.nu>
References: <442064E3.8070600@lassoweb.nu>
Message-ID: <afd80b540603211252j4caeef3lf3704b1f49f26204@mail.gmail.com>

nope, thanks for reporting!

-g.

On 3/21/06, Lars Olsson <lasso at lassoweb.nu> wrote:
> Hello!
>
> The addition of aggregating functions in Og is great, but unfortunatly
> it seems a bit brokenin places. When working with some timestamp columns
> in SQLite i ran in to trouble. The following code illustrates the problem:
>
> <snip>
> require 'rubygems'
> require 'og'
>
> # Turn on Og debugging
> $DBG = true
>
> # Sample class
> class Foo
>
>         property :timestamp, Time
>
>         def initialize(timestamp = Time.now())
>                 self.timestamp = timestamp
>         end
>
> end
>
> # Initialize Og
> Og.setup(:destroy => true,
>         :name => 'foo',
>         :store => 'sqlite')
>
> # Create some Foo objects
> foo1 = Foo.new
> foo2 = Foo.new(Time.now() - 86400)
> foo3 = Foo.new(Time.now() - 172800)
>
> # Save object to SQLite
> foo1.save
> foo2.save
> foo3.save
>
> # Fetch maximum, minimum and average values
> max = Foo.max(:timestamp)
> min = Foo.min(:timestamp)
> avg = Foo.avg(:timestamp)
>
> # Print values
> puts max, min, avg
> </snip>
>
> The above code yields:
> 2006.0
> 2006.0
> 2006.0
>
> ...which obviously isn't correct. When using the same SQL queries that
> Og generated directly into SQLite everything works fine.
>
> SELECT MAX(timestamp) FROM ogfoo; ==> 2006-03-21 21:27:07
> SELECT MIN(timestamp) FROM ogfoo; ==> 2006-03-19 21:27:07
> # With AVG Og is consistent with the standard SQLite client
> SELECT AVG(timestamp) FROM ogfoo; ==> 2006.0
>
>
> Is this a known bug?
>
>
> Kindly
>
> /Lasso
>
> --
> ________________________________________
> Lars Olsson
> lasso at lassoweb.nu
> http://www.lassoweb.nu/
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Tue Mar 21 15:55:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 21 Mar 2006 12:55:23 -0800
Subject: [Nitro] Path to 0.30.0
Message-ID: <bd93db0d0603211255n7e8f6070q9ca5e593f1b62f8d@mail.gmail.com>

Okay everyone, we need to figure out what we're shooting for with the
next release. We've gotten some good input of late, have tickets and
plenty of patches. We just don't have a vision. Here's my
contributions. Some probably more related to our mythical 1.0.

Please respond with your wishlist. Let's figure out what's acceptable
and it's place in our project timeline.


Documentation. I really like the idea behind http://www.oxyliquit.de .
I encourage everyone who hasn't signed up yet to please do so. Ask and
answer questions. Do better than I have with my measly two points. :(

The results of the documentation survey (posted on the trac wiki)
shows that everyone wants docs I'm not sure who, other than George,
can provide. I'm trying to learn more about web apps (I'm a newcomer
who wanted to increase his skill set) and hopefully will be able to
provide some after more study. I'm becoming more familiar with the
codebase with every patch and bug report (a reason why patches take so
long for me to review if you were wondering. I'm more familiar with Og
than Nitro). If you are in a position where you can write some and
have a question, I'd be happy to dig through code to find an answer.
-
I agree with Trans. Glue can be removed. Validations and properties
should be moved to Og. Aspects to calibre or facets or something. I'm
not quite sure what the difference is. The rest on a case by case
basis. This, of course, makes things rather interesting since Nitro
expects there to be properties. But perhaps it's for the better to
remove that expectation.
-
Gen, I think has some potential. It's a good idea, it just hasn't been
worked. As an example, imagine having models created for you via:
$ gen models --from-db db_name --adapter mysql
$ gen models --from-schema-file create.sql

Support for index fields other than oid would need to be verified.
-
Zed has written gem_plugins, which I think we can use. At least as a
start towards our own plugin system if we don't use it outright.
-
zimba has been working towards making the compiler pipeline per
controller. I'm thinking this can be used, say for named
transformations. Imagine in your controller:

def some_action
  if client.accepts? :pdf
     @generator = :pdf
  end
  # ...
end # content rendered as pdf.

Each compiler pipeline would have it's own templates, but still be
based off the controller (it is just another view if you think about
it). It could render a pdf, csv, excel spreadsheet, mobile web, etc.
-
Compiler improvements. As it stands, the compiler pipeline is...
inflexible. I don't know how it can be improved though yet.
-
Scaffolding. It could use some improvements. The goal I envision is
that someone who doesn't really know how to program can scaffold there
entire application and just use CSS to make it pretty. Probably
unrealistic, but good to shoot for.
-
A user authentication system.
-
Og should be multi-store friendly.
-
I'd like to see less version bumping. I.e. if gen didn't change, there
shouldn't be any reason to release a new gem, I'd say.
-
Beta gems for testing. Not everyone uses darcs.
-
Better Ajax integration. There's a lot of intersting things that can
be done. Most of which can just be generated via Nitro.


Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From george.moschovitis at gmail.com  Tue Mar 21 16:03:49 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Tue, 21 Mar 2006 22:03:49 +0100
Subject: [Nitro] Using Nitro
In-Reply-To: <44206713.3050104@neurogami.com>
References: <4b6f054f0603201300n73095ae7r1a4bed6c8b735af1@mail.gmail.com>
	<441F2B46.7070009@neurogami.com>
	<afd80b540603211227q4c6fea81o4a6e20e820c96b39@mail.gmail.com>
	<44206713.3050104@neurogami.com>
Message-ID: <afd80b540603211303p4458f32j644eb43f6bdaeeb5@mail.gmail.com>

> I confess to having resorted to assorted hacks to accomplish caching.
> I'll have to look at this closer.

I will have to write a post about caching best practices. I am using
some nice techniques to combine output caching and different content
per user.

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From guillaume.pierronnet at gmail.com  Tue Mar 21 16:45:15 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Tue, 21 Mar 2006 22:45:15 +0100
Subject: [Nitro] Repo update.
In-Reply-To: <bd93db0d0603211141u63fbb425ne1082c8e1711e138@mail.gmail.com>
References: <bd93db0d0603211141u63fbb425ne1082c8e1711e138@mail.gmail.com>
Message-ID: <6a7d49ca0603211345t4d32c65cx@mail.gmail.com>

2006/3/21, Bryan Soto <bryan.a.soto at gmail.com>:

> [Nitro] PATCH: nitro-fix-dot-in-url   Bryan Soto
> I've checked Webrick and Mongrel. If someone could verify SCgi,
> FastCgi or even Cgi, I'd feel comfortable applying this.
>
for these three adapters, this is the web server's job to delegate to
nitro or not so that looks good to me.

...

Thanks for this review!



From bryan.a.soto at gmail.com  Tue Mar 21 17:25:24 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 21 Mar 2006 14:25:24 -0800
Subject: [Nitro] Repo update.
In-Reply-To: <6a7d49ca0603211345t4d32c65cx@mail.gmail.com>
References: <bd93db0d0603211141u63fbb425ne1082c8e1711e138@mail.gmail.com>
	<6a7d49ca0603211345t4d32c65cx@mail.gmail.com>
Message-ID: <bd93db0d0603211425o23f4dd1kfbc89469ec82016e@mail.gmail.com>

On 3/21/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> 2006/3/21, Bryan Soto <bryan.a.soto at gmail.com>:
>
> > [Nitro] PATCH: nitro-fix-dot-in-url   Bryan Soto
> > I've checked Webrick and Mongrel. If someone could verify SCgi,
> > FastCgi or even Cgi, I'd feel comfortable applying this.
> >
> for these three adapters, this is the web server's job to delegate to
> nitro or not so that looks good to me.

Works for me. Applied, thanks.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 21 18:07:31 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 21 Mar 2006 15:07:31 -0800
Subject: [Nitro] STI (Again =))
In-Reply-To: <bd93db0d0603150025o523e730ev5336e07c0d2405ca@mail.gmail.com>
References: <30CF8422-9A3A-46FB-8739-CBBF8BBE05EC@motionpath.com>
	<bd93db0d0603150025o523e730ev5336e07c0d2405ca@mail.gmail.com>
Message-ID: <bd93db0d0603211507u569a6178k7e789bae872b7672@mail.gmail.com>

I've added this as a test case to the repo.

On 3/15/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> <testcase>
> require File.join(File.dirname(__FILE__), '..', 'CONFIG.rb')
>
> require 'test/unit'
>
> require 'og'
>
> class TC_Sti_OgType < Test::Unit::TestCase
>   class Human
>     property :name
>
>     schema_inheritance
>
>     def initialize(name)
>       @name = name
>     end
>   end
>
>   class Parent < Human
>     property :job
>   end
>
>   class Child < Human
>     property :toys
>   end
>
>   $og1.manage_classes(Human, Parent, Child)
>
>   def setup
>     Parent.create('mom')
>     Parent.create('dad')
>     Child.create('son')
>   end
>
>   def test_sti_all
>     all = Human.find(:condition => 'name IS NOT NULL')
>     assert_equal(3, all.size)
>     parents = all.select {|h| h.class == Parent }
>     children = all.select {|h| h.class == Child }
>     assert_equal(2, parents.size)
>     assert_equal(1, children.size)
>     assert_equal(%w[dad mom], parents.sort {|x,y| x.name <=>
> y.name}.map {|p| p.name } )
>     assert_equal(%w[son], children.map {|c| c.name } )
>   end
> end
> </testcase>

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 21 18:08:35 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 21 Mar 2006 15:08:35 -0800
Subject: [Nitro] [PATCH] make Nitro work with facets 1.1.0
In-Reply-To: <op.s6nu53cwg3glhz@jo.local>
References: <op.s6nu53cwg3glhz@jo.local>
Message-ID: <bd93db0d0603211508h6066aa3cuac7c5274fa1d923e@mail.gmail.com>

Kashia,

Thanks for the patch. I've applied to the repo.

On 3/19/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hi, since zimba advertised his new facets I absolutely wanted those fuctions.
> Here's a patch to make Nitro work with the new facets and calibre.
>
> I can't verify the errors come from this change though, I'd like other people
> to test it first before integrating.
>
> Tests:
>
> Nitro:
>
> Finished in 94.812502 seconds.
>
>    1) Failure:
> test_all(TC_Element) [./test/nitro/tc_element.rb:80]:
> <"<x:box color=\"#ff0\">\n    xhtml mode\n  </x:box>  \n"> expected to be =~
> </div style/>.
>
> 34 tests, 117 assertions, 1 failures, 0 errors
>
> Glue:
>
> Finished in 0.408187 seconds.
>
>    1) Failure:
> test_manage_entities(TC_MixinsTest) [./test/glue/tc_property_mixins.rb:55]:
> <Pure>
> of type <Class>
> expected to respond_to?<:properties>.
>
>    2) Failure:
> test_managing_mixins(TC_MixinsTest) [./test/glue/tc_property_mixins.rb:64]:
> <Mixin>
> of type <Module>
> expected to respond_to?<:properties>.
>
>    3) Error:
> test_macro_params(Test::TC_N_Properties):
> NoMethodError: undefined method `properties' for Test::Msg:Class
>      /Volumes/Data/build/cvs/nitro/og/lib/og/entity.rb:477:in `method_missing'
>      ./test/glue/tc_property.rb:100:in `test_macro_params'
>
>    4) Error:
> test_props(Test::TC_N_Properties):
> NoMethodError: undefined method `properties' for Test::SubMsg:Class
>      /Volumes/Data/build/cvs/nitro/og/lib/og/entity.rb:477:in `method_missing'
>      ./test/glue/tc_property.rb:78:in `test_props'
>
> 36 tests, 160 assertions, 2 failures, 2 errors
>
> Sadly can't run Og test atm.
>
>
> --
> Feel the love
> http://pinkjuice.com/pics/ruby.png
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Tue Mar 21 20:21:10 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 22 Mar 2006 01:21:10 +0000
Subject: [Nitro] Reap features on the way
Message-ID: <4b6f054f0603211721k6abcdd6er56e2aab1098ef15b@mail.gmail.com>

Just finished adding a new feature to Reap. I have now integrated
setup.rb directly into the install task. (Which I may rename to
'setup' in honor of it. We'll see.)

This means I'm forking setup.rb --but inside of reap. I've found a
couple bugs with setup.rb and haven't gotten any reponse from Minero
in some time, so I figure a fork is the way to go. It will take a
while to get it fully intgrated, but it does the standard install just
fine.

The nicest thing about this is that you no longer need to have
setup.rb included in your package to support manual installation. As
long as a project follows standard dir layout and has a ProjectInfo
file one can just 'reap install' it.

Also, once I have one of those outstanding setup bugs fixed, I should
have the .deb packaging task finished. Generate .deb packages just as
easy as you do .tar.bz2 and .gem!!!

T.

(P.S. I never quite completed the RubyForge release task. I suppose
I'll get to it eventually. But if someone else would like it finsh it,
feel free. Most of the needed code is in place.)



From kgergely at mcl.hu  Tue Mar 21 20:27:10 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Wed, 22 Mar 2006 02:27:10 +0100
Subject: [Nitro] Path to 0.30.0
In-Reply-To: <bd93db0d0603211255n7e8f6070q9ca5e593f1b62f8d@mail.gmail.com>
References: <bd93db0d0603211255n7e8f6070q9ca5e593f1b62f8d@mail.gmail.com>
Message-ID: <20060322012710.GA20685@mlabdial.hit.bme.hu>

Eeer, maybe that's not at the top of the priority list (workaround 
exist), but is the self join problem solved?

Point: directed many_to_many connection between the same type of 
objects, which can be traversed in both direction ("to parents", "to 
childrens").

( If I think it further, it would be also good, if I can express ANY 
undirected graph. Well, here, the connector table has 2 fields, and if 
I swap the 2 fields, I get the same graphs. Woof. Maybe I should google 
for graphs in SQL for this )

thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060321/133c48e4/attachment.bin 

From bryan.a.soto at gmail.com  Tue Mar 21 20:46:18 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 21 Mar 2006 17:46:18 -0800
Subject: [Nitro] Reap features on the way
In-Reply-To: <4b6f054f0603211721k6abcdd6er56e2aab1098ef15b@mail.gmail.com>
References: <4b6f054f0603211721k6abcdd6er56e2aab1098ef15b@mail.gmail.com>
Message-ID: <bd93db0d0603211746y4384bb73h5c9e739900e9bfb1@mail.gmail.com>

Nice work Trans. :)

On 3/21/06, TRANS <transfire at gmail.com> wrote:
> Just finished adding a new feature to Reap. I have now integrated
> setup.rb directly into the install task. (Which I may rename to
> 'setup' in honor of it. We'll see.)
>
> This means I'm forking setup.rb --but inside of reap. I've found a
> couple bugs with setup.rb and haven't gotten any reponse from Minero
> in some time, so I figure a fork is the way to go. It will take a
> while to get it fully intgrated, but it does the standard install just
> fine.
>
> The nicest thing about this is that you no longer need to have
> setup.rb included in your package to support manual installation. As
> long as a project follows standard dir layout and has a ProjectInfo
> file one can just 'reap install' it.
>
> Also, once I have one of those outstanding setup bugs fixed, I should
> have the .deb packaging task finished. Generate .deb packages just as
> easy as you do .tar.bz2 and .gem!!!
>
> T.
>
> (P.S. I never quite completed the RubyForge release task. I suppose
> I'll get to it eventually. But if someone else would like it finsh it,
> feel free. Most of the needed code is in place.)
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 21 20:52:39 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 21 Mar 2006 17:52:39 -0800
Subject: [Nitro] Path to 0.30.0
In-Reply-To: <20060322012710.GA20685@mlabdial.hit.bme.hu>
References: <bd93db0d0603211255n7e8f6070q9ca5e593f1b62f8d@mail.gmail.com>
	<20060322012710.GA20685@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603211752s5b9764e7wc6145afdf16e28e0@mail.gmail.com>

On 3/21/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Eeer, maybe that's not at the top of the priority list (workaround
> exist), but is the self join problem solved?
>

I'm not sure. :)

I have those threads marked as unread so I can go through them. I just
haven't gone through them yet. Sorry...

> Point: directed many_to_many connection between the same type of
> objects, which can be traversed in both direction ("to parents", "to
> childrens").
>
> ( If I think it further, it would be also good, if I can express ANY
> undirected graph. Well, here, the connector table has 2 fields, and if
> I swap the 2 fields, I get the same graphs. Woof. Maybe I should google
> for graphs in SQL for this )
>

Your definitely taking databases further than I ever have. :)

I don't know if you've done it already, but could you provide a test
case? We can open a ticket and see what we can come up with. :)

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From timh at dirtymonday.net  Tue Mar 21 22:52:33 2006
From: timh at dirtymonday.net (TimH)
Date: Tue, 21 Mar 2006 19:52:33 -0800
Subject: [Nitro] Spark, Og, ruby-postgres snapshots
Message-ID: <20060321195233.1af87833.timh@dirtymonday.net>

I hope it's OK to bring this back up, but using the latest Nitro and Spark from the repo.nitrohq.com repository, I still get this error when using the recent (starting late November 2005) ruby-postgres snapshots:

E, [2006-03-21T19:30:05.240018 #19264] ERROR -- : Error while handling '/view/Home'.
E, [2006-03-21T19:30:05.251746 #19264] ERROR -- : private method `gsub!' called for #<DateTime: 106004828101/43200,0,2299161>

The basic issue is that in the newer API, a date object will be returned by a query for a date field in a postgres table instead of a string containing the date information.  This isn't the only change, but it was the one preventing me from running Spark.  I found a simple fix that seems to completely solve this issue for me.  Small patch fragment below.

hunk ./og/lib/og/store/sql.rb 68
-    return Time.parse(str)
+    return Time.parse(str.to_str)

--TimH


From billk at cts.com  Tue Mar 21 22:53:37 2006
From: billk at cts.com (Bill Kelly)
Date: Tue, 21 Mar 2006 19:53:37 -0800
Subject: [Nitro] Mixing Nitro and static pages?
Message-ID: <05ba01c64d64$332b4560$6442a8c0@musicbox>

Hi,

I have an existing website consisting of static .html pages,
but I'd like to add some new links to the site which are
rendered dynamically by Nitro.  

Is this possible?  Or would I be better off converting the
site to .xhtml and rendering everything with Nitro?  (I would
prefer to keep the .html pages as they are, because someone
else is creating them with a Dreamweaver-like program.)


Thanks,

Bill




From timh at dirtymonday.net  Tue Mar 21 22:56:04 2006
From: timh at dirtymonday.net (TimH)
Date: Tue, 21 Mar 2006 19:56:04 -0800
Subject: [Nitro] Spark, Og, ruby-postgres snapshots
In-Reply-To: <20060321195233.1af87833.timh@dirtymonday.net>
References: <20060321195233.1af87833.timh@dirtymonday.net>
Message-ID: <20060321195604.72542604.timh@dirtymonday.net>

On Tue, 21 Mar 2006 19:52:33 -0800
TimH <timh at dirtymonday.net> wrote:

> hunk ./og/lib/og/store/sql.rb 68
> -    return Time.parse(str)
> +    return Time.parse(str.to_str)


Damn, that should say "to_s"

--TimH


From manveru at weez.co.jp  Tue Mar 21 23:12:45 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Wed, 22 Mar 2006 13:12:45 +0900
Subject: [Nitro] Mixing Nitro and static pages?
In-Reply-To: <05ba01c64d64$332b4560$6442a8c0@musicbox>
References: <05ba01c64d64$332b4560$6442a8c0@musicbox>
Message-ID: <200603221312.45623.manveru@weez.co.jp>

This shouldn't pose a problem, you just put the pages into the 
templates-directory
however, nitro is looking for xhtml not html - so you might have to do a 
batch-renaming or patch nitro...

require 'FileUtils'
Dir['*.html'].each{|f| FileUtils.mv(f, File.basename(f)+'.html')}

it's not the best solution of course... but you could also do something in the 
controller like:

def method_missing(*args)
  File.readlines('static_files_dir/'+args.first).each do |line|
    @out << line
  end
end

which is a nice little hack imho :)
however, there are nice ways to do that also in 3 lines... but this is not 
tested, so please don't complain if it's not working correctly:

class Render
  setting :template_extension, :default => 'html'
end

have fun :)

On Wednesday 22 March 2006 12:53, Bill Kelly wrote:
> Hi,
>
> I have an existing website consisting of static .html pages,
> but I'd like to add some new links to the site which are
> rendered dynamically by Nitro.
>
> Is this possible?  Or would I be better off converting the
> site to .xhtml and rendering everything with Nitro?  (I would
> prefer to keep the .html pages as they are, because someone
> else is creating them with a Dreamweaver-like program.)
>
>
> Thanks,
>
> Bill
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general


From james_b at neurogami.com  Wed Mar 22 00:06:18 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 21 Mar 2006 22:06:18 -0700
Subject: [Nitro] Mixing Nitro and static pages?
In-Reply-To: <200603221312.45623.manveru@weez.co.jp>
References: <05ba01c64d64$332b4560$6442a8c0@musicbox>
	<200603221312.45623.manveru@weez.co.jp>
Message-ID: <4420DB4A.1020206@neurogami.com>

Michael Fellinger wrote:
> This shouldn't pose a problem, you just put the pages into the 
> templates-directory
> however, nitro is looking for xhtml not html - so you might have to do a 
> batch-renaming or patch nitro...

Well, if you are using Apache, for example, you can use mod_rewrite to 
server disk files directly, and bypass Nitro.

> 


-- 
James Britt

Judge a man by his questions, rather than his answers.
  - Voltaire


From billk at cts.com  Wed Mar 22 05:36:30 2006
From: billk at cts.com (Bill Kelly)
Date: Wed, 22 Mar 2006 02:36:30 -0800
Subject: [Nitro] Mixing Nitro and static pages?
References: <05ba01c64d64$332b4560$6442a8c0@musicbox>
	<200603221312.45623.manveru@weez.co.jp>
Message-ID: <069401c64d9c$7bb9d110$6442a8c0@musicbox>

Hi,

From: "Michael Fellinger" <manveru at weez.co.jp>
>
> This shouldn't pose a problem, you just put the pages into the 
> templates-directory
> however, nitro is looking for xhtml not html - so you might have to do a 
> batch-renaming or patch nitro...

Cool... But doesn't Nitro parse these files?  For ex. in one of 
my "real" .xhtml files:

    <tr for="prod in @products">
      <td>
        <img src="#{prod.image_url}" /> <br />
        <b>#{prod.name}</b>
      </td>
      <td>#{prod.description}</td>
      <td>$#{prod.price} <br /> (<a href="admin/products/edit/#{prod.oid}">edit</a>)</td>
    </tr>

I presume Nitro is parsing the xhtml to find the closing </tr>.

If I rename a .html file as .xhtml, and it has html-style <br>
tags, etc. that are not valid xhtml because there's no closing
tag and they are not self-closed, would that be a problem?


Regards,

Bill




From manveru at weez.co.jp  Wed Mar 22 05:47:29 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Wed, 22 Mar 2006 19:47:29 +0900
Subject: [Nitro] Mixing Nitro and static pages?
In-Reply-To: <069401c64d9c$7bb9d110$6442a8c0@musicbox>
References: <05ba01c64d64$332b4560$6442a8c0@musicbox>
	<200603221312.45623.manveru@weez.co.jp>
	<069401c64d9c$7bb9d110$6442a8c0@musicbox>
Message-ID: <200603221947.29773.manveru@weez.co.jp>

I think that is quite possible, nitro checks the validity of a xhtml-file to 
be xhtml-strict... tho i'm not sure why it does that, but it helped me some 
times already to stay at standard.

Not sure how to get around this one... i guess you could do a new pipeline for 
the old HTML4.01, but this being really speculative and atm it's one of the 
hardest things you can do in nitro because of it's constant flux in that 
area.
Maybe somebody else has some advice on that? Could you please also post that 
question at http://www.oxyliquit.de - we are collecting documentation there.

~~~~manveru


On Wednesday 22 March 2006 19:36, Bill Kelly wrote:
> Hi,
>
> From: "Michael Fellinger" <manveru at weez.co.jp>
>
> > This shouldn't pose a problem, you just put the pages into the
> > templates-directory
> > however, nitro is looking for xhtml not html - so you might have to do a
> > batch-renaming or patch nitro...
>
> Cool... But doesn't Nitro parse these files?  For ex. in one of
> my "real" .xhtml files:
>
>     <tr for="prod in @products">
>       <td>
>         <img src="#{prod.image_url}" /> <br />
>         <b>#{prod.name}</b>
>       </td>
>       <td>#{prod.description}</td>
>       <td>$#{prod.price} <br /> (<a
> href="admin/products/edit/#{prod.oid}">edit</a>)</td> </tr>
>
> I presume Nitro is parsing the xhtml to find the closing </tr>.
>
> If I rename a .html file as .xhtml, and it has html-style <br>
> tags, etc. that are not valid xhtml because there's no closing
> tag and they are not self-closed, would that be a problem?
>
>
> Regards,
>
> Bill
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general


From james_b at neurogami.com  Wed Mar 22 10:02:31 2006
From: james_b at neurogami.com (James Britt)
Date: Wed, 22 Mar 2006 08:02:31 -0700
Subject: [Nitro] Mixing Nitro and static pages?
In-Reply-To: <200603221947.29773.manveru@weez.co.jp>
References: <05ba01c64d64$332b4560$6442a8c0@musicbox>	<200603221312.45623.manveru@weez.co.jp>	<069401c64d9c$7bb9d110$6442a8c0@musicbox>
	<200603221947.29773.manveru@weez.co.jp>
Message-ID: <44216707.30403@neurogami.com>

Michael Fellinger wrote:
> I think that is quite possible, nitro checks the validity of a xhtml-file to 
> be xhtml-strict... tho i'm not sure why it does that, but it helped me some 
> times already to stay at standard.

I believe the pages are loaded by REXML, so they must be proper XML, but 
I don't think there is any DTD validation going on, so the pages need 
not be XHTML per se.



-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://refreshingcities.org   - Design, technology, usability


From kashia at vfemail.net  Wed Mar 22 13:48:41 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Wed, 22 Mar 2006 19:48:41 +0100
Subject: [Nitro] Using Nitro
In-Reply-To: <afd80b540603211303p4458f32j644eb43f6bdaeeb5@mail.gmail.com>
References: <4b6f054f0603201300n73095ae7r1a4bed6c8b735af1@mail.gmail.com>
	<441F2B46.7070009@neurogami.com>
	<afd80b540603211227q4c6fea81o4a6e20e820c96b39@mail.gmail.com>
	<44206713.3050104@neurogami.com>
	<afd80b540603211303p4458f32j644eb43f6bdaeeb5@mail.gmail.com>
Message-ID: <op.s6tvvfm6g3glhz@jo.local>

Hey, big G :D

> I will have to write a post about caching best practices. I am using
> some nice techniques to combine output caching and different content
> per user.

If you do, please fill out http://oxyliquit.de/question/17 ;)

If this is going to be something bigger, please wait a few days, I'm
adding tutorials to Oxyliquit. (If you're willing to use the Nitro Q&A
as platform for that ;))

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From reid.thompson at ateb.com  Wed Mar 22 16:58:26 2006
From: reid.thompson at ateb.com (Reid Thompson)
Date: Wed, 22 Mar 2006 16:58:26 -0500
Subject: [Nitro] Ruby performance
In-Reply-To: <75e4ed690603220551i26925201k1e45eb18a9362b8f@mail.gmail.com>
References: <75e4ed690603220551i26925201k1e45eb18a9362b8f@mail.gmail.com>
Message-ID: <4421C882.2080701@ateb.com>

Eric Kidd wrote:
> On Mar 22, 2006, at 2:26 PM, Randy Kramer wrote:
>> 3 million updates per day doesn't mean much to me.  With a little 
>> arithmetic,
>> that looks like a sustained average load of ~ 33 TPS.
>
> By an interesting coincidence, Rails sites tend to support about 30 
> hits/second/server on decent hardware, assuming they have to go all 
> the way to the database and render views. With action caching (which 
> bypasses the database and view rendering, but still runs Ruby code), 
> I've seen benchmarks in the 500 hits/second range.
>
> So Ruby might very well be a plausible solution, depending on a number 
> of factors. Given the sweet simplicity of ActiveRecord, you could even 
> spend a couple of days building a prototype and seeing how fast it 
> goes. :-)
>
> Cheers,
> Eric
>
>
>
Simple test, ran from
  within RDE editor
  windows xp,
  testog=# select version();
                                        
version                                         
------------------------------------------------------------------------------------------ 

PostgreSQL 8.1.3 on i686-pc-mingw32, compiled by GCC gcc.exe (GCC) 3.4.2 
(mingw-special)
(1 row)
  with 119 odd processes running ( windows, cygwin, etc, etc)
  1GB ram

10000 inserts in < 60 seconds = 167 tps.

testog=# truncate ogcomment;
TRUNCATE TABLE
testog=# select * from ogcomment;
title | body | author | create_time | oid
-------+------+--------+-------------+-----
(0 rows)

I, [2006-03-22T16:44:48.183000 #3220]  INFO -- : Og uses the Psql store.
Wed Mar 22 16:44:49 Eastern Standard Time 2006
Wed Mar 22 16:45:44 Eastern Standard Time 2006
D, [2006-03-22T16:44:49.464000 #3220] DEBUG -- : Table ogcomment already 
exists
D, [2006-03-22T16:44:49.495000 #3220] DEBUG -- : PostgreSQL processing 
foreign key constraints
D, [2006-03-22T16:44:49.495000 #3220] DEBUG -- : PostgreSQL finished 
setting constraints. No action was taken in 0.00 seconds.
Completed(0)



testog=# select count(*) from ogcomment;
count
-------
10000
(1 row)

testog=# select min(create_time),max(create_time) from ogcomment;
        min         |         max        
---------------------+---------------------
2006-03-22 16:44:49 | 2006-03-22 16:45:44
(1 row)



require 'og'

class Comment
   property :title, String
   property :body, String
   property :author, String
   property :create_time, Time
end
og_psql = {
 :destroy => true,
 :store => :psql,
 :user => 'rthompso',
 :password => 'rthompso',
 :name => 'testog'
}

Og.setup(og_psql)

c = Comment.new
c.title = 'Hello'
c.body = 'World'
c.create_time = Time.now
c.author = 'tml'

puts Time.now
# save the object in the database
1.upto(10000)  { |i|
 c = Comment.new
 c.title = 'Hello'
 c.body = 'World'
 c.create_time = Time.now
 c.author = 'tml'
 c.save
}
puts Time.now


From nissl at tiscali.it  Wed Mar 22 17:25:29 2006
From: nissl at tiscali.it (Massimo Maria Ghisalberti)
Date: Wed, 22 Mar 2006 23:25:29 +0100
Subject: [Nitro] AJAX helper generated javascript
Message-ID: <1143066329.12494.5.camel@nissl.mammuth>

hi all...

I'm new to nitro and I've made some experiment...

question...

is it possiblo to pass some parameters in the AJAX helper generated
javascript function?

ex: <a href="#" id="grabber" onclick="__nc_grab('param'); return
false">link a client</a>

thanks
-- 
Massimo Maria Ghisalberti <nissl at tiscali.it>



From bryan.a.soto at gmail.com  Wed Mar 22 18:57:42 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 22 Mar 2006 15:57:42 -0800
Subject: [Nitro] AJAX helper generated javascript
In-Reply-To: <1143066329.12494.5.camel@nissl.mammuth>
References: <1143066329.12494.5.camel@nissl.mammuth>
Message-ID: <bd93db0d0603221557i69f031b1j81bb55255422f0d9@mail.gmail.com>

Hi,

On 3/22/06, Massimo Maria Ghisalberti <nissl at tiscali.it> wrote:
> hi all...
>
> I'm new to nitro and I've made some experiment...
>
> question...
>
> is it possiblo to pass some parameters in the AJAX helper generated
> javascript function?
>

To me, it looks like it's hard-coded to no params. I'd be interested
in being proven wrong though. :)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Wed Mar 22 19:22:51 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 22 Mar 2006 16:22:51 -0800
Subject: [Nitro] nested templates calls and shared data
In-Reply-To: <6a7d49ca0603150715o7a13c22br@mail.gmail.com>
References: <6a7d49ca0603150715o7a13c22br@mail.gmail.com>
Message-ID: <bd93db0d0603221622y66313a58mf363b22c6bb4e1f5@mail.gmail.com>

On 3/15/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> i spent all my week-end on trying to implement a sort of 'global
> compiling pipeline', with no success.
>
> does anybody have a idea on how to fix it ?

Hi Guill,

Check out nitro/lib/nitro/compiler/script.rb:90

Currently, the script compiler looks for a </body> tag and writes the
script out at the end of it. As a "fugly hack", you can add

<!-- </body> -->

to the end of your album_list.xhtml file.

Perhaps a real fix would be to check if text =~ /<\/body>/ and,
failing that, just appending the the script at the end of text? What
do you think?

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Wed Mar 22 21:09:35 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 22 Mar 2006 18:09:35 -0800
Subject: [Nitro] Spark, Og, ruby-postgres snapshots
In-Reply-To: <20060321195604.72542604.timh@dirtymonday.net>
References: <20060321195233.1af87833.timh@dirtymonday.net>
	<20060321195604.72542604.timh@dirtymonday.net>
Message-ID: <bd93db0d0603221809o2c8a9e43x4d085b0bb2ea6708@mail.gmail.com>

On 3/21/06, TimH <timh at dirtymonday.net> wrote:
> On Tue, 21 Mar 2006 19:52:33 -0800
> TimH <timh at dirtymonday.net> wrote:
>
> > hunk ./og/lib/og/store/sql.rb 68
> > -    return Time.parse(str)
> > +    return Time.parse(str.to_str)
>
>
> Damn, that should say "to_s"

I don't have Postgres handy. Is there any difference in the results of
the Og test suite before and after the patch? I can verify Mysql and
Sqlite3.

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From timh at dirtymonday.net  Wed Mar 22 21:58:03 2006
From: timh at dirtymonday.net (TimH)
Date: Wed, 22 Mar 2006 18:58:03 -0800
Subject: [Nitro] Spark, Og, ruby-postgres snapshots
In-Reply-To: <bd93db0d0603221809o2c8a9e43x4d085b0bb2ea6708@mail.gmail.com>
References: <20060321195233.1af87833.timh@dirtymonday.net>
	<20060321195604.72542604.timh@dirtymonday.net>
	<bd93db0d0603221809o2c8a9e43x4d085b0bb2ea6708@mail.gmail.com>
Message-ID: <20060322185803.7ebc8047.timh@dirtymonday.net>

On Wed, 22 Mar 2006 18:09:35 -0800
"Bryan Soto" <bryan.a.soto at gmail.com> wrote:

> On 3/21/06, TimH <timh at dirtymonday.net> wrote:
> > On Tue, 21 Mar 2006 19:52:33 -0800
> > TimH <timh at dirtymonday.net> wrote:
> >
> > > hunk ./og/lib/og/store/sql.rb 68
> > > -    return Time.parse(str)
> > > +    return Time.parse(str.to_str)
> >
> >
> > Damn, that should say "to_s"
> 
> I don't have Postgres handy. Is there any difference in the results of
> the Og test suite before and after the patch? I can verify Mysql and
> Sqlite3.

	Uh, I don't know.  I started using the postgres release when I ran into other issues I couldn't solve.  Where can I read some info on this test suite?

--TimH


From bryan.a.soto at gmail.com  Thu Mar 23 01:11:00 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 22 Mar 2006 22:11:00 -0800
Subject: [Nitro] Spark, Og, ruby-postgres snapshots
In-Reply-To: <20060322185803.7ebc8047.timh@dirtymonday.net>
References: <20060321195233.1af87833.timh@dirtymonday.net>
	<20060321195604.72542604.timh@dirtymonday.net>
	<bd93db0d0603221809o2c8a9e43x4d085b0bb2ea6708@mail.gmail.com>
	<20060322185803.7ebc8047.timh@dirtymonday.net>
Message-ID: <bd93db0d0603222211iac40d34ue806e62a72d3f667@mail.gmail.com>

On 3/22/06, TimH <timh at dirtymonday.net> wrote:
> On Wed, 22 Mar 2006 18:09:35 -0800
> "Bryan Soto" <bryan.a.soto at gmail.com> wrote:
> > I don't have Postgres handy. Is there any difference in the results of
> > the Og test suite before and after the patch? I can verify Mysql and
> > Sqlite3.
>
>         Uh, I don't know.  I started using the postgres release when I ran into other issues I
> couldn't solve.  Where can I read some info on this test suite?

My apologies Tim. :)

I would have bet money that you had submitted darcs bundles in the
past... I meant the Og test suite in the repo. No worries. Thanks for
the patch. We'll check it out and see about getting it committed.

Many thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From timh at dirtymonday.net  Thu Mar 23 01:35:21 2006
From: timh at dirtymonday.net (TimH)
Date: Wed, 22 Mar 2006 22:35:21 -0800
Subject: [Nitro] Spark, Og, ruby-postgres snapshots
In-Reply-To: <bd93db0d0603222211iac40d34ue806e62a72d3f667@mail.gmail.com>
References: <20060321195233.1af87833.timh@dirtymonday.net>
	<20060321195604.72542604.timh@dirtymonday.net>
	<bd93db0d0603221809o2c8a9e43x4d085b0bb2ea6708@mail.gmail.com>
	<20060322185803.7ebc8047.timh@dirtymonday.net>
	<bd93db0d0603222211iac40d34ue806e62a72d3f667@mail.gmail.com>
Message-ID: <20060322223521.533d6df1.timh@dirtymonday.net>

On Wed, 22 Mar 2006 22:11:00 -0800
"Bryan Soto" <bryan.a.soto at gmail.com> wrote:

> On 3/22/06, TimH <timh at dirtymonday.net> wrote:
> > On Wed, 22 Mar 2006 18:09:35 -0800
> > "Bryan Soto" <bryan.a.soto at gmail.com> wrote:
> > > I don't have Postgres handy. Is there any difference in the results of
> > > the Og test suite before and after the patch? I can verify Mysql and
> > > Sqlite3.
> >
> >         Uh, I don't know.  I started using the postgres release when I ran into other issues I
> > couldn't solve.  Where can I read some info on this test suite?
> 
> My apologies Tim. :)
> 
> I would have bet money that you had submitted darcs bundles in the
> past... I meant the Og test suite in the repo. No worries. Thanks for
> the patch. We'll check it out and see about getting it committed.
> 
> Many thanks,
> 
> Bryan

	Well, on its own it isn't worth much.  I just thought it might help people using the latest ruby-postgres.  but so much other stuff needs to change that just using that to solve the date issue (and I only know that it solves it for Spark, something else may still have trouble) doesn't give anyone much.  I just thought if anybody else was thinking about that problem that might help a little.

--TimH


From nissl at tiscali.it  Thu Mar 23 05:13:52 2006
From: nissl at tiscali.it (Massimo Maria Ghisalberti)
Date: Thu, 23 Mar 2006 11:13:52 +0100
Subject: [Nitro] AJAX helper generated javascript
In-Reply-To: <bd93db0d0603221557i69f031b1j81bb55255422f0d9@mail.gmail.com>
References: <1143066329.12494.5.camel@nissl.mammuth>
	<bd93db0d0603221557i69f031b1j81bb55255422f0d9@mail.gmail.com>
Message-ID: <1143108832.3997.0.camel@nissl.mammuth>

for send params to the generated js code in ClientMorpher I've made this
change:
	
nitro/helper/javascript/morphing.rb:

class ClientMorpher < SharedMorpher
  def before_start(buffer)
    require_script_file :prototype, :effects

	params = @attributes['params']
	case @name
      when 'form'
        @attributes['onsubmit'] = "__nc_#@value(#{params}); return
false"
      else
        @attributes['href'] = '#'
        @attributes['onclick'] = "__nc_#@value(#{params}); return false"
    end
    @attributes.delete(@key)
	@attributes.delete('params')
  end
end
	  
nitro/compiler/script.rb:
	  
...
	  
functions = {}
params = {}

text.scan(/__nc_(.*)\((.*)\)/) do |match|
	functions[match.first] = true
	params[match.last] = true
end	

...

script << %{
function __nc_#{fun}(params) {
  #{function_code}
}

...

with the attribute params in xhtml code is it possible to pass data ti
javascript function:

<a id="grabber_111" client="grab" params="this.id">link a client</a>

<a href="#" id="grabber_111" onclick="__nc_grab(this.id); return
false">link a client</a>

with no params attribute generated code:

<a href="#" id="grabber_111" onclick="__nc_grab(); return false">link a
client</a>

useful? yes, no... boh! :-D

ciao

(Nitro is a very great framework...)


Il giorno mer, 22/03/2006 alle 15.57 -0800, Bryan Soto ha scritto:
> Hi,
> 
> On 3/22/06, Massimo Maria Ghisalberti <nissl at tiscali.it> wrote:
> > hi all...
> >
> > I'm new to nitro and I've made some experiment...
> >
> > question...
> >
> > is it possiblo to pass some parameters in the AJAX helper generated
> > javascript function?
> >
> 
> To me, it looks like it's hard-coded to no params. I'd be interested
> in being proven wrong though. :)
> 
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
> 
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
-- 
Massimo Maria Ghisalberti <nissl at tiscali.it>




From rob at motionpath.com  Thu Mar 23 07:02:12 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 23 Mar 2006 12:02:12 +0000
Subject: [Nitro] how to tell my boss to use nitro?
In-Reply-To: <20060320232900.GB10859@mlabdial.hit.bme.hu>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
Message-ID: <ED7081C1-F2D8-44C8-B731-1F6154124EDE@motionpath.com>

Affected package: php4-4.4.0
Type of problem: PHP -- multiple vulnerabilities.
Reference: <http://www.FreeBSD.org/ports/portaudit/ 
6821a2db-4ab7-11da-932d-00055d790c25.html>

I see this kind of thing often...

Ask him to read Bugtraq. PHP has an appalling security record.

On 20 Mar 2006, at 23:29, Gergely Kontra wrote:

> Hi!
>
> Switching to nitro at my company is not easy, because I'm not a boss.
> Despite of having very serious problems and stucks with PHP, it is  
> very
> hard to persuade him to begin developing in Nitro.
> The facts, and things he said:
> 1) Now pure php is used for web developement
> 2) We can use CakePHP, which is a Rails clone!
> 3) He said: everyone knows PHP, but who knows ruby? If you can't
>    continue developing, what will I do.
>    However, he knows, that yes, ruby is a cool language and has a very
>    good learning curve.
> 4) "Show me a *COMPLEX* site, which uses nitro!"
> 5) "php+apache+mysql is VERY fast"
> 6) "cgi is slow"
>
> thx
> Gergo
> -- 
> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/ 
> ~kgergely ]------+
> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356  
> 9656 |
> | Budapest University of Technology and Economics        ICQ#  
> 175564914 |
> +------------- "Olyan langesz vagyok, hogy poroltoval kellene  
> jarnom!" -+
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

Rob Pitt
Technical Architect

Motionpath Digital Media Ltd.
St Georges Road, Brighton, BN2 1ED.
Office: 01273 608708





From rob at motionpath.com  Thu Mar 23 07:05:31 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 23 Mar 2006 12:05:31 +0000
Subject: [Nitro] how to tell my boss to use nitro?
In-Reply-To: <20060320232900.GB10859@mlabdial.hit.bme.hu>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
Message-ID: <CE2EC037-741E-4165-9B52-7D5E827285B3@motionpath.com>

fastcgi isn't slow either.

php is faster than ruby there is no question but assembler is faster  
than php... let's all develop in assembler...

you have to trade off speed of execution vs ease of maintaining code  
and speed of development.

ruby/nitro beats php all over the place for RAD, and making changes  
to sites is again a lot easier and faster with the cleaner code of ruby.

On 20 Mar 2006, at 23:29, Gergely Kontra wrote:

> Hi!
>
> Switching to nitro at my company is not easy, because I'm not a boss.
> Despite of having very serious problems and stucks with PHP, it is  
> very
> hard to persuade him to begin developing in Nitro.
> The facts, and things he said:
> 1) Now pure php is used for web developement
> 2) We can use CakePHP, which is a Rails clone!
> 3) He said: everyone knows PHP, but who knows ruby? If you can't
>    continue developing, what will I do.
>    However, he knows, that yes, ruby is a cool language and has a very
>    good learning curve.
> 4) "Show me a *COMPLEX* site, which uses nitro!"
> 5) "php+apache+mysql is VERY fast"
> 6) "cgi is slow"
>
> thx
> Gergo
> -- 
> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/ 
> ~kgergely ]------+
> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356  
> 9656 |
> | Budapest University of Technology and Economics        ICQ#  
> 175564914 |
> +------------- "Olyan langesz vagyok, hogy poroltoval kellene  
> jarnom!" -+
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

Rob Pitt
Technical Architect

Motionpath Digital Media Ltd.
St Georges Road, Brighton, BN2 1ED.
Office: 01273 608708





From mischa.kroon at gmail.com  Thu Mar 23 07:45:24 2006
From: mischa.kroon at gmail.com (Mischa Kroon)
Date: Thu, 23 Mar 2006 13:45:24 +0100
Subject: [Nitro] how to tell my boss to use nitro?
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
	<CE2EC037-741E-4165-9B52-7D5E827285B3@motionpath.com>
Message-ID: <000901c64e77$a804a7f0$0a01a8c0@mischabak>

To be honest I wouldn't try tro convince my boss to use Nitro at this point.

1: It's development is in the hands of a couple of persons, if for some 
reason the core team calls it quits then thats the end.

2: Unstable API, Nitro will be changing a lot in the future... or at least 
thats what I'm seeing :)
While change is good it will bring with it a lot of migration whoes.

3: Mass, there are a lot of tutorials out there for PHP. Courses, jobs etc.
For Nitro there is very little in terms of docs.

Etc.

I know this isn't what you wanted to hear.
Good luck either way.




----- Original Message ----- 
From: "Rob Pitt" <rob at motionpath.com>
To: "General discussion about Nitro" <nitro-general at rubyforge.org>
Sent: Thursday, March 23, 2006 1:05 PM
Subject: Re: [Nitro] how to tell my boss to use nitro?


> fastcgi isn't slow either.
>
> php is faster than ruby there is no question but assembler is faster
> than php... let's all develop in assembler...
>
> you have to trade off speed of execution vs ease of maintaining code
> and speed of development.
>
> ruby/nitro beats php all over the place for RAD, and making changes
> to sites is again a lot easier and faster with the cleaner code of ruby.
>
> On 20 Mar 2006, at 23:29, Gergely Kontra wrote:
>
>> Hi!
>>
>> Switching to nitro at my company is not easy, because I'm not a boss.
>> Despite of having very serious problems and stucks with PHP, it is
>> very
>> hard to persuade him to begin developing in Nitro.
>> The facts, and things he said:
>> 1) Now pure php is used for web developement
>> 2) We can use CakePHP, which is a Rails clone!
>> 3) He said: everyone knows PHP, but who knows ruby? If you can't
>>    continue developing, what will I do.
>>    However, he knows, that yes, ruby is a cool language and has a very
>>    good learning curve.
>> 4) "Show me a *COMPLEX* site, which uses nitro!"
>> 5) "php+apache+mysql is VERY fast"
>> 6) "cgi is slow"
>>
>> thx
>> Gergo
>> -- 
>> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/
>> ~kgergely ]------+
>> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356
>> 9656 |
>> | Budapest University of Technology and Economics        ICQ#
>> 175564914 |
>> +------------- "Olyan langesz vagyok, hogy poroltoval kellene
>> jarnom!" -+
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>
> Rob Pitt
> Technical Architect
>
> Motionpath Digital Media Ltd.
> St Georges Road, Brighton, BN2 1ED.
> Office: 01273 608708
>
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general 



From rob at motionpath.com  Thu Mar 23 07:52:01 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 23 Mar 2006 12:52:01 +0000
Subject: [Nitro] how to tell my boss to use nitro?
In-Reply-To: <CE2EC037-741E-4165-9B52-7D5E827285B3@motionpath.com>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
	<CE2EC037-741E-4165-9B52-7D5E827285B3@motionpath.com>
Message-ID: <CEFB23A9-6CED-411C-9866-99059B4697EC@motionpath.com>

i should mention unless your site receives *heavy* traffic you won't  
notice either.

most of the major speed bumps encountered can be got around by using  
sql queries at the bottlenecks.

if your site is receiving so much traffic general interpreter  
overhead is an issue you should consider using C not php.

php does have it's place though... in 1999 :D

On 23 Mar 2006, at 12:05, Rob Pitt wrote:

> fastcgi isn't slow either.
>
> php is faster than ruby there is no question but assembler is faster
> than php... let's all develop in assembler...
>
> you have to trade off speed of execution vs ease of maintaining code
> and speed of development.
>
> ruby/nitro beats php all over the place for RAD, and making changes
> to sites is again a lot easier and faster with the cleaner code of  
> ruby.
>
> On 20 Mar 2006, at 23:29, Gergely Kontra wrote:
>
>> Hi!
>>
>> Switching to nitro at my company is not easy, because I'm not a boss.
>> Despite of having very serious problems and stucks with PHP, it is
>> very
>> hard to persuade him to begin developing in Nitro.
>> The facts, and things he said:
>> 1) Now pure php is used for web developement
>> 2) We can use CakePHP, which is a Rails clone!
>> 3) He said: everyone knows PHP, but who knows ruby? If you can't
>>    continue developing, what will I do.
>>    However, he knows, that yes, ruby is a cool language and has a  
>> very
>>    good learning curve.
>> 4) "Show me a *COMPLEX* site, which uses nitro!"
>> 5) "php+apache+mysql is VERY fast"
>> 6) "cgi is slow"
>>
>> thx
>> Gergo
>> -- 
>> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/
>> ~kgergely ]------+
>> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356
>> 9656 |
>> | Budapest University of Technology and Economics        ICQ#
>> 175564914 |
>> +------------- "Olyan langesz vagyok, hogy poroltoval kellene
>> jarnom!" -+
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>
> Rob Pitt
> Technical Architect
>
> Motionpath Digital Media Ltd.
> St Georges Road, Brighton, BN2 1ED.
> Office: 01273 608708
>
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From rob at motionpath.com  Thu Mar 23 08:01:40 2006
From: rob at motionpath.com (Rob Pitt)
Date: Thu, 23 Mar 2006 13:01:40 +0000
Subject: [Nitro] how to tell my boss to use nitro?
In-Reply-To: <000901c64e77$a804a7f0$0a01a8c0@mischabak>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
	<CE2EC037-741E-4165-9B52-7D5E827285B3@motionpath.com>
	<000901c64e77$a804a7f0$0a01a8c0@mischabak>
Message-ID: <F50674A2-16F4-4ACD-BCE7-2B23F5409243@motionpath.com>

You take a specific version of nitro (or do what we do, take  
snapshots of glycerin and freeze it by making our own gems) to skip  
the unstable API part.

Don't migrate the old sites.

I am sure the guy already knows PHP (most of us do) so the jobs point  
is moot.

He has a point about the dev team but I don't think it will happen.

On 23 Mar 2006, at 12:45, Mischa Kroon wrote:

> To be honest I wouldn't try tro convince my boss to use Nitro at  
> this point.
>
> 1: It's development is in the hands of a couple of persons, if for  
> some
> reason the core team calls it quits then thats the end.
>
> 2: Unstable API, Nitro will be changing a lot in the future... or  
> at least
> thats what I'm seeing :)
> While change is good it will bring with it a lot of migration whoes.
>
> 3: Mass, there are a lot of tutorials out there for PHP. Courses,  
> jobs etc.
> For Nitro there is very little in terms of docs.
>
> Etc.
>
> I know this isn't what you wanted to hear.
> Good luck either way.
>
>
>
>
> ----- Original Message -----
> From: "Rob Pitt" <rob at motionpath.com>
> To: "General discussion about Nitro" <nitro-general at rubyforge.org>
> Sent: Thursday, March 23, 2006 1:05 PM
> Subject: Re: [Nitro] how to tell my boss to use nitro?
>
>
>> fastcgi isn't slow either.
>>
>> php is faster than ruby there is no question but assembler is faster
>> than php... let's all develop in assembler...
>>
>> you have to trade off speed of execution vs ease of maintaining code
>> and speed of development.
>>
>> ruby/nitro beats php all over the place for RAD, and making changes
>> to sites is again a lot easier and faster with the cleaner code of  
>> ruby.
>>
>> On 20 Mar 2006, at 23:29, Gergely Kontra wrote:
>>
>>> Hi!
>>>
>>> Switching to nitro at my company is not easy, because I'm not a  
>>> boss.
>>> Despite of having very serious problems and stucks with PHP, it is
>>> very
>>> hard to persuade him to begin developing in Nitro.
>>> The facts, and things he said:
>>> 1) Now pure php is used for web developement
>>> 2) We can use CakePHP, which is a Rails clone!
>>> 3) He said: everyone knows PHP, but who knows ruby? If you can't
>>>    continue developing, what will I do.
>>>    However, he knows, that yes, ruby is a cool language and has a  
>>> very
>>>    good learning curve.
>>> 4) "Show me a *COMPLEX* site, which uses nitro!"
>>> 5) "php+apache+mysql is VERY fast"
>>> 6) "cgi is slow"
>>>
>>> thx
>>> Gergo
>>> -- 
>>> +-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/
>>> ~kgergely ]------+
>>> | PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356
>>> 9656 |
>>> | Budapest University of Technology and Economics        ICQ#
>>> 175564914 |
>>> +------------- "Olyan langesz vagyok, hogy poroltoval kellene
>>> jarnom!" -+
>>> _______________________________________________
>>> Nitro-general mailing list
>>> Nitro-general at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>> Rob Pitt
>> Technical Architect
>>
>> Motionpath Digital Media Ltd.
>> St Georges Road, Brighton, BN2 1ED.
>> Office: 01273 608708
>>
>>
>>
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general

Rob Pitt
Technical Architect

Motionpath Digital Media Ltd.
St Georges Road, Brighton, BN2 1ED.
Office: 01273 608708





From george.moschovitis at gmail.com  Thu Mar 23 10:11:46 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 17:11:46 +0200
Subject: [Nitro] Reap features on the way
In-Reply-To: <4b6f054f0603211721k6abcdd6er56e2aab1098ef15b@mail.gmail.com>
References: <4b6f054f0603211721k6abcdd6er56e2aab1098ef15b@mail.gmail.com>
Message-ID: <afd80b540603230711j3a7223d0qcfeb03a3f5f56936@mail.gmail.com>

> (P.S. I never quite completed the RubyForge release task. I suppose
> I'll get to it eventually. But if someone else would like it finsh it,
> feel free. Most of the needed code is in place.)


RubyForge release is extremely important. Hope you will find time for
it soon ;-)

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar 23 10:50:39 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 17:50:39 +0200
Subject: [Nitro] Mixing Nitro and static pages?
In-Reply-To: <44216707.30403@neurogami.com>
References: <05ba01c64d64$332b4560$6442a8c0@musicbox>
	<200603221312.45623.manveru@weez.co.jp>
	<069401c64d9c$7bb9d110$6442a8c0@musicbox>
	<200603221947.29773.manveru@weez.co.jp> <44216707.30403@neurogami.com>
Message-ID: <afd80b540603230750r27f78689u3ee2e9d588fe5a23@mail.gmail.com>

If you put your .html files in the 'public' directory, Nitro (in all
adapters, webrick, fcgi, scgi etc) will  serve them as they are,
completely bypassing the rendering process. This is in fact how page
caching works. Xhtml files that support caching generate respective
.html files in public. The next time the webserver requests the file
it looks in public, finds the html and serves it as is, completely
bypassing nitro. When the page is invalidated nitro deletes the
generated html file and the rendering pipeline generates a new one.

regards,
George.


On 3/22/06, James Britt <james_b at neurogami.com> wrote:
> Michael Fellinger wrote:
> > I think that is quite possible, nitro checks the validity of a xhtml-file to
> > be xhtml-strict... tho i'm not sure why it does that, but it helped me some
> > times already to stay at standard.
>
> I believe the pages are loaded by REXML, so they must be proper XML, but
> I don't think there is any DTD validation going on, so the pages need
> not be XHTML per se.
>
>
>
> --
> James Britt
>
> http://www.ruby-doc.org       - Ruby Help & Documentation
> http://www.artima.com/rubycs/ - The Journal By & For Rubyists
> http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
> http://refreshingcities.org   - Design, technology, usability
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar 23 11:09:59 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 18:09:59 +0200
Subject: [Nitro] Small patch
Message-ID: <afd80b540603230809i7b8ed8fy5e8b4c071281ad06@mail.gmail.com>

Bryan,

a small patch, can you apply it against your repo?

thanks!
-g.



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar 23 11:17:44 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 18:17:44 +0200
Subject: [Nitro] Patch
Message-ID: <afd80b540603230817l6954baeesa1a1da82ba459f9f@mail.gmail.com>

here is the patch I forgot, against bryan's repo.

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle
Type: application/octet-stream
Size: 40226 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060323/d20d6cd8/attachment.obj 

From reid.thompson at ateb.com  Thu Mar 23 12:24:35 2006
From: reid.thompson at ateb.com (Reid Thompson)
Date: Thu, 23 Mar 2006 12:24:35 -0500
Subject: [Nitro] How do I force Og to NOT prefix table names with og
Message-ID: <4422D9D3.2060202@ateb.com>

How do I force Og to NOT prefix the table names.
Research led me to try setting :table_prefix to an empty string ala the 
example below.
However, the below still results in table prefixed with og.

og_psql = {
  :destroy => false,
  :table_prefix => '',
  :store => :psql,
  :user => 'rthompso',
  :password => 'rthompso',
  :name => 'testog'
 
}

Og.setup(og_psql)

thanks,
reid


From bryan.a.soto at gmail.com  Thu Mar 23 13:03:10 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 10:03:10 -0800
Subject: [Nitro] How do I force Og to NOT prefix table names with og
In-Reply-To: <4422D9D3.2060202@ateb.com>
References: <4422D9D3.2060202@ateb.com>
Message-ID: <bd93db0d0603231003l35d40a35nc4f82d017796d95@mail.gmail.com>

Hi Reid,

> og_psql = {
>   :destroy => false,
>   :store => :psql,
>   :user => 'rthompso',
>   :password => 'rthompso',
>   :name => 'testog'
>
> }
>

Og.table_prefix = '' # Prefix setting goes here.

> Og.setup(og_psql)

I believe that'll work. :)

Bryan




On 3/23/06, Reid Thompson <reid.thompson at ateb.com> wrote:
> How do I force Og to NOT prefix the table names.
> Research led me to try setting :table_prefix to an empty string ala the
> example below.
> However, the below still results in table prefixed with og.
>
> og_psql = {
>   :destroy => false,
>   :table_prefix => '',
>   :store => :psql,
>   :user => 'rthompso',
>   :password => 'rthompso',
>   :name => 'testog'
>
> }
>
> Og.setup(og_psql)
>
> thanks,
> reid
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From kashia at vfemail.net  Thu Mar 23 12:59:34 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 23 Mar 2006 18:59:34 +0100
Subject: [Nitro] How do I force Og to NOT prefix table names with og
In-Reply-To: <4422D9D3.2060202@ateb.com>
References: <4422D9D3.2060202@ateb.com>
Message-ID: <op.s6vn9kumg3glhz@jo.local>

Hi,

That "should" work... but..

> og_psql = {
>   :destroy => false,
>   :table_prefix => '',
>   :store => :psql,
>   :user => 'rthompso',
>   :password => 'rthompso',
>   :name => 'testog'
>}
>
> Og.setup(og_psql)

Try adding:

Og.table_prefix = ''

right below that.

Kashia


-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From george.moschovitis at gmail.com  Thu Mar 23 13:15:59 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 20:15:59 +0200
Subject: [Nitro] How do I force Og to NOT prefix table names with og
In-Reply-To: <bd93db0d0603231003l35d40a35nc4f82d017796d95@mail.gmail.com>
References: <4422D9D3.2060202@ateb.com>
	<bd93db0d0603231003l35d40a35nc4f82d017796d95@mail.gmail.com>
Message-ID: <afd80b540603231015w3d240852s7c8d154ff3f80118@mail.gmail.com>

> Og.table_prefix = '' # Prefix setting goes here.

correct!

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Thu Mar 23 13:21:59 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 10:21:59 -0800
Subject: [Nitro] AJAX helper generated javascript
In-Reply-To: <1143108832.3997.0.camel@nissl.mammuth>
References: <1143066329.12494.5.camel@nissl.mammuth>
	<bd93db0d0603221557i69f031b1j81bb55255422f0d9@mail.gmail.com>
	<1143108832.3997.0.camel@nissl.mammuth>
Message-ID: <bd93db0d0603231021i1b722e02l9db206695e454ca6@mail.gmail.com>

On 3/23/06, Massimo Maria Ghisalberti <nissl at tiscali.it> wrote:
> useful? yes, no... boh! :-D
>

Flexibility is always useful. :)

Thanks for the patch. I'll check it out and see about getting it applied.

> ciao
>
> (Nitro is a very great framework...)

I hope George sees that. :)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 23 13:29:32 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 10:29:32 -0800
Subject: [Nitro] How do I force Og to NOT prefix table names with og
In-Reply-To: <op.s6vn9kumg3glhz@jo.local>
References: <4422D9D3.2060202@ateb.com> <op.s6vn9kumg3glhz@jo.local>
Message-ID: <bd93db0d0603231029h1ed58567p96313a45c96b7862@mail.gmail.com>

On 3/23/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hi,
>
> That "should" work... but..

You were thinking that too, huh? I wonder if it should be made too.
Any thoughts?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From george.moschovitis at gmail.com  Thu Mar 23 13:32:09 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 20:32:09 +0200
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
	<441F2842.8000305@neurogami.com>
	<4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>
Message-ID: <afd80b540603231032g49a50c0csb35dfc51862fb744@mail.gmail.com>

> Let start with glue. Now besides the fact that it's being asked for
> twice, the more important fact is that we don;t even need it. Please,
> let's get rid of glue altogether. If a glue lib is generic it can go
> into calibre. Otherwise put it in nitro itself and be done with it.

Ok, can you help here? Ie getting the general parts in calibre and the
rest in Nitro and/or Og ?

> Secondly, 'gen' doesn't need to be a seprate project. Maybe one day it
> will be worth having as such, but not today. Put it in nitro too.

lets postpone this decision, I think we will find a use for gen.

> We don't need cmdparse, calibre has consoleapp.rb which works very
> well (and if it lacks anything then fix it or tell me and I will).

Ok, could you perhaps help here? Ie replace cmdparse with the calibre library?

> Are Redhat, ruby-breakpoint and daemons really _required_? Or can they
> be optional? If optional make them optional. It would be nice if we
> could print a  list of optional recommended libraries to this process.

RedCloth and daemons are needed. Ruby-breakpoint could be optional,
but it is just a small lib. Is this such a big problem?

> Then there is that dynamic require warning. We need to do something
> about that but I'm not sure what at the moment. Can we add a :nodoc:
> option to that line?

There is a dynamic reqquire in calibre 1.2.0 i just downloaded ;-)
Anw, this is easy to fix, dont worry about it.

> Notice how nicely this comes across? It may seem silly but this is the
> kind of little thing that helps sell rails.

I think this looks almost the same. It is easy to fix though, perhaps
you can help here as I suggested.

>   Attempting local installation of 'nitro'
> Which is much nicer.

agreed ;-)

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From kashia at vfemail.net  Thu Mar 23 13:41:30 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 23 Mar 2006 19:41:30 +0100
Subject: [Nitro] How do I force Og to NOT prefix table names with og
In-Reply-To: <bd93db0d0603231029h1ed58567p96313a45c96b7862@mail.gmail.com>
References: <4422D9D3.2060202@ateb.com> <op.s6vn9kumg3glhz@jo.local>
	<bd93db0d0603231029h1ed58567p96313a45c96b7862@mail.gmail.com>
Message-ID: <op.s6vp7gbeg3glhz@jo.local>

Hi,

>> That "should" work... but..
>
> You were thinking that too, huh? I wonder if it should be made too.
> Any thoughts?

well, my problem are those "setting :foo ... => ..." thingies in general.
I don't know how to handle them.

Where do I set them? (I know that, but where is that documented)
What does the :doc => "" do? do I need that information at runtime?
Do setting's integrate into rdoc?
Are they @@class_variables or @instance_variables?
If either, why?
How are the setting's different from "options"?

lotsa questions...

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From george.moschovitis at gmail.com  Thu Mar 23 13:48:36 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 20:48:36 +0200
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <op.s6rwjhi4g3glhz@jo.local>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
	<op.s6rwjhi4g3glhz@jo.local>
Message-ID: <afd80b540603231048x329c4da1ke5852aed197ef5f1@mail.gmail.com>

> If you want Nitro to scale, I think caching with "memcached" would be
> the way to go, that means, it'd use the Og store less, which is maybe
> a tad slow because of the database.

Nitro and Og support caching (there is even a memcached caching store
provided by guill).

> I just tested my Nitro FAQ with ab (apache bench) using 400 connects,
> 40 concurrent connections, with a result of 12 requests per second.
> This is on a crappy virtual server.

Here are my results for www.joyful.gr with the same parameters, As you
can see, I get
438.82 req/sec with a dynamic site and per user personalization
(admin/normal views).
The trick is in caching ;-)


ab -n 400 -c 40 http://www.joyful.gr/index

This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Copyright (c) 1998-2002 The Apache Software Foundation, http://www.apache.org/

Benchmarking www.joyful.gr (be patient)
Completed 100 requests
Completed 200 requests
Completed 300 requests
Finished 400 requests


Server Software:        lighttpd/1.4.6
Server Hostname:        www.joyful.gr
Server Port:            80

Document Path:          /index
Document Length:        20482 bytes

Concurrency Level:      40
Time taken for tests:   0.911532 seconds
Complete requests:      400
Failed requests:        0
Write errors:           0
Total transferred:      8327332 bytes
HTML transferred:       8233764 bytes
Requests per second:    438.82 [#/sec] (mean)
Time per request:       91.153 [ms] (mean)
Time per request:       2.279 [ms] (mean, across all concurrent requests)
Transfer rate:          8921.24 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    0   0.4      0       2
Processing:     0   44 134.6     16     908
Waiting:        0   43 134.4     16     906
Total:          0   44 134.9     16     909



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Thu Mar 23 13:50:14 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Thu, 23 Mar 2006 20:50:14 +0200
Subject: [Nitro] How do I force Og to NOT prefix table names with og
In-Reply-To: <op.s6vp7gbeg3glhz@jo.local>
References: <4422D9D3.2060202@ateb.com> <op.s6vn9kumg3glhz@jo.local>
	<bd93db0d0603231029h1ed58567p96313a45c96b7862@mail.gmail.com>
	<op.s6vp7gbeg3glhz@jo.local>
Message-ID: <afd80b540603231050h27b542a4x2328e578b361d88d@mail.gmail.com>

I just supplied a patch with a better implementation of
configuration/settings. There are some docs in the source file, have a
look at

glue/configuration.rb

-g.


On 3/23/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hi,
>
> >> That "should" work... but..
> >
> > You were thinking that too, huh? I wonder if it should be made too.
> > Any thoughts?
>
> well, my problem are those "setting :foo ... => ..." thingies in general.
> I don't know how to handle them.
>
> Where do I set them? (I know that, but where is that documented)
> What does the :doc => "" do? do I need that information at runtime?
> Do setting's integrate into rdoc?
> Are they @@class_variables or @instance_variables?
> If either, why?
> How are the setting's different from "options"?
>
> lotsa questions...
>
> Kashia
>
> --
> Feel the love
> http://pinkjuice.com/pics/ruby.png
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From billk at cts.com  Thu Mar 23 14:15:16 2006
From: billk at cts.com (Bill Kelly)
Date: Thu, 23 Mar 2006 11:15:16 -0800
Subject: [Nitro] Mixing Nitro and static pages?
References: <05ba01c64d64$332b4560$6442a8c0@musicbox><200603221312.45623.manveru@weez.co.jp><069401c64d9c$7bb9d110$6442a8c0@musicbox><200603221947.29773.manveru@weez.co.jp>
	<44216707.30403@neurogami.com>
	<afd80b540603230750r27f78689u3ee2e9d588fe5a23@mail.gmail.com>
Message-ID: <097601c64eae$1eeea470$6442a8c0@musicbox>

From: "George Moschovitis" <george.moschovitis at gmail.com>
>
> If you put your .html files in the 'public' directory, Nitro (in all
> adapters, webrick, fcgi, scgi etc) will  serve them as they are,
> completely bypassing the rendering process. This is in fact how page
> caching works. Xhtml files that support caching generate respective
> .html files in public. The next time the webserver requests the file
> it looks in public, finds the html and serves it as is, completely
> bypassing nitro. When the page is invalidated nitro deletes the
> generated html file and the rendering pipeline generates a new one.

Awesome, thanks !!!!!!!!!!


Regards,

Bill




From kashia at vfemail.net  Thu Mar 23 14:15:45 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 23 Mar 2006 20:15:45 +0100
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <afd80b540603231048x329c4da1ke5852aed197ef5f1@mail.gmail.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
	<op.s6rwjhi4g3glhz@jo.local>
	<afd80b540603231048x329c4da1ke5852aed197ef5f1@mail.gmail.com>
Message-ID: <op.s6vrsj0rg3glhz@jo.local>

Hurry up and write something about caching, how that works, how I
integrate that, I'm envious ;D

> 438.82 req/sec with a dynamic site and per user personalization

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From manveru at weez.co.jp  Wed Mar 22 22:39:48 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Thu, 23 Mar 2006 12:39:48 +0900
Subject: [Nitro] Digging around in the Compiler and what it turned up...
Message-ID: <200603231239.48088.manveru@weez.co.jp>

Ok, Vagabond and I have made a stab at documentation again, this time to one 
of our everyday-friends: the templates - the V part of MVC.
This is what we met on our way, many things have been omitted since we simply 
couldn't grasp the concept in the larger picture (the javascriptmixin is 
_very, very_ meta, guess even for _why  it would take a while to understand 
it...)
There are some things that have not been documented like the localization. 
only the stuff in the standard-pipeline that we use every day.

Hope this makes its way to Oxyliquit and can be used by others for further 
extension.
Have fun - and enjoy, you won't stumble upon new nitro-docu everyday (even tho 
i hope we can change that)



def transform_template(template)
template = StaticInclude.transform(template, self)

Performs static includes. Typically you should include this compiler
as the first stage of the compile pipeline.
This compiler is extremely helpful, typically you would want
to use static includes in many many cases.
Statically include sub-template files.
The target file is included at compile time.
If the given path is relative, the template_root stack of
the controller is traversed. If an absolute path is provided,
templates are searched only in Template.root
gmosx: must be xformed before the <?r pi.
=== Example    
<?include href="root/myfile.xhtml" ?>



template = Morphing.transform(template, self)

:section: A collection of standard morphers.
The base morpher class. Morphers are triggered
by a special 'key' attribute in the xml stream and
transform the owner element. 
###################################################
# attribute: if, unless
#
# <div prop1="one" if="@mycond" prop2="two">@mycond is true</div>
#
# becomes
#
# <?r if @mycond ?>
#   <div prop1="one" prop2="two">@mycond is true</div>
# <?r end ?>
###################################################
# attribute: each, for
#
# <li each="item in array">my item is #{item}</li>
#
# becomes
#
# <?r for item in array ?>
#   <li>my item is #{item}</li>
# <?r end ?>
###################################################
# attribute: times
#
# <li times="3">...</li>
# 
# becomes
#
# <?r 3.times do ?>
#   <li>...</li>
# <?r end ?>
###################################################
    
    
template = LayoutCompiler.transform(template, self)

We have to confess that we've got no idea what the heck is going on in there
but for anybody interested in methods to continue from a rails-application 
might be able to figure this one out. Its a very simple Compiler anyway.
Apart from that, it is commented out in the pipeline, so there is the slight 
feeling of it being redunant.


template = Elements.transform(template, self)

Finds element tags, looks up an Element class for them, creates an instance
of that Element and sets its instance vars to any variables passed in the tag. 
It
then pushes this new instance onto the stack, adding it to its parent if it 
has
one.
Elements are compiled at startup, they are static to some extent and can only
change their behaviour based on instance variables.
One way to have access to the outside world of an Element are sessions, which 
are
globally available.

There are some problems due to its place in the pipeline, so you have to be 
careful
how to use variables in what scope:
<?r if session[:foo] ?>
  \#{session[:foo]}
<?r end ?>
<?r if flash[:error] ?>
  \#{flash[:error]}
<?r end ?>
<?r if @context.path =\~ /financial/ ?>
  #{financial_menu} is a method in the Element.
  the =\~ is because we are within a %~~ string.
<?r end ?>


Example:

template.xhtml:
<?r bar = "foobar" ?>
<Page> Some Content </Page>
<Page foo="#{bar}" />
<Page />

skin.rb:
class Page < Nitro::Element
  def render
    %~
      <div class="page">
        #{foo} #{content} #{foo}
      </div>
    ~
  end
end

becomes:

<div class="page">
  Some Content
</div>
<div class="page">
  foobar  foobar
</div>
<div class="page">
  
</div>

template = Markup.transform(template)

Transform the markup macros.
Maps #(..) to :sanitize. (escapes XML elements?)
Maps #|..| to :markup. (should do RedCloth markup)

glue/lib/glue/markup.rb is where you can try to figure out what the heck
is going on.


template = ScriptCompiler.transform(template, self)

This one is too metaphysical to be explained by a mere human being. So we 
leave
it as an exercise to the reader to figure out how that black magic works.
Important for the understanding is especially the FlickR-example and it helps
when you dig a bit through the rest of nitro too :)

However, if you want to use it, you can just try stuff like 
#{alert("foobar")}

i think that should work... somehow... 


template = Cleanup.transform(template)

This one is best explained by what it really does:

<style something="red"></style> => <style something="red"/>
<include href="magic.xhtml"></include> => <include href="magic.xhtml"/>

Essentially it reduces empty tags while preserving any attributes. It tries to
do something to textareas, but apparently it just ends up doing nothing but in
an expensive fashion.

  def self.cleanup(buf)
    out = buf.dup
    elements = "input|img|br|hr|link|style|render|include|inject|base|meta"
    out.gsub! /<textarea ([^>]*)><\/textarea>/, '<textarea \1>#{}</textarea>'
    out.gsub! /<(#{elements}) ([^>]*)><\/\1>/, '<\1 \2 />'
    out.gsub! /<(#{elements})><\/\1>/, '<\1 />'
    out
  end


template = Template.transform(template)

Now to the very heart of the Compilers, the Template itself.
It basically parses the files in Template.root combined with an method in the 
Controller (method and template together are called action and can act without 
the
other part, so you can have a method without template and a template without 
method)

It basically enables you to use:
<?r some ruby-code ?>
<% some ruby-code %>
<ruby> some ruby-code </ruby>

(why the heck is this defined two times?)
However, this one is static (compiled at startup)?
<include href="foo.xhtml" /> or <inject href="foo.xhtml" />

And this one is rendered at every view... calls the corresponding action
<render href="foo" />

To use your variables in the template, you can use one of these methods
#{my_val} for compilation at parse-time
#\my_val\ for compilation at parse-time (handy for XSL-style)
#[my_val] for compilation at render/eval-time, makes most sense with 
localization.

Additionally, XHTML entities can be used like %rquo; and they'll be converted 
to &rquo;


From bryan.a.soto at gmail.com  Thu Mar 23 15:57:13 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 12:57:13 -0800
Subject: [Nitro] Repo Updated.
Message-ID: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>

darcs pull http://devlab.oree.ch/darcs/nitrohq

Let me know of any problems.

http://devlab.oree.ch/darcs/nitrohq is updated with:

Wed Mar 22 16:33:16 PST 2006  bryan.a.soto at gmail.com
  * glue-cache-file-fix
This time, I'm really adding my file-cacheable-fix.

Win32 doesn't seem to like opening an existing file flocked with exclusive.
Also opens cache files in binary mode. Doesn't seem to make much difference
as Marshall seems to convert it to binary, but just to be safe.


Thu Mar 16 22:01:21 PST 2006  bryan.a.soto at gmail.com
  * nitro-test-config
Adds a CONFIG.rb file to the nitro test suite. Sets
$NITRO_NO_ENVIRONMENT to true and requires nitro. Ensures that the
chdir isn't performed.

(The above was causing some problems in test runs. It wasn't always
set, so occasionally it would perform the chdir and you'd get an error
message about not being able to load a file.)


Thu Mar 23 11:52:06 PST 2006  bryan.a.soto at gmail.com
  * georges-glue-configuration-improvements
Reimplemented Configuration system to be more flexible, added more
comments and documentation.

Written by gmosx. Ported by bryan.


Thu Mar 23 12:38:31 PST 2006  bryan.a.soto at gmail.com
  * test-fix-glue-tc_configuration
This test was displaying evil knowledge of internals. It deleted _all_
settings currently stored causing things to fail in later tests.

--

Be warned on the changes to glue configuration. The tests broke, so
something was changed in an incompatible way. I couldn't figure out
exactly what...

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 23 16:15:04 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 13:15:04 -0800
Subject: [Nitro] [PATCH] two small flare-patches
In-Reply-To: <bd93db0d0603141737od20e6b7i990cd1eb8e2616bf@mail.gmail.com>
References: <200603141625.40585.manveru@weez.co.jp>
	<bd93db0d0603141659o5cbed8d8y9b4dbc6f1454ac08@mail.gmail.com>
	<bd93db0d0603141737od20e6b7i990cd1eb8e2616bf@mail.gmail.com>
Message-ID: <bd93db0d0603231315r3ffc324fia07e3774e5548cf7@mail.gmail.com>

On 3/14/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/14/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> > Hmm... Is your blog running with this patch? It'd be nice to see it in
> > action. By the way, nice redesign. I like the pictures too. :)
>
> Yes, you are. Standard flare just gives an access denied error
> (verifiable from gmosx.com).
>
> Any other opinions on how to handle this? I'm not too familiar with
> blogs... Do they typically link back to the article or to a separate
> comment page?

No other opinions, so you get in by default. Both have been applied, thanks.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 23 16:25:19 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 13:25:19 -0800
Subject: [Nitro] Outstanding Patches.
Message-ID: <bd93db0d0603231325o6b3483fald64198a82cd1a431@mail.gmail.com>

Here's a list of outstanding patches. If you've tried them out, please
mention it. It saves some time, shows knowledge of the code base, etc.
Basically, it helps them get applied faster.

[Nitro] AJAX helper generated javascript
Not submitted as a darcs bundle. Allows params to be added to the
Client functions in the script compiler.
http://rubyforge.org/pipermail/nitro-general/2006-March/003460.html
-
[Nitro] Spark, Og, ruby-postgres snapshots
Not submitted as a darcs bundle. Postgress compatibility fix for newer
versions that cast db fields to their corresponding Ruby types.
http://rubyforge.org/pipermail/nitro-general/2006-March/003445.html
-
[Nitro] PATCH: tests-camel-case   Bryan Soto
I don't want to apply this one without a fix. But I didn't want people
to miss it. Try it out. It's an interesting one... I spent a few hours
trying to figure out a fix instead of reviewing patches. ;)

http://rubyforge.org/pipermail/nitro-general/2006-March/003372.html
-
[Nitro] [PATCH] Moved the transformer pipeline at the controller   zimba.tm
I like where this is going, but it's unfinished. It has at least one
problem reported in the thread.

http://rubyforge.org/pipermail/nitro-general/2006-March/003338.html
-
[Nitro] Add :table property to join relationships   Rob Pitt
I missed that this was a patch. Needs to be reviewed.

http://rubyforge.org/pipermail/nitro-general/2006-March/003326.html
-
[Nitro] kirby patches and a test fix   Dylan Bruzenak
I like this one. There is one thing that it does which I'd like to
port to the other stores. It makes ogtype a property and Og managed
rather than inserting a field via sql. Seems like a no brainer. The
only thing holding this up is a comparison of the tests. I just want
to make sure it doesn't break anything that wasn't broken before.

http://rubyforge.org/pipermail/nitro-general/2006-March/003306.html
-
[Nitro] [PATCH] Compiler, Action and Element   zimba.tm
I like that this makes all compilers have an identical signature. It
makes dynamic dispatching easier. It did break a test, so it needs a
little bit of work.

http://rubyforge.org/pipermail/nitro-general/2006-March/003279.html
-


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 23 16:44:12 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 13:44:12 -0800
Subject: [Nitro] Controller resolution order
In-Reply-To: <bd93db0d0603062237o1bc293dbw51a03e6af69492c6@mail.gmail.com>
References: <440C65DC.7090200@cataclysm-software.net>
	<bd93db0d0603062237o1bc293dbw51a03e6af69492c6@mail.gmail.com>
Message-ID: <bd93db0d0603231344n2da952bbr3b2d925f385cd54e@mail.gmail.com>

On 3/6/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/6/06, Andrew Thompson <vagabond at cataclysm-software.net> wrote:
> > Nitro::Server.map= {'/'=> BaseController,
> > '/admin/users'=>UserController, '/admin'=>AdminController}
> >
> > now /admin points to the admin controller, fine. However /admin/users
> > ALSO points to the admin controller and looks for a 'users' method on
> > it. This seems a bit counter-intuitive, why doesn't Nitro look for a
> > controller to match before looking for a method in a controller?
> >
> > It also does a similar thing for methods with a double underscore (I
> > think foo__bar() maps to foo/bar), if I define foo(), and go to foo/bar,
> > instead of finding foo__bar(), it passes 'bar' to foo(). This works fine
> > if foo() is not defined in the controller however.
> >
>
> Hmm... you make a good point. I suppose George would probably be the
> person to ask as to why current behaviour is what it is. Any opinions
> on whether this is a bug or a feature? :)
>
> Bryan
>

I forgot about this one, but saw you mention it in my log for #nitro. :)

I'm leaning towards making this one a bug. Nitro's philosophy is being
flexible and it's current behaviour here seems to be rather
dictatorial... ;)

Unless someone objects, I think we should create a ticket for this.

By the way, are there any more of these 'forgotten items'? If so,
could I ask everyone to create a ticket on devlab for their
outstanding requests/bugs/etc.? I still have at least 20 threads
marked as unread I keep meaning to go through. I'm sorry I'm not able
to keep up.

My current plan is to get through all the outstanding patches, then
tickets, read my unread threads and finally get back to outlining some
docs when I'm caught up with that. If anyone can spare some time to
pitch in, it'd be greatly appreciated. :)

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 23 16:45:34 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 13:45:34 -0800
Subject: [Nitro] Digging around in the Compiler and what it turned up...
In-Reply-To: <200603231239.48088.manveru@weez.co.jp>
References: <200603231239.48088.manveru@weez.co.jp>
Message-ID: <bd93db0d0603231345h7221b5adgd1194887886c7e52@mail.gmail.com>

Three cheers for Vagabond and manveru! :)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From kgergely at mcl.hu  Thu Mar 23 18:07:20 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Fri, 24 Mar 2006 00:07:20 +0100
Subject: [Nitro] how to tell my boss to use nitro? (docs?, debug?)
In-Reply-To: <9c00d3e00603210015n647a10eag38e3419bd678588a@mail.gmail.com>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
	<9c00d3e00603210015n647a10eag38e3419bd678588a@mail.gmail.com>
Message-ID: <20060323230720.GA10288@mlabdial.hit.bme.hu>

Hi all!

Good news. My boss give Nitro a chance .
I can teach a php man to <anything> (nitro or rails)

But what if he asks about the documentation? (it seems, that rails has 
a wider range of documentation. I still prefer Nitro, if it has docs)

Well, I can show them screencasts, but what then after?

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060323/16b4df4b/attachment.bin 

From transfire at gmail.com  Thu Mar 23 18:11:07 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 23 Mar 2006 23:11:07 +0000
Subject: [Nitro] To require a facet
Message-ID: <4b6f054f0603231511v2d519904ya8b635ea69e8712d@mail.gmail.com>

With comparisons of Facets and ActiveSupport flying about, I have a
hankering to up the ante.

So in that spirit, what do you think of being able to require facets
without any need for the 'facet/...' prefix? Eg.

  require 'facets'
  require 'string/blank'

I can do this simply by appending the facets lib dir to the load path.
I realize this could cause a conflict with other libs, but I very much
doubt that anyone's lib will be named after a Ruby core class
('string' or 'nilclass', etc.) And if such a conflict should occur,
Facets is the last path checked, so it will be overridden by the other
lib, and one can just require the facet the old fashion way to get
around it.

Thoughts?

T.



From kgergely at mcl.hu  Thu Mar 23 19:39:08 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Fri, 24 Mar 2006 01:39:08 +0100
Subject: [Nitro] Why should I use Nitro instead of Rails - missing proof
Message-ID: <20060324003908.GA10705@mlabdial.hit.bme.hu>

Hi!

The $subj is a pretty list, but it doesn't have argues.
URL:
http://www.nitrohq.com/view/FAQ%3AWhy+should+I+use+Nitro+instead+of+Rails

Maybe links to mini-docs with examples would be useful...
Just to prove and show how easy (for mee too :)

BTW: is there any plan how to document nitro, and how can anybody help?
thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060323/ae2edeb7/attachment.bin 

From bryan.a.soto at gmail.com  Thu Mar 23 20:10:48 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 17:10:48 -0800
Subject: [Nitro] how to tell my boss to use nitro? (docs?, debug?)
In-Reply-To: <20060323230720.GA10288@mlabdial.hit.bme.hu>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
	<9c00d3e00603210015n647a10eag38e3419bd678588a@mail.gmail.com>
	<20060323230720.GA10288@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603231710s12f0d720ldd2dcded8d69614f@mail.gmail.com>

Hi,

On 3/23/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Hi all!
>
> Good news. My boss give Nitro a chance .
> I can teach a php man to <anything> (nitro or rails)
>

That'll be fun. PHP4 to Ruby is a pretty big difference in thought.
PHP5 might be easier to transition from. :)

> But what if he asks about the documentation? (it seems, that rails has
> a wider range of documentation. I still prefer Nitro, if it has docs)
>

Quite honestly, the best forms of documentation we have currently are:
1) www.oxyliquit.de
2) irc and the mailing list.
3) the RDoc shipped with the gems available via gem_server command.
4) this is a good place for someone to make a surprise announcement. ;)

We've made a _lot_ of progress in stabilizing Nitro and Og and fixing
bugs in the past few months. And with big thanks to Kashia, Fabian,
manveru, Vagabond and Aidan (I'm sorry if I missed someone), docs are
starting to appear. They're digging through the code and figuring
things out and doing good things for NItro. But the majority of the
documentation is still written in Ruby (the source).

So hopefully that will be enough, but I understand if it's not. It is
your boss and job, after all.

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 23 20:34:05 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 17:34:05 -0800
Subject: [Nitro] Why should I use Nitro instead of Rails - missing proof
In-Reply-To: <20060324003908.GA10705@mlabdial.hit.bme.hu>
References: <20060324003908.GA10705@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603231734p20721962ya058bde1e37378ea@mail.gmail.com>

On 3/23/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Hi!
>
> The $subj is a pretty list, but it doesn't have argues.
> URL:
> http://www.nitrohq.com/view/FAQ%3AWhy+should+I+use+Nitro+instead+of+Rails
>
> Maybe links to mini-docs with examples would be useful...
> Just to prove and show how easy (for mee too :)
>

That would be nice. :)
Unfortunately, the wiki is still broken and showing the old cached
version of the page. :(

> BTW: is there any plan how to document nitro, and how can anybody help?

Most of the current good work in documentation is coming from the
#nitro crowd on irc. :)

I don't know if they have specific plans made. I'd love for one of
them to post if they do. I like oxyliquit.de, the question and answer
idea behind it is very good. They've added a new tips and tutorial
section and are even starting to fill it up. :)

In my free time, I'm outlining things in book form. I'd like to see a
hierarki type web site for community editing that can be converted to
pdf or chm for easy reading offline and printing. It's just an
outline, but it's something that might help others if they'd like to
contribute and aren't sure how. I'll try and put what I have up on
devlab for review and improvements. I was thinking in book form, so
most of it is what would be at the beginning of a book. How to
install, the different web servers, types of cgi and databases.

Anyone have any other thoughts or plans?

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 23 20:56:48 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 17:56:48 -0800
Subject: [Nitro] To require a facet
In-Reply-To: <4b6f054f0603231511v2d519904ya8b635ea69e8712d@mail.gmail.com>
References: <4b6f054f0603231511v2d519904ya8b635ea69e8712d@mail.gmail.com>
Message-ID: <bd93db0d0603231756g12822f8bt91e8c94d455ed177@mail.gmail.com>

On 3/23/06, TRANS <transfire at gmail.com> wrote:
> With comparisons of Facets and ActiveSupport flying about, I have a
> hankering to up the ante.
>

Up the ante or add fuel to the fire? ;)

> So in that spirit, what do you think of being able to require facets
> without any need for the 'facet/...' prefix? Eg.
>
>   require 'facets'
>   require 'string/blank'
>
> I can do this simply by appending the facets lib dir to the load path.
> I realize this could cause a conflict with other libs, but I very much
> doubt that anyone's lib will be named after a Ruby core class
> ('string' or 'nilclass', etc.) And if such a conflict should occur,
> Facets is the last path checked, so it will be overridden by the other
> lib, and one can just require the facet the old fashion way to get
> around it.
>
> Thoughts?
>

Mucking with require. You have an evil sense of fun. :)

It definitely seems as though the facet is a part of the standard
library that way. Though I wonder how many people would continue using
require 'facet/string/blank' just to avoid the chance of conflicts
though.

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From aidan at infurious.com  Thu Mar 23 22:08:58 2006
From: aidan at infurious.com (Aidan Rogers)
Date: Fri, 24 Mar 2006 14:08:58 +1100
Subject: [Nitro] Nitro docs
Message-ID: <442362CA.7030707@infurious.com>


 > Anyone have any other thoughts or plans?

Apologies for being unresponsive for so long - real life distracted me :-)

I have a plan to write a simple script to count up all the methods we 
have in each file that are undocumented and post that list on devlab. 
That way, we can see which files are least documented and work on those, 
and we'll also have some indication of how far we have to go to get API 
docs.

Take a look at http://api.rubyonrails.org.  While it's far from perfect, 
it's really useful once you start working on a Rails app to have that 
reference right there for you.

Second part of my plan only came to light after Kashia showed me the 
excellent work that's been done on Oxyliquit's tutorial site - check out 
my Og tutorial once it's been Liquitised ;-) 
(http://oxyliquit.de/tutorial/1).  These tutorials can be added to very 
easily, and will make a really great intro on any topic for Nitro.

Kashia/Manveru: Tutorials don't show up when I click on an individual 
tag (I see questions and tips but not tutorials).  Also, it would rock 
if people could make comments on any given tutorial page, and/or relate 
questions to tutorials easily.

Aidan


From timh at dirtymonday.net  Fri Mar 24 00:30:25 2006
From: timh at dirtymonday.net (TimH)
Date: Thu, 23 Mar 2006 21:30:25 -0800
Subject: [Nitro] FastCGI startup issues
Message-ID: <20060323213025.2d4f2f67.timh@dirtymonday.net>

I am using Spark, lighttpd, and spawn-fcgi to run my Spark fcgi process under my own user permissions.

My run.rb looks like this:
<code>
require 'rubygems'
require 'facet/time/stamp'
require '/wr/bendug/repo.nitrohq.com/glycerin'
require 'nitro_and_og'

require 'part/admin'

require 'src/skin'
require 'src/diff'
require 'src/controller'

include Nitro

class MyConfig
  attr_accessor :db, :webserver
  def initialize
    @db = {
# :destroy => true,
# :store => :sqlite,
# :store => :kirby,
  :store => :psql,
  :name => 'bendug_spark',
  :user => 'bendug',
  :password => 'password'
#      :socket => "/var/run/mysql/mysql.sock"
#      :destroy => false,
#      :connection_count => 1
    }
    @webserver = {
      :BindAddress => "0.0.0.0",
      :Port => 9999,
      :log_filename => "webrick.log",
      :log_level => 5,
      :mode => :live,
    }
  end
end

config = MyConfig.new
puts "Setting up connection to ### #{config.db[:address]}:#{config.db[:name]}"
puts "            for web port ### #{config.webserver[:BindAddress]}:#{config.webserver[:Port]}\n\n"

Og.setup(config.db)

#Compiler.setup_template_transform do |template|
#  template = Elements.transform(template, self)
#  template = Markup.transform(template)
#  template = Template.transform(template)
#end

Wiki.name = 'BendUG'
Wiki.password = 'password'

Glue::Template.root = 'src/template'

Server.map = {
  '/' => WikiController,
  '/admin' => AdminController
}

runner = Runner.new
Runner.mode = config.webserver[:mode]
runner.setup_options
runner.setup_mode
runner.daemonize if runner.daemon

server = Server.new
server.address = config.webserver[:BindAddress]
server.port    = config.webserver[:Port]
server.start()
runner.invoke(server) unless $NITRO_NO_INVOKE
</code>

Then I start it with spawn-fcgi.  I can do this from the command line or with a ruby script like this:

<code>
$user = 'bendug'
$group = 'bendug'
$fcgi_program = '/wr/bendug/spark/public/index.fcgi'
$socket_name = '/tmp/bendug-spark-fcgi.socket'
$spawn_fcgi = '/usr/local/bin/spawn-fcgi'

system($spawn_fcgi,"-s",$socket_name,"-f",$fcgi_program,"-u",$user,"-g",$group)
system("chmod","777",$socket_name)
</code>

The problem is once the runner starts, it prints some info to the screen, but never returns control back to the terminal.

I see info like this:

 D, [2006-03-23T21:06:40.027174 #8114] DEBUG -- : Using Memory sessions.
Setting up connection to ### :bendug_spark
            for web port ### 0.0.0.0:9999

I, [2006-03-23T21:06:42.684206 #8114]  INFO -- : Og uses the Psql store.
D, [2006-03-23T21:06:43.839676 #8114] DEBUG -- : Table ogwikipage already exists
D, [2006-03-23T21:06:43.872080 #8114] DEBUG -- : Join table 'ogj_glue_tag_wikipage' already exists.
D, [2006-03-23T21:06:43.959164 #8114] DEBUG -- : Table ogwikipage_revision already exists
D, [2006-03-23T21:06:44.141410 #8114] DEBUG -- : Join table 'ogj_glue_tag_wikipage' already exists.
D, [2006-03-23T21:06:44.220408 #8114] DEBUG -- : Table ogglue_tag already exists
D, [2006-03-23T21:06:44.244832 #8114] DEBUG -- : Join table 'ogj_glue_tag_wikipage' already exists.
D, [2006-03-23T21:06:44.513942 #8114] DEBUG -- : PostgreSQL processing foreign key constraints
D, [2006-03-23T21:06:44.515770 #8114] DEBUG -- : PostgreSQL finished setting constraints. No action was taken in 0.00 seconds.


But then I have to press return to get a prompt again.

This also means I can't run it with backticks.  This also seems to prevent me from starting Spark when my server boots from rc.local...  The process spawns and my socket file gets there, but spark doesn't run.

If there is an option to make it startup silently, that would probably work.  If it sent that data to the log file and never took control of the terminal, I would be in good shape.

Is there any way to do this?

--TimH


From transfire at gmail.com  Fri Mar 24 01:53:38 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 24 Mar 2006 06:53:38 +0000
Subject: [Nitro] how to tell my boss to use nitro? (docs?, debug?)
In-Reply-To: <bd93db0d0603231710s12f0d720ldd2dcded8d69614f@mail.gmail.com>
References: <20060320232900.GB10859@mlabdial.hit.bme.hu>
	<9c00d3e00603210015n647a10eag38e3419bd678588a@mail.gmail.com>
	<20060323230720.GA10288@mlabdial.hit.bme.hu>
	<bd93db0d0603231710s12f0d720ldd2dcded8d69614f@mail.gmail.com>
Message-ID: <4b6f054f0603232253l6b7a050flbad05204cd852c81@mail.gmail.com>

> They're digging through the code and figuring
> things out and doing good things for NItro. But the majority of the
> documentation is still written in Ruby (the source).

Just a reminder to take some time to expand the source docs while
working on things. That's a great way to improve documentation via
good RDocs.

T.



From bryan.a.soto at gmail.com  Fri Mar 24 02:58:38 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 23 Mar 2006 23:58:38 -0800
Subject: [Nitro] Add :table property to join relationships
In-Reply-To: <DED1373C-1A55-433C-AE7C-719686931C3F@motionpath.com>
References: <BC2D5047-5212-416C-B721-A89818C66950@motionpath.com>
	<200603131126.48485.zimba.tm@gmail.com>
	<DED1373C-1A55-433C-AE7C-719686931C3F@motionpath.com>
Message-ID: <bd93db0d0603232358s374cc1bwe4c306a4bbf8f240@mail.gmail.com>

Wow. Took me long enough. :)

Background for everyone:
Here
http://rubyforge.org/pipermail/nitro-general/2006-February/002901.html
and
http://rubyforge.org/pipermail/nitro-general/2006-February/002930.html

Barring objections, I'm planning to apply this one tomorrow (well,
today in a few minutes). Probably not a good idea to commit at
midnight.

It needs a slight modification to bring it up to date. Had a conflict
in og/lib/glue/taggable.rb. I'll take care of that as it was up to
date when it was sent.

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Fri Mar 24 03:47:13 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 24 Mar 2006 08:47:13 +0000
Subject: [Nitro] What sets Nitro apart from Rails?
In-Reply-To: <afd80b540603231032g49a50c0csb35dfc51862fb744@mail.gmail.com>
References: <4b6f054f0603201310u5daf1ee4w67791b3a8c15e14@mail.gmail.com>
	<441F2842.8000305@neurogami.com>
	<4b6f054f0603201841n1164b41yd7b962351b0f3d70@mail.gmail.com>
	<afd80b540603231032g49a50c0csb35dfc51862fb744@mail.gmail.com>
Message-ID: <4b6f054f0603240047n10cfc73dgbc76fdd68e8a895@mail.gmail.com>

On 3/23/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Let start with glue. Now besides the fact that it's being asked for
> > twice, the more important fact is that we don;t even need it. Please,
> > let's get rid of glue altogether. If a glue lib is generic it can go
> > into calibre. Otherwise put it in nitro itself and be done with it.
>
> Ok, can you help here? Ie getting the general parts in calibre and the
> rest in Nitro and/or Og ?

Yes.

> Ok, could you perhaps help here? Ie replace cmdparse with the calibre library?

Yes.

T.



From transfire at gmail.com  Fri Mar 24 04:08:56 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 24 Mar 2006 09:08:56 +0000
Subject: [Nitro] Stop kicking me Austin! What George said...
Message-ID: <4b6f054f0603240108v45d2ced5gac92c7b01c5eb0ff@mail.gmail.com>

Yes, George. When you said you liked 'core' and 'more', you were
right. I just couldn't hold a rational schema of integration together
until Austin Ziegler set out to punta slap me like bad tequilla on
Cinco de Mio (me thinks he doth protest too much!). I just can't let
him get away with it like that. So I put on the Ritz...

  http://facets.rubyforge.org

Plusa new release of Facets, v1.2.0. Please adjust Nitro as needed.
Many lashes to my back side for the inconvenience. But as you can see
my troubles are well rewarding.

As of 1.3.0 I'll work in the suitable Glue parts. All submissions,
imporvements, fixes and doumentations, of course, are well taken.

"There. That's better."

T.



From george.moschovitis at gmail.com  Fri Mar 24 06:55:03 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Mar 2006 13:55:03 +0200
Subject: [Nitro] Digging around in the Compiler and what it turned up...
In-Reply-To: <200603231239.48088.manveru@weez.co.jp>
References: <200603231239.48088.manveru@weez.co.jp>
Message-ID: <afd80b540603240355g5626d5d3u958c7e20f2e20269@mail.gmail.com>

Why dont you convert this to rdoc comments?

-g.



On 3/23/06, Michael Fellinger <manveru at weez.co.jp> wrote:
> Ok, Vagabond and I have made a stab at documentation again, this time to one
> of our everyday-friends: the templates - the V part of MVC.
> This is what we met on our way, many things have been omitted since we simply
> couldn't grasp the concept in the larger picture (the javascriptmixin is
> _very, very_ meta, guess even for _why  it would take a while to understand
> it...)
> There are some things that have not been documented like the localization.
> only the stuff in the standard-pipeline that we use every day.
>
> Hope this makes its way to Oxyliquit and can be used by others for further
> extension.
> Have fun - and enjoy, you won't stumble upon new nitro-docu everyday (even tho
> i hope we can change that)
>
>
>
> def transform_template(template)
> template = StaticInclude.transform(template, self)
>
> Performs static includes. Typically you should include this compiler
> as the first stage of the compile pipeline.
> This compiler is extremely helpful, typically you would want
> to use static includes in many many cases.
> Statically include sub-template files.
> The target file is included at compile time.
> If the given path is relative, the template_root stack of
> the controller is traversed. If an absolute path is provided,
> templates are searched only in Template.root
> gmosx: must be xformed before the <?r pi.
> === Example
> <?include href="root/myfile.xhtml" ?>
>
>
>
> template = Morphing.transform(template, self)
>
> :section: A collection of standard morphers.
> The base morpher class. Morphers are triggered
> by a special 'key' attribute in the xml stream and
> transform the owner element.
> ###################################################
> # attribute: if, unless
> #
> # <div prop1="one" if="@mycond" prop2="two">@mycond is true</div>
> #
> # becomes
> #
> # <?r if @mycond ?>
> #   <div prop1="one" prop2="two">@mycond is true</div>
> # <?r end ?>
> ###################################################
> # attribute: each, for
> #
> # <li each="item in array">my item is #{item}</li>
> #
> # becomes
> #
> # <?r for item in array ?>
> #   <li>my item is #{item}</li>
> # <?r end ?>
> ###################################################
> # attribute: times
> #
> # <li times="3">...</li>
> #
> # becomes
> #
> # <?r 3.times do ?>
> #   <li>...</li>
> # <?r end ?>
> ###################################################
>
>
> template = LayoutCompiler.transform(template, self)
>
> We have to confess that we've got no idea what the heck is going on in there
> but for anybody interested in methods to continue from a rails-application
> might be able to figure this one out. Its a very simple Compiler anyway.
> Apart from that, it is commented out in the pipeline, so there is the slight
> feeling of it being redunant.
>
>
> template = Elements.transform(template, self)
>
> Finds element tags, looks up an Element class for them, creates an instance
> of that Element and sets its instance vars to any variables passed in the tag.
> It
> then pushes this new instance onto the stack, adding it to its parent if it
> has
> one.
> Elements are compiled at startup, they are static to some extent and can only
> change their behaviour based on instance variables.
> One way to have access to the outside world of an Element are sessions, which
> are
> globally available.
>
> There are some problems due to its place in the pipeline, so you have to be
> careful
> how to use variables in what scope:
> <?r if session[:foo] ?>
>   \#{session[:foo]}
> <?r end ?>
> <?r if flash[:error] ?>
>   \#{flash[:error]}
> <?r end ?>
> <?r if @context.path =\~ /financial/ ?>
>   #{financial_menu} is a method in the Element.
>   the =\~ is because we are within a %~~ string.
> <?r end ?>
>
>
> Example:
>
> template.xhtml:
> <?r bar = "foobar" ?>
> <Page> Some Content </Page>
> <Page foo="#{bar}" />
> <Page />
>
> skin.rb:
> class Page < Nitro::Element
>   def render
>     %~
>       <div class="page">
>         #{foo} #{content} #{foo}
>       </div>
>     ~
>   end
> end
>
> becomes:
>
> <div class="page">
>   Some Content
> </div>
> <div class="page">
>   foobar  foobar
> </div>
> <div class="page">
>
> </div>
>
> template = Markup.transform(template)
>
> Transform the markup macros.
> Maps #(..) to :sanitize. (escapes XML elements?)
> Maps #|..| to :markup. (should do RedCloth markup)
>
> glue/lib/glue/markup.rb is where you can try to figure out what the heck
> is going on.
>
>
> template = ScriptCompiler.transform(template, self)
>
> This one is too metaphysical to be explained by a mere human being. So we
> leave
> it as an exercise to the reader to figure out how that black magic works.
> Important for the understanding is especially the FlickR-example and it helps
> when you dig a bit through the rest of nitro too :)
>
> However, if you want to use it, you can just try stuff like
> #{alert("foobar")}
>
> i think that should work... somehow...
>
>
> template = Cleanup.transform(template)
>
> This one is best explained by what it really does:
>
> <style something="red"></style> => <style something="red"/>
> <include href="magic.xhtml"></include> => <include href="magic.xhtml"/>
>
> Essentially it reduces empty tags while preserving any attributes. It tries to
> do something to textareas, but apparently it just ends up doing nothing but in
> an expensive fashion.
>
>   def self.cleanup(buf)
>     out = buf.dup
>     elements = "input|img|br|hr|link|style|render|include|inject|base|meta"
>     out.gsub! /<textarea ([^>]*)><\/textarea>/, '<textarea \1>#{}</textarea>'
>     out.gsub! /<(#{elements}) ([^>]*)><\/\1>/, '<\1 \2 />'
>     out.gsub! /<(#{elements})><\/\1>/, '<\1 />'
>     out
>   end
>
>
> template = Template.transform(template)
>
> Now to the very heart of the Compilers, the Template itself.
> It basically parses the files in Template.root combined with an method in the
> Controller (method and template together are called action and can act without
> the
> other part, so you can have a method without template and a template without
> method)
>
> It basically enables you to use:
> <?r some ruby-code ?>
> <% some ruby-code %>
> <ruby> some ruby-code </ruby>
>
> (why the heck is this defined two times?)
> However, this one is static (compiled at startup)?
> <include href="foo.xhtml" /> or <inject href="foo.xhtml" />
>
> And this one is rendered at every view... calls the corresponding action
> <render href="foo" />
>
> To use your variables in the template, you can use one of these methods
> #{my_val} for compilation at parse-time
> #\my_val\ for compilation at parse-time (handy for XSL-style)
> #[my_val] for compilation at render/eval-time, makes most sense with
> localization.
>
> Additionally, XHTML entities can be used like %rquo; and they'll be converted
> to &rquo;
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From aglarond at gmail.com  Fri Mar 24 09:01:36 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Fri, 24 Mar 2006 15:01:36 +0100
Subject: [Nitro] Repo Updated.
In-Reply-To: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
References: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
Message-ID: <55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>

On 3/23/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> darcs pull http://devlab.oree.ch/darcs/nitrohq

Just tried to 'get' the repository, but I get a 404:

darcs get http://devlab.oree.ch/darcs/nitrohq
Invalid repository:  http://devlab.oree.ch/darcs/nitrohq

darcs failed:  Failed to download URL
http://devlab.oree.ch/darcs/nitrohq/_darcs/inventory
libcurl: HTTP error (404?)


zimba - have you pulled the repo?  I noticed that a 'wee'-repo is
there in its place.

- Dimitri



From aglarond at gmail.com  Fri Mar 24 09:15:38 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Fri, 24 Mar 2006 15:15:38 +0100
Subject: [Nitro] Repo Updated.
In-Reply-To: <55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>
References: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
	<55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>
Message-ID: <55c107bf0603240615x14a33e73t5ccb4e9faf084ae5@mail.gmail.com>

On 3/24/06, Dimitri Aivaliotis <aglarond at gmail.com> wrote:

> zimba - have you pulled the repo?  I noticed that a 'wee'-repo is
> there in its place.

It's back, now.  Thanks, zimba!

- Dimitri



From fabian at oggu.de  Fri Mar 24 09:32:59 2006
From: fabian at oggu.de (Fabian Buch)
Date: Fri, 24 Mar 2006 15:32:59 +0100
Subject: [Nitro] Digging around in the Compiler and what it turned up...
In-Reply-To: <200603231239.48088.manveru@weez.co.jp>
References: <200603231239.48088.manveru@weez.co.jp>
Message-ID: <4b5ba4c82e408b7466e87ea825e8f1c0@oggu.de>


Am 23.03.2006 um 04:39 schrieb Michael Fellinger:
> Hope this makes its way to Oxyliquit and can be used by others for 
> further
> extension.

It made it into oxyliquit -> http://oxyliquit.de/tip/5
I added some explaining text at the beginning and smothed the rest a 
little.

Fabian



From zimba.tm at gmail.com  Fri Mar 24 09:43:39 2006
From: zimba.tm at gmail.com (zimba.tm)
Date: Fri, 24 Mar 2006 15:43:39 +0100
Subject: [Nitro] Repo Updated.
In-Reply-To: <55c107bf0603240615x14a33e73t5ccb4e9faf084ae5@mail.gmail.com>
References: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
	<55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>
	<55c107bf0603240615x14a33e73t5ccb4e9faf084ae5@mail.gmail.com>
Message-ID: <200603241543.39513.zimba.tm@gmail.com>

On Friday 24 March 2006 15:15, Dimitri Aivaliotis wrote:
> On 3/24/06, Dimitri Aivaliotis <aglarond at gmail.com> wrote:
> > zimba - have you pulled the repo?  I noticed that a 'wee'-repo is
> > there in its place.
>
> It's back, now.  Thanks, zimba!

Yeah sorry :-p

I'm currently restructuring devlab to allow multiple projects easily. It is 
now ready to host a splitted version of nitro (nitro, og, ...)

-- 
Cheers,
  zimba.tm

weblog : http://zimba.oree.ch


From aglarond at gmail.com  Fri Mar 24 09:59:16 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Fri, 24 Mar 2006 15:59:16 +0100
Subject: [Nitro] Repo Updated.
In-Reply-To: <200603241543.39513.zimba.tm@gmail.com>
References: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
	<55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>
	<55c107bf0603240615x14a33e73t5ccb4e9faf084ae5@mail.gmail.com>
	<200603241543.39513.zimba.tm@gmail.com>
Message-ID: <55c107bf0603240659i706cfd58k868212bbe27f61a9@mail.gmail.com>

On 3/24/06, zimba.tm <zimba.tm at gmail.com> wrote:

> I'm currently restructuring devlab to allow multiple projects easily. It is
> now ready to host a splitted version of nitro (nitro, og, ...)

cool.

- Dimitri



From fabian at oggu.de  Fri Mar 24 10:02:37 2006
From: fabian at oggu.de (Fabian Buch)
Date: Fri, 24 Mar 2006 16:02:37 +0100
Subject: [Nitro] Why should I use Nitro instead of Rails - missing proof
In-Reply-To: <bd93db0d0603231734p20721962ya058bde1e37378ea@mail.gmail.com>
References: <20060324003908.GA10705@mlabdial.hit.bme.hu>
	<bd93db0d0603231734p20721962ya058bde1e37378ea@mail.gmail.com>
Message-ID: <1cdb6083d10033a00d45f045131653d7@oggu.de>


Am 24.03.2006 um 02:34 schrieb Bryan Soto:
> On 3/23/06, Gergely Kontra <kgergely at mcl.hu> wrote:
>> BTW: is there any plan how to document nitro, and how can anybody 
>> help?
>
> Most of the current good work in documentation is coming from the
> #nitro crowd on irc. :)
>
> I don't know if they have specific plans made. I'd love for one of
> them to post if they do. I like oxyliquit.de, the question and answer
> idea behind it is very good. They've added a new tips and tutorial
> section and are even starting to fill it up. :)

Dunno whether I can call it specific plans, but there are some. 
Especially Michael Fellinger, Kashia and me are collecting thoughts 
that appear on #nitro and try to convert them to useful doc peaces that 
can be posted as Questions, Tips or Tutorials on Oxyliquit 
(www.oxyliquit.de).
Oxyliquit is only about 2 weeks old and has quite some content already. 
I'm positive it'll grow more and more so that it won't take too long to 
be the number 1 place for new (and even old) Nitro programmers to look 
for documentation (especially for "Howto.."-kinda stuff, a.k.a. not 
plain api (rdoc)).

As soon as we understand what and how Nitro is doing something we try 
to post it to Oxyliquit (see the helper tips for example). Of course 
time is also limited for us though.

> In my free time, I'm outlining things in book form. I'd like to see a
> hierarki type web site for community editing that can be converted to
> pdf or chm for easy reading offline and printing. It's just an

I don't know Hieraki, I guess it's a wiki with the possibility to 
convert the content to pdf?

For community editing:
there is nitrohq.com wiki already which hopefully will work again in 
the future. I think that's the place where community editing should 
take place.
And as I said on #nitro yesterday. Oxyliquit is not meant as wiki.
"16:48 < Oggu> I think if a tutorial doesn't have only one author (who 
publishes his compledet work on oxyliquit), then it should be done in a 
wiki (like on nitrohq.com) or somewhere else where people can freely 
edit and stuff and at the end               just finish the completed 
work as tutorial on oxyliquit.de"

PDF:
Kashia thought about adding pdf-support on oxyliquit.de so that at 
least tutorials can be downloaded as .pdf. I had a look at pdf-writer 
for this purpose, but it's pretty low level, so it'd take quite some 
work (and time) to create an appropriate html2pdf with it. There are 
more planned features for Oxyliquit, so it's not sure yet when this 
will be implemented.
I made a small research for html2pdf, but could only find pdf-writer 
for ruby, anyone who knows more?

Fabian



From fabian at oggu.de  Fri Mar 24 10:52:47 2006
From: fabian at oggu.de (Fabian Buch)
Date: Fri, 24 Mar 2006 16:52:47 +0100
Subject: [Nitro] Nitro docs
In-Reply-To: <442362CA.7030707@infurious.com>
References: <442362CA.7030707@infurious.com>
Message-ID: <5516b9d941b49a0779215ef1ec1ee051@oggu.de>


Am 24.03.2006 um 04:08 schrieb Aidan Rogers:
> Second part of my plan only came to light after Kashia showed me the
> excellent work that's been done on Oxyliquit's tutorial site - check 
> out
> my Og tutorial once it's been Liquitised ;-)
> (http://oxyliquit.de/tutorial/1).  These tutorials can be added to very
> easily, and will make a really great intro on any topic for Nitro.

Tutorials will be improved soon on Oxyliquit (when Kashia is back after 
this weekend). Don't hesitate to post your tutorials already though.
oxyliquit.de even has a proper database backup strategy so that 
hopefully there'll never be a lost tutorial, tip or question.

> Kashia/Manveru: Tutorials don't show up when I click on an individual
> tag (I see questions and tips but not tutorials).  Also, it would rock

That'll be the first Kashia fixes after weekend.

> if people could make comments on any given tutorial page, and/or relate
> questions to tutorials easily.

we'll think about it.

Fabian



From george.moschovitis at gmail.com  Fri Mar 24 12:59:53 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Mar 2006 19:59:53 +0200
Subject: [Nitro] Nitro docs
In-Reply-To: <5516b9d941b49a0779215ef1ec1ee051@oggu.de>
References: <442362CA.7030707@infurious.com>
	<5516b9d941b49a0779215ef1ec1ee051@oggu.de>
Message-ID: <afd80b540603240959w2c9d26a7t12028aa8012fcf5c@mail.gmail.com>

Nice to see the progress on the docs ;-)

keep up the good work!

-g.

On 3/24/06, Fabian Buch <fabian at oggu.de> wrote:
>
> Am 24.03.2006 um 04:08 schrieb Aidan Rogers:
> > Second part of my plan only came to light after Kashia showed me the
> > excellent work that's been done on Oxyliquit's tutorial site - check
> > out
> > my Og tutorial once it's been Liquitised ;-)
> > (http://oxyliquit.de/tutorial/1).  These tutorials can be added to very
> > easily, and will make a really great intro on any topic for Nitro.
>
> Tutorials will be improved soon on Oxyliquit (when Kashia is back after
> this weekend). Don't hesitate to post your tutorials already though.
> oxyliquit.de even has a proper database backup strategy so that
> hopefully there'll never be a lost tutorial, tip or question.
>
> > Kashia/Manveru: Tutorials don't show up when I click on an individual
> > tag (I see questions and tips but not tutorials).  Also, it would rock
>
> That'll be the first Kashia fixes after weekend.
>
> > if people could make comments on any given tutorial page, and/or relate
> > questions to tutorials easily.
>
> we'll think about it.
>
> Fabian
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar 24 13:01:21 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Mar 2006 20:01:21 +0200
Subject: [Nitro] Repo Updated.
In-Reply-To: <55c107bf0603240659i706cfd58k868212bbe27f61a9@mail.gmail.com>
References: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
	<55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>
	<55c107bf0603240615x14a33e73t5ccb4e9faf084ae5@mail.gmail.com>
	<200603241543.39513.zimba.tm@gmail.com>
	<55c107bf0603240659i706cfd58k868212bbe27f61a9@mail.gmail.com>
Message-ID: <afd80b540603241001g237de873nfc2a8a7fae89506b@mail.gmail.com>

> > I'm currently restructuring devlab to allow multiple projects easily. It is
> > now ready to host a splitted version of nitro (nitro, og, ...)
>
> cool.


*not* so cool...

-g.



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar 24 13:04:51 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Mar 2006 20:04:51 +0200
Subject: [Nitro] Repo Updated.
In-Reply-To: <afd80b540603241001g237de873nfc2a8a7fae89506b@mail.gmail.com>
References: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
	<55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>
	<55c107bf0603240615x14a33e73t5ccb4e9faf084ae5@mail.gmail.com>
	<200603241543.39513.zimba.tm@gmail.com>
	<55c107bf0603240659i706cfd58k868212bbe27f61a9@mail.gmail.com>
	<afd80b540603241001g237de873nfc2a8a7fae89506b@mail.gmail.com>
Message-ID: <afd80b540603241004p20b6d8d8wf6d7ce2efa7a0e30@mail.gmail.com>

> *not* so cool...

...but, lets try and see how this work out (even though I ve tried
this in the past and
it *didn't* work out)

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar 24 13:20:45 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Mar 2006 20:20:45 +0200
Subject: [Nitro] Controller resolution order
In-Reply-To: <bd93db0d0603231344n2da952bbr3b2d925f385cd54e@mail.gmail.com>
References: <440C65DC.7090200@cataclysm-software.net>
	<bd93db0d0603062237o1bc293dbw51a03e6af69492c6@mail.gmail.com>
	<bd93db0d0603231344n2da952bbr3b2d925f385cd54e@mail.gmail.com>
Message-ID: <afd80b540603241020v7dceb6f9rd98d5d331bcb63f2@mail.gmail.com>

> Unless someone objects, I think we should create a ticket for this.

Hello, there is a reason for this behaviour (at the moment it supports
the 'inteligent' dispatcher, ie auto nice urls and stuff). I do not
have the time to explain this at the moment. However I know how to
make the dispatcher even more intelligent to handle cases such as
this. I just havent needed it till now, so I didn't bother.

Bryan, please create a ticket, and I will either improve this in the
next week or post a detailed explanation, with my tips how to improve
it so someone else can try it.

regards,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Fri Mar 24 13:25:39 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Fri, 24 Mar 2006 20:25:39 +0200
Subject: [Nitro] Outstanding Patches.
In-Reply-To: <bd93db0d0603231325o6b3483fald64198a82cd1a431@mail.gmail.com>
References: <bd93db0d0603231325o6b3483fald64198a82cd1a431@mail.gmail.com>
Message-ID: <afd80b540603241025v5a4be44bxa3236a63c1b52187@mail.gmail.com>

Bryan, can you please try to integrate as many patches as you can ? I
would like to pull your repo on Monday and work from what you have.
Most probably I will be able to code a few things over the weekend
too, that I will try to merge on Monday.

-g.


On 3/23/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Here's a list of outstanding patches. If you've tried them out, please
> mention it. It saves some time, shows knowledge of the code base, etc.
> Basically, it helps them get applied faster.
>
> [Nitro] AJAX helper generated javascript
> Not submitted as a darcs bundle. Allows params to be added to the
> Client functions in the script compiler.
> http://rubyforge.org/pipermail/nitro-general/2006-March/003460.html
> -
> [Nitro] Spark, Og, ruby-postgres snapshots
> Not submitted as a darcs bundle. Postgress compatibility fix for newer
> versions that cast db fields to their corresponding Ruby types.
> http://rubyforge.org/pipermail/nitro-general/2006-March/003445.html
> -
> [Nitro] PATCH: tests-camel-case   Bryan Soto
> I don't want to apply this one without a fix. But I didn't want people
> to miss it. Try it out. It's an interesting one... I spent a few hours
> trying to figure out a fix instead of reviewing patches. ;)
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003372.html
> -
> [Nitro] [PATCH] Moved the transformer pipeline at the controller   zimba.tm
> I like where this is going, but it's unfinished. It has at least one
> problem reported in the thread.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003338.html
> -
> [Nitro] Add :table property to join relationships   Rob Pitt
> I missed that this was a patch. Needs to be reviewed.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003326.html
> -
> [Nitro] kirby patches and a test fix   Dylan Bruzenak
> I like this one. There is one thing that it does which I'd like to
> port to the other stores. It makes ogtype a property and Og managed
> rather than inserting a field via sql. Seems like a no brainer. The
> only thing holding this up is a comparison of the tests. I just want
> to make sure it doesn't break anything that wasn't broken before.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003306.html
> -
> [Nitro] [PATCH] Compiler, Action and Element   zimba.tm
> I like that this makes all compilers have an identical signature. It
> makes dynamic dispatching easier. It did break a test, so it needs a
> little bit of work.
>
> http://rubyforge.org/pipermail/nitro-general/2006-March/003279.html
> -
>
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Fri Mar 24 13:47:35 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 10:47:35 -0800
Subject: [Nitro] Controller resolution order
In-Reply-To: <afd80b540603241020v7dceb6f9rd98d5d331bcb63f2@mail.gmail.com>
References: <440C65DC.7090200@cataclysm-software.net>
	<bd93db0d0603062237o1bc293dbw51a03e6af69492c6@mail.gmail.com>
	<bd93db0d0603231344n2da952bbr3b2d925f385cd54e@mail.gmail.com>
	<afd80b540603241020v7dceb6f9rd98d5d331bcb63f2@mail.gmail.com>
Message-ID: <bd93db0d0603241047m10b70620s225dec728e320811@mail.gmail.com>

On 3/24/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Unless someone objects, I think we should create a ticket for this.
>
> Hello, there is a reason for this behaviour (at the moment it supports
> the 'inteligent' dispatcher, ie auto nice urls and stuff). I do not
> have the time to explain this at the moment. However I know how to
> make the dispatcher even more intelligent to handle cases such as
> this. I just havent needed it till now, so I didn't bother.
>
> Bryan, please create a ticket, and I will either improve this in the
> next week or post a detailed explanation, with my tips how to improve
> it so someone else can try it.
>

Created last night so I wouldn't forget again. :)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Fri Mar 24 13:50:27 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 10:50:27 -0800
Subject: [Nitro] Outstanding Patches.
In-Reply-To: <afd80b540603241025v5a4be44bxa3236a63c1b52187@mail.gmail.com>
References: <bd93db0d0603231325o6b3483fald64198a82cd1a431@mail.gmail.com>
	<afd80b540603241025v5a4be44bxa3236a63c1b52187@mail.gmail.com>
Message-ID: <bd93db0d0603241050p329436aare16da7e345067026@mail.gmail.com>

On 3/24/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Bryan, can you please try to integrate as many patches as you can ? I
> would like to pull your repo on Monday and work from what you have.
> Most probably I will be able to code a few things over the weekend
> too, that I will try to merge on Monday.
>
> -g.

I'm trying too. Please work from

http://devlab.oree.ch/darcs/nitrohq

though. That is the up to date one with the most patches integrated.
It's also stable which is a plus.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Fri Mar 24 13:51:49 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 10:51:49 -0800
Subject: [Nitro] FastCGI startup issues
In-Reply-To: <20060323213025.2d4f2f67.timh@dirtymonday.net>
References: <20060323213025.2d4f2f67.timh@dirtymonday.net>
Message-ID: <bd93db0d0603241051o7c37f5d2jf009e85b32080b6d@mail.gmail.com>

I've never used spawn-fcgi before. Looks like I have some installing to do. :)

Will let you know if I can dig anything up.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From timh at dirtymonday.net  Fri Mar 24 14:05:51 2006
From: timh at dirtymonday.net (Tim Howe)
Date: Fri, 24 Mar 2006 11:05:51 -0800
Subject: [Nitro] FastCGI startup issues
In-Reply-To: <bd93db0d0603241051o7c37f5d2jf009e85b32080b6d@mail.gmail.com>
References: <20060323213025.2d4f2f67.timh@dirtymonday.net>
	<bd93db0d0603241051o7c37f5d2jf009e85b32080b6d@mail.gmail.com>
Message-ID: <20060324110551.62b4b7b8.timh@dirtymonday.net>

On Fri, 24 Mar 2006 10:51:49 -0800
"Bryan Soto" <bryan.a.soto at gmail.com> wrote:

> I've never used spawn-fcgi before. Looks like I have some installing to do. :)
> 
> Will let you know if I can dig anything up.

I can, if need be, distill it down to a single command you can run.  spawn-fcgi returns control to the terminal right away; it is not the issue.  In fact, I have also used it with Rails and PHP.  After it has spawned the Nitro fcgi process, the Nitro process then starts printing data to the screen and has to be interrupted.

--TimH


From bryan.a.soto at gmail.com  Fri Mar 24 14:35:05 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 11:35:05 -0800
Subject: [Nitro] FastCGI startup issues
In-Reply-To: <20060324110551.62b4b7b8.timh@dirtymonday.net>
References: <20060323213025.2d4f2f67.timh@dirtymonday.net>
	<bd93db0d0603241051o7c37f5d2jf009e85b32080b6d@mail.gmail.com>
	<20060324110551.62b4b7b8.timh@dirtymonday.net>
Message-ID: <bd93db0d0603241135g2aa9265m69022f4797234a13@mail.gmail.com>

On 3/24/06, Tim Howe <timh at dirtymonday.net> wrote:
> On Fri, 24 Mar 2006 10:51:49 -0800
> "Bryan Soto" <bryan.a.soto at gmail.com> wrote:
>
> > I've never used spawn-fcgi before. Looks like I have some installing to do. :)
> >
> > Will let you know if I can dig anything up.
>
> I can, if need be, distill it down to a single command you can run.  spawn-fcgi returns control to the terminal right away; it is not the issue.  In fact, I have also used it with Rails and PHP.  After it has spawned the Nitro fcgi process, the Nitro process then starts printing data to the screen and has to be interrupted.
>

Hi, I was poking through runner.rb and noticed it changes behaviour
depending on an environment variable ENV['NITRO_INVOKE']. Does it make
a difference if you set this to 'fgci_proc' in your runner file?

If not, another alternative is to redirect the logger output. I did
something similar for the Og test suite.

require 'glue'
require 'glue/logger'

# This would increase memory consumption but would be okay for testing
Logger.set(Logger.new(StringIO.new))

Let me see about redirecting to a file. I'll get back to you in a bit.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Fri Mar 24 14:42:16 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 11:42:16 -0800
Subject: [Nitro] FastCGI startup issues
In-Reply-To: <bd93db0d0603241135g2aa9265m69022f4797234a13@mail.gmail.com>
References: <20060323213025.2d4f2f67.timh@dirtymonday.net>
	<bd93db0d0603241051o7c37f5d2jf009e85b32080b6d@mail.gmail.com>
	<20060324110551.62b4b7b8.timh@dirtymonday.net>
	<bd93db0d0603241135g2aa9265m69022f4797234a13@mail.gmail.com>
Message-ID: <bd93db0d0603241142u32828530r5a57a2b40c8c1fdf@mail.gmail.com>

On 3/24/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/24/06, Tim Howe <timh at dirtymonday.net> wrote:
> > On Fri, 24 Mar 2006 10:51:49 -0800
> > "Bryan Soto" <bryan.a.soto at gmail.com> wrote:
> >
> > > I've never used spawn-fcgi before. Looks like I have some installing to do. :)
> > >
> > > Will let you know if I can dig anything up.
> >
> > I can, if need be, distill it down to a single command you can run.  spawn-fcgi returns control to the terminal right away; it is not the issue.  In fact, I have also used it with Rails and PHP.  After it has spawned the Nitro fcgi process, the Nitro process then starts printing data to the screen and has to be interrupted.
> >
>
> Hi, I was poking through runner.rb and noticed it changes behaviour
> depending on an environment variable ENV['NITRO_INVOKE']. Does it make
> a difference if you set this to 'fgci_proc' in your runner file?
>
> If not, another alternative is to redirect the logger output. I did
> something similar for the Og test suite.
>
> require 'glue'
> require 'glue/logger'
>
> # This would increase memory consumption but would be okay for testing
> Logger.set(Logger.new(StringIO.new))
>
> Let me see about redirecting to a file. I'll get back to you in a bit.
>

Logger.set(Logger.new(File.open('foo.log', File::WRONLY | File::APPEND)))

Interesting. Logger has some options for log file rotation and aging.
Perhaps Nitro should take advantage of these features? Anyone got some
long running apps that generate large log files?


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Fri Mar 24 15:29:28 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 12:29:28 -0800
Subject: [Nitro] Facets/Og problem.
Message-ID: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>

When running rake test in og:

The offending line in the test suite is:

require 'facets'

If I dig up anything more, I'll let you know.

Backtrace follows:

/usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:59:
warning: already initialized constant EXCLUDE
/usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:96:in
`bind': singleton method called for a different object (TypeError)
        from /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:96:in
`method_added'
        from /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:95
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:21:in
`require'
        from /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/functor.rb:25
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:21:in
`require'
        from /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/core/kernel/as.rb:2
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:21:in
`require'
        from /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets.rb:102
        from /usr/lib/ruby/site_ruby/1.8/rubygems/custom_require.rb:21:in
`require'
        from ./test/og/tc_delete_all.rb:4
        from /usr/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb:5
        from /usr/lib/ruby/gems/1.8/gems/rake-0.7.0/lib/rake/rake_test_loader.rb:5
rake aborted!
Command failed with status (1): [/usr/bin/ruby18 -Ilib:test:/mnt/zip/nitro....]

(See full trace by running task with --trace)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Fri Mar 24 16:41:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 13:41:23 -0800
Subject: [Nitro] Add :table property to join relationships
In-Reply-To: <bd93db0d0603232358s374cc1bwe4c306a4bbf8f240@mail.gmail.com>
References: <BC2D5047-5212-416C-B721-A89818C66950@motionpath.com>
	<200603131126.48485.zimba.tm@gmail.com>
	<DED1373C-1A55-433C-AE7C-719686931C3F@motionpath.com>
	<bd93db0d0603232358s374cc1bwe4c306a4bbf8f240@mail.gmail.com>
Message-ID: <bd93db0d0603241341v5a4ed4bdv5aa8010baa227656@mail.gmail.com>

Applied to nitrohq.

On 3/23/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Wow. Took me long enough. :)
>
> Background for everyone:
> Here
> http://rubyforge.org/pipermail/nitro-general/2006-February/002901.html
> and
> http://rubyforge.org/pipermail/nitro-general/2006-February/002930.html
>
> Barring objections, I'm planning to apply this one tomorrow (well,
> today in a few minutes). Probably not a good idea to commit at
> midnight.
>
> It needs a slight modification to bring it up to date. Had a conflict
> in og/lib/glue/taggable.rb. I'll take care of that as it was up to
> date when it was sent.
>
> Thanks,
>
> Bryan
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Fri Mar 24 18:48:50 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 24 Mar 2006 23:48:50 +0000
Subject: [Nitro] Why should I use Nitro instead of Rails - missing proof
In-Reply-To: <1cdb6083d10033a00d45f045131653d7@oggu.de>
References: <20060324003908.GA10705@mlabdial.hit.bme.hu>
	<bd93db0d0603231734p20721962ya058bde1e37378ea@mail.gmail.com>
	<1cdb6083d10033a00d45f045131653d7@oggu.de>
Message-ID: <4b6f054f0603241548l6ba7d1e8w6c45dd38ee1b9542@mail.gmail.com>

> I made a small research for html2pdf, but could only find pdf-writer
> for ruby, anyone who knows more?

Search for Austin Ziegler's PDF project. I forget what it's called.
But it shouldn't be too hard to track down. You can always ask on
RubyForge too if need be.

T.



From transfire at gmail.com  Fri Mar 24 18:52:06 2006
From: transfire at gmail.com (TRANS)
Date: Fri, 24 Mar 2006 23:52:06 +0000
Subject: [Nitro] Repo Updated.
In-Reply-To: <afd80b540603241004p20b6d8d8wf6d7ce2efa7a0e30@mail.gmail.com>
References: <bd93db0d0603231257k4fbdee9bk8ec73c4d7d5389c@mail.gmail.com>
	<55c107bf0603240601r4e220781ia470d2fbee3aee16@mail.gmail.com>
	<55c107bf0603240615x14a33e73t5ccb4e9faf084ae5@mail.gmail.com>
	<200603241543.39513.zimba.tm@gmail.com>
	<55c107bf0603240659i706cfd58k868212bbe27f61a9@mail.gmail.com>
	<afd80b540603241001g237de873nfc2a8a7fae89506b@mail.gmail.com>
	<afd80b540603241004p20b6d8d8wf6d7ce2efa7a0e30@mail.gmail.com>
Message-ID: <4b6f054f0603241552q5adc4059v5630f3baf567f594@mail.gmail.com>

On 3/24/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > *not* so cool...
>
> ...but, lets try and see how this work out (even though I ve tried
> this in the past and
> it *didn't* work out)

It requires APIs to be fairly stable. If the Nitro/Og/... APIs will
change alot and rapidly so, then it won't work so well. It all depends
on where the projects' are at, really.

T.



From transfire at gmail.com  Fri Mar 24 19:44:04 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 25 Mar 2006 00:44:04 +0000
Subject: [Nitro] Facets/Og problem.
In-Reply-To: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
Message-ID: <4b6f054f0603241644t4357caa2y807d930727a5116d@mail.gmail.com>

Do me a favor and uninstall facets 1.2.0 and then reinstall it and try
it again. I've updated this version with some fixed requires. My tests
now all pass.

Thanks,
T.



From bryan.a.soto at gmail.com  Fri Mar 24 20:02:11 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 17:02:11 -0800
Subject: [Nitro] Facets/Og problem.
In-Reply-To: <4b6f054f0603241644t4357caa2y807d930727a5116d@mail.gmail.com>
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
	<4b6f054f0603241644t4357caa2y807d930727a5116d@mail.gmail.com>
Message-ID: <bd93db0d0603241702h23edab02vb91469c5d37935b9@mail.gmail.com>

On 3/24/06, TRANS <transfire at gmail.com> wrote:
> Do me a favor and uninstall facets 1.2.0 and then reinstall it and try
> it again. I've updated this version with some fixed requires. My tests
> now all pass.

I got the same backtrace. :(

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Fri Mar 24 20:16:47 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 25 Mar 2006 01:16:47 +0000
Subject: [Nitro] Facets/Og problem.
In-Reply-To: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
Message-ID: <4b6f054f0603241716r28e7646byacdee27c1439f5bb@mail.gmail.com>

On 3/24/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> When running rake test in og:
>
> The offending line in the test suite is:
>
> require 'facets'
>
> If I dig up anything more, I'll let you know.
>
> Backtrace follows:
>
> /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:59:
> warning: already initialized constant EXCLUDE

This would indicate that BasicObject is being loaded twice. I wonder
if there is a stray calibre require in there or whether it some sort
of circular reference? BTW uninstall calibre and change all references
to 'calibre/' to either 'facet/' or 'facets/more/'

> /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:96:in
> `bind': singleton method called for a different object (TypeError)

Yea. Somthing stange is going on for sure. But I can't quite place it.

T.



From bryan.a.soto at gmail.com  Fri Mar 24 20:26:11 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 17:26:11 -0800
Subject: [Nitro] Facets/Og problem.
In-Reply-To: <4b6f054f0603241716r28e7646byacdee27c1439f5bb@mail.gmail.com>
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
	<4b6f054f0603241716r28e7646byacdee27c1439f5bb@mail.gmail.com>
Message-ID: <bd93db0d0603241726u6a5b6e8egea3eb8a0afd0fae3@mail.gmail.com>

On 3/24/06, TRANS <transfire at gmail.com> wrote:
> > /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:59:
> > warning: already initialized constant EXCLUDE
>
> This would indicate that BasicObject is being loaded twice. I wonder
> if there is a stray calibre require in there or whether it some sort
> of circular reference? BTW uninstall calibre and change all references
> to 'calibre/' to either 'facet/' or 'facets/more/'
>

Ah, okay. Let me try that too.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From billk at cts.com  Fri Mar 24 20:33:25 2006
From: billk at cts.com (Bill Kelly)
Date: Fri, 24 Mar 2006 17:33:25 -0800
Subject: [Nitro] Facets/Og problem.
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
	<4b6f054f0603241716r28e7646byacdee27c1439f5bb@mail.gmail.com>
Message-ID: <021e01c64fac$1cb17940$6442a8c0@musicbox>

From: "TRANS" <transfire at gmail.com>
>
>> /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:96:in
>> `bind': singleton method called for a different object (TypeError)
> 
> Yea. Somthing stange is going on for sure. But I can't quite place it.

Do you think we should ask Austin to look into it?

<grin>

/me ducks






From bryan.a.soto at gmail.com  Fri Mar 24 20:47:39 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 17:47:39 -0800
Subject: [Nitro] Facets/Og problem.
In-Reply-To: <021e01c64fac$1cb17940$6442a8c0@musicbox>
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
	<4b6f054f0603241716r28e7646byacdee27c1439f5bb@mail.gmail.com>
	<021e01c64fac$1cb17940$6442a8c0@musicbox>
Message-ID: <bd93db0d0603241747u5e13ac0en762dd68f05212403@mail.gmail.com>

On 3/24/06, Bill Kelly <billk at cts.com> wrote:
> From: "TRANS" <transfire at gmail.com>
> >
> >> /usr/lib/ruby/gems/1.8/gems/facets-1.2.0/lib/facets/more/basicobject.rb:96:in
> >> `bind': singleton method called for a different object (TypeError)
> >
> > Yea. Somthing stange is going on for sure. But I can't quite place it.
>
> Do you think we should ask Austin to look into it?
>
> <grin>
>
> /me ducks

Thats a good one. :)

Trans, I'll have to go through and change the requires. For some
reason I can't unpull the patch that changed them. Not sure why. I'll
let you know later how it worked out. Thanks for your help.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Fri Mar 24 20:53:12 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 17:53:12 -0800
Subject: [Nitro] Og test suite, postgres and Sqlite
Message-ID: <bd93db0d0603241753t643f8cduc74dd224e239b5e7@mail.gmail.com>

Apparently I've been living in a mysql fantasy land where the Og test
suite completes. To check Tim's recent patch, I tried to run it and to
my surprise, it died. After that, I thought I'd check out Sqlite as
well and got similar results. :(

I'm okay with Sqlite but know nothiing about Postgres. Is there
someone who'd be interested in helping me get the test suites actually
finishing for these stores?

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Fri Mar 24 20:57:58 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 25 Mar 2006 01:57:58 +0000
Subject: [Nitro] Facets/Og problem.
In-Reply-To: <bd93db0d0603241747u5e13ac0en762dd68f05212403@mail.gmail.com>
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
	<4b6f054f0603241716r28e7646byacdee27c1439f5bb@mail.gmail.com>
	<021e01c64fac$1cb17940$6442a8c0@musicbox>
	<bd93db0d0603241747u5e13ac0en762dd68f05212403@mail.gmail.com>
Message-ID: <4b6f054f0603241757q8826e35p21038cbf010e1436@mail.gmail.com>

A bit brutish but this helps:

  perl -pi -w -e 's/calibre\//facet\//g;' *.rb

Be sure to have a backup first.

T.



From bryan.a.soto at gmail.com  Sat Mar 25 01:54:38 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 22:54:38 -0800
Subject: [Nitro] Facets/Og problem.
In-Reply-To: <4b6f054f0603241757q8826e35p21038cbf010e1436@mail.gmail.com>
References: <bd93db0d0603241229l62fefdb5l872d5b5daf37814b@mail.gmail.com>
	<4b6f054f0603241716r28e7646byacdee27c1439f5bb@mail.gmail.com>
	<021e01c64fac$1cb17940$6442a8c0@musicbox>
	<bd93db0d0603241747u5e13ac0en762dd68f05212403@mail.gmail.com>
	<4b6f054f0603241757q8826e35p21038cbf010e1436@mail.gmail.com>
Message-ID: <bd93db0d0603242254p17601758pb0ba14dc8ce6af1@mail.gmail.com>

On 3/24/06, TRANS <transfire at gmail.com> wrote:
> A bit brutish but this helps:
>
>   perl -pi -w -e 's/calibre\//facet\//g;' *.rb
>
> Be sure to have a backup first.

It just gave me an excuse to figure out macros in Emacs. ;)

Changing the calls to calibre did it. So that must have been your
second loading of BasicObject. I don't know if you need to track that
down or not. Here's the requires to calibre that were made. Just in
case.

Thanks for your help. :)

flare/src/controller.rb:require 'calibre/times'
glue/lib/glue/aspects.rb:require 'calibre/inheritor'
glue/lib/glue/fixture.rb:require 'calibre/ormsupport'
glue/lib/glue/configuration.rb:require 'calibre/synchash'
glue/lib/glue/configuration.rb:require 'calibre/dictionary'
glue/lib/glue/accumulate.rb:require 'calibre/functor'
glue/lib/glue/markup.rb:require 'calibre/classinherit'
glue/lib/glue/property.rb:require 'calibre/annotation'
glue/lib/glue/property.rb:require 'calibre/inheritor'
glue/lib/glue/property.rb:require 'calibre/dictionary'
glue/lib/glue/cache/memory.rb:require 'calibre/synchash'
glue/lib/glue.rb:require 'calibre/nullclass'
glue-cache-file-fix:[Nitro + facets 1.1.0 + calibre 1.2.0
nitro/lib/nitro/compiler/errors.rb:require 'calibre/ormsupport'
nitro/lib/nitro/compiler/morphing.rb:require 'calibre/dictionary'
nitro/lib/nitro/element.rb:require 'calibre/annotation'
nitro/lib/nitro/server/drb.rb:require 'calibre/synchash'
nitro/lib/nitro/helper/javascript/morphing.rb:require 'calibre/dictionary'
nitro/lib/nitro/controller.rb:require 'calibre/annotation'
nitro/lib/nitro/controller.rb:require 'calibre/inheritor'
nitro/lib/nitro/session.rb:require 'calibre/synchash'
nitro/lib/nitro/session.rb:require 'calibre/times'
nitro/lib/glue/webfile.rb:require 'calibre/inheritor'
nitro/test/nitro/tc_render.rb:require 'calibre/mock'
og/lib/og/entity.rb:require 'calibre/classinherit'
og/lib/og/manager.rb:require 'calibre/pool'
og/lib/og/relation/joins_many.rb:require 'calibre/ormsupport'
og/lib/og/relation/has_many.rb:require 'calibre/ormsupport'
og/lib/og/relation.rb:require 'calibre/ormsupport'
og/lib/og/relation.rb:require 'calibre/inheritor'
og/lib/og/relation.rb:require 'calibre/annotation'
og/lib/og/test/testcase.rb:require 'calibre/ormsupport'
og/lib/glue/hierarchical.rb:require 'calibre/ormsupport'
og/lib/glue/cacheable.rb:require 'calibre/classinherit'
og/lib/og.rb:require 'calibre/synchash'
og/lib/og.rb:require 'calibre/syncarray'

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Sat Mar 25 02:00:30 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Fri, 24 Mar 2006 23:00:30 -0800
Subject: [Nitro] Repo updated.
Message-ID: <bd93db0d0603242300l6e9f48ffv4730667f808abfbc@mail.gmail.com>

I've updated devlab.oree.ch/darcs/nitrohq for Facets-1.2.0.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Sat Mar 25 03:06:38 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Sat, 25 Mar 2006 00:06:38 -0800
Subject: [Nitro] PATCH: adds params to AJAX helper generated functions.
Message-ID: <bd93db0d0603250006l33aa4112y509a9102251c0ce6@mail.gmail.com>

Adds code submitted via mailing list and a testcase.

See http://rubyforge.org/pipermail/nitro-general/2006-March/003453.html
for more info.

I wanted to get some more opinions on this one.

Thanks.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ajax-helper-params.zip
Type: application/zip
Size: 13556 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060325/0503118c/attachment.zip 

From m.fellinger at gmail.com  Sat Mar 25 03:27:51 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Sat, 25 Mar 2006 17:27:51 +0900
Subject: [Nitro] FastCGI startup issues
In-Reply-To: <bd93db0d0603241142u32828530r5a57a2b40c8c1fdf@mail.gmail.com>
References: <20060323213025.2d4f2f67.timh@dirtymonday.net>
	<bd93db0d0603241051o7c37f5d2jf009e85b32080b6d@mail.gmail.com>
	<20060324110551.62b4b7b8.timh@dirtymonday.net>
	<bd93db0d0603241135g2aa9265m69022f4797234a13@mail.gmail.com>
	<bd93db0d0603241142u32828530r5a57a2b40c8c1fdf@mail.gmail.com>
Message-ID: <9c00d3e00603250027n7403cc38r44302ae7e8b7ffe7@mail.gmail.com>

yeap... just had some problems with a 15Gig-logfile ;)thx for that info...
On 3/25/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:> On 3/24/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:> > On 3/24/06, Tim Howe <timh at dirtymonday.net> wrote:> > > On Fri, 24 Mar 2006 10:51:49 -0800> > > "Bryan Soto" <bryan.a.soto at gmail.com> wrote:> > >> > > > I've never used spawn-fcgi before. Looks like I have some installing to do. :)> > > >> > > > Will let you know if I can dig anything up.> > >> > > I can, if need be, distill it down to a single command you can run.  spawn-fcgi returns control to the terminal right away; it is not the issue.  In fact, I have also used it with Rails and PHP.  After it has spawned the Nitro fcgi process, the Nitro process then starts printing data to the screen and has to be interrupted.> > >> >> > Hi, I was poking through runner.rb and noticed it changes behaviour> > depending on an environment variable ENV['NITRO_INVOKE']. Does it make> > a difference if you set this to 'fgci_proc' in your runner file?> >> > If not, another alternative is to redirect the logger output. I did> > something similar for the Og test suite.> >> > require 'glue'> > require 'glue/logger'> >> > # This would increase memory consumption but would be okay for testing> > Logger.set(Logger.new(StringIO.new))> >> > Let me see about redirecting to a file. I'll get back to you in a bit.> >>> Logger.set(Logger.new(File.open('foo.log', File::WRONLY | File::APPEND)))>> Interesting. Logger has some options for log file rotation and aging.> Perhaps Nitro should take advantage of these features? Anyone got some> long running apps that generate large log files?>>> --> "Never tell people how to do things. Tell them what to do and they> will surprise you with their ingenuity." ?General George S. Patton>> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From m.fellinger at gmail.com  Sat Mar 25 03:29:55 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Sat, 25 Mar 2006 17:29:55 +0900
Subject: [Nitro] Why should I use Nitro instead of Rails - missing proof
In-Reply-To: <4b6f054f0603241548l6ba7d1e8w6c45dd38ee1b9542@mail.gmail.com>
References: <20060324003908.GA10705@mlabdial.hit.bme.hu>
	<bd93db0d0603231734p20721962ya058bde1e37378ea@mail.gmail.com>
	<1cdb6083d10033a00d45f045131653d7@oggu.de>
	<4b6f054f0603241548l6ba7d1e8w6c45dd38ee1b9542@mail.gmail.com>
Message-ID: <9c00d3e00603250029v7fc272adi3d85d28329cda25b@mail.gmail.com>

I did some research on that, but there seems to be no ruby-solutionfor that so far, neither html2pdf nor markdown/textile2pdf... but ithink some of the rails-guys have something in mind, however, askingon ruby-lang should bring up something nice...
On 3/25/06, TRANS <transfire at gmail.com> wrote:> > I made a small research for html2pdf, but could only find pdf-writer> > for ruby, anyone who knows more?>> Search for Austin Ziegler's PDF project. I forget what it's called.> But it shouldn't be too hard to track down. You can always ask on> RubyForge too if need be.>> T.>> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From nissl at tiscali.it  Sat Mar 25 04:32:40 2006
From: nissl at tiscali.it (Massimo Maria Ghisalberti)
Date: Sat, 25 Mar 2006 10:32:40 +0100
Subject: [Nitro] PATCH: adds params to AJAX helper generated functions.
In-Reply-To: <bd93db0d0603250006l33aa4112y509a9102251c0ce6@mail.gmail.com>
References: <bd93db0d0603250006l33aa4112y509a9102251c0ce6@mail.gmail.com>
Message-ID: <1143279160.3992.15.camel@nissl.mammuth>

in my opinion, but this patch is mine ;)

it is useful because I need to have the id of the element that send
click event (for dispatching to a single page method handler), but in
general a javascript event handler will must to have an object event as
first parameter, in special case for mozilla browser where the object
event is internal to event handler, not is the case of IE where exist a
global window.event.

the right js generated will must be (pseudo):

event_handler_function(event, param){
	
	var Ev;
	if (window.event)
		Ev = event = window.event
	else
		Ev = event 
	...
	handler code
	...
}

in the single param you can pass any sort of data... simple type or
complete object, obviously is you responsability to handle this right.

ciao

Massimo

Il giorno sab, 25/03/2006 alle 00.06 -0800, Bryan Soto ha scritto:
> Adds code submitted via mailing list and a testcase.
> 
> See http://rubyforge.org/pipermail/nitro-general/2006-March/003453.html
> for more info.
> 
> I wanted to get some more opinions on this one.
> 
> Thanks.
> 
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
-- 
Massimo Maria Ghisalberti <nissl at tiscali.it>




From transfire at gmail.com  Sat Mar 25 08:13:56 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 25 Mar 2006 13:13:56 +0000
Subject: [Nitro] Glue Undo
Message-ID: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>

I think the best course of action with Glue is to move it's libs
directly to Nitro (with the obvious things related to Og moved to Og,
like property). In other words:

  glue/lib/glue  -> nitro/lib/nitro/glue

Aso nitro has some glue stuff already which needs to be moved in a level.

  nitro/lib/glue/*  -> nitro/lib/nitro/glue/

Moving this stuff will get us past the glue dependency right away, and
from there we can start moving the generic libs to Facets or move them
elsewhere in nitro/og if they are nitro/og specific, as we have time,

(glue/lib/html will also have to be moved but I'm not sure why that's
out in lib space anyway rather then in a lib directory)

Here's a run down of some of the things I've have done so far for "Glue2Facets".

* I've moved aspect.rb to Facets (I assume this is completely generic,
please tell me if not).

* I've moved accumulate.rb to Facets Enumberable.

  require 'facet/enumerable/accumulate'

(seems like a neat method btw!)

* I moved on_included.rb to Facets Module:

  require 'facet/module/on_included'

* And Facets already has expirable.rb.

The cooresponding requires will need to be changed in the rest of the
system. I will release a 1.2.1 version of Facets on Monday with these
changes and  probably a few more.

Some notes about the things in Glue:

* Flexob isn't needed. It is essentially an OpenStruct with a few
extra methods.  Facets provides means for doing the same things with
OpenStruct itself. So this should be easily replaced with an
OpenStruct. (Or use Facets own OpenObject).
So somtime after Monday it would be good if someone could address
this. Here's where it is used:

  glue/lib/glue/configuration.rb:require 'glue/flexob'
  glue/lib/glue/template.rb:require 'glue/flexob'

The lib template.rb, I recall, is too tightly tied to Nitro. Not sure
about configuration.rb.

* I think sanatize.rb should be a module method of some web/html
module. Then if you want a String method, in the same file do:

  class String
    def sanitize
      SomeHTMLModule.sanitize(self)
    end
  end

Okay, that's all I have time for today. I'm going to Disney World for
the day. Be back tomorrow.

T.



From transfire at gmail.com  Sat Mar 25 08:28:40 2006
From: transfire at gmail.com (TRANS)
Date: Sat, 25 Mar 2006 13:28:40 +0000
Subject: [Nitro] Repo updated.
In-Reply-To: <bd93db0d0603242300l6e9f48ffv4730667f808abfbc@mail.gmail.com>
References: <bd93db0d0603242300l6e9f48ffv4730667f808abfbc@mail.gmail.com>
Message-ID: <4b6f054f0603250528p52d164ecued4a3b09c290fa86@mail.gmail.com>

Cool. I'll have it ready to go in this week sometime. Lets get past
the Glue dependency.

T.



From james_b at neurogami.com  Sat Mar 25 22:20:03 2006
From: james_b at neurogami.com (James Britt)
Date: Sat, 25 Mar 2006 20:20:03 -0700
Subject: [Nitro] Dynamic Settings
Message-ID: <44260863.5080501@neurogami.com>

I've an application that needs to run with varying settings depending on 
the environment.  For example, locally I may want to use sqlite, but in 
production I may prefer mysql. Similarly,  I have multiple instances of 
the same app running on a development server, each needing a its own 
scgi port.

I can think of various ways to stop writing the gnarly conditionals that 
have begun littering run.rb, but before I head off on something, what do 
other people do?  Is there a Nitro Way to assign Og and Server settings 
based on some environment setting or command line argument?


Thanks!

-- 
James Britt

"In physics the truth is rarely perfectly clear, and that is certainly
  universally the case in human affairs. Hence, what is not surrounded by
  uncertainty cannot be the truth."
  - R. Feynman



From news at stephan.walter.name  Sun Mar 26 09:18:00 2006
From: news at stephan.walter.name (Stephan Walter)
Date: Sun, 26 Mar 2006 16:18:00 +0200
Subject: [Nitro] form controls
Message-ID: <pan.2006.03.26.14.17.53.96319@stephan.walter.name>

Hi,

I'm new to Ruby and Nitro, and while writing a web site I found the
HTML form helper not so flexible (or maybe I just misunderstood it?
Documentation is unfortunately a bit sparse). Currently you have the
choice of either letting Nitro do everything by calling form_for or
controls_for. But I wanted only a subset of the properties and relations
to be shown. So I wrote these two functions:

  def control_for_relation(obj, target, options={})
    rel = obj.class.relations.find { |r| r[:name] == target }
    FormBuilder.element(rel, Form::Control.fetch(obj, rel, options).render)
  end                                                                               

  def control_for_property(obj, prop, options={})
    pr = obj.class.properties[prop]                                           
    FormBuilder.element(pr, Form::Control.fetch(obj, pr, options).render)     
  end

Using these functions you can write in your templates, for example:

  #{ control_for_property(user, :name) }
  #{ control_for_relation(user, :tags) }

A few questions: These are currently in my controller.rb, but this might
not be the right place. They could of course be incorporated into
lib/nitro/helper/form.rb, but what do the experts think of my
implementation? (I think calling Array#find in the first function is
clumsy, is there a better way?)

Besides that, are there any plans to support error messages in the form
builder? e.g. if you forgot to fill in a field, it will make a red border
across that field.

Greetings,
Stephan





From transfire at gmail.com  Sun Mar 26 12:08:38 2006
From: transfire at gmail.com (TRANS)
Date: Sun, 26 Mar 2006 12:08:38 -0500
Subject: [Nitro] Glue move
Message-ID: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>

I did a little more work on working Glue out. But as a proceeded I
realized I probably would not be able to just submit a patch for this
given that I pulled down George's repo then applied all the patches
from the other development repo; plus the fact the it is such a major
change.

It seems like it will difficult to incorporate such a major change
into two seperate repos. How should we proceed on this?

Outline of what I'm doing:

  move glue/lib/glue to nitro/lib/nitro/glue
  move nitro/lib/glue/* to nitro/lib/nitro/glue
  move glue/lib/html to nitro/lib/html
  merge glue/lib/glue.rb into nitro/lib/nitro.rb (with some imporvements)
  update all realted requires

>From there I can migrate certain glue libs to facets.

T.



From fabian at oggu.de  Sun Mar 26 15:37:00 2006
From: fabian at oggu.de (Fabian Buch)
Date: Sun, 26 Mar 2006 22:37:00 +0200
Subject: [Nitro] Dynamic Settings
In-Reply-To: <44260863.5080501@neurogami.com>
References: <44260863.5080501@neurogami.com>
Message-ID: <509369bc93828575beeb9fea6fab30e7@oggu.de>


Am 26.03.2006 um 05:20 schrieb James Britt:
> I can think of various ways to stop writing the gnarly conditionals 
> that
> have begun littering run.rb, but before I head off on something, what 
> do
> other people do?  Is there a Nitro Way to assign Og and Server settings
> based on some environment setting or command line argument?

The blog example in the nitro-examples package uses a conf directory 
for various settings both in ruby and yaml. Dunno if that does exactly 
what you want. I didn't use it myself yet, but wanted to have a closer 
look at this somewhen.

Apart from that there are some environment settings. One that comes to 
my mind is $DBG (true, false), anotherone $NITRO_INVOKE (proc_cgi, 
proc_scgi, proc_fcgi) and $NITRO_MODE (debug, live).

Fabian



From kgergely at mcl.hu  Sun Mar 26 17:20:39 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Mon, 27 Mar 2006 00:20:39 +0200
Subject: [Nitro] form controls
In-Reply-To: <pan.2006.03.26.14.17.53.96319@stephan.walter.name>
References: <pan.2006.03.26.14.17.53.96319@stephan.walter.name>
Message-ID: <20060326222039.GC3704@mlabdial.hit.bme.hu>

Stephan Walter azt ?rta, hogy: 
> Hi,
Hi!

I also run into this problem!

The problems/questions:

Can I simply build a form, which is not database related? (you can store 
info eg. in your session).

What the hell options can be?

> Besides that, are there any plans to support error messages in the form
> builder? e.g. if you forgot to fill in a field, it will make a red border
> across that field.
About the error handling:
It is matter of taste what and how do you signal errors.
Somehow you should be able to tell to nitro how would you like to signal 
errors. (borders, css styles, alternative error messages/html fragments, 
which are rendered as title, rendered after the element, or go to some 
variable, so, you can print all error messages at one place in your 
program.

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060326/2efa6afb/attachment.bin 

From lionel.orry at gmail.com  Mon Mar 27 04:15:53 2006
From: lionel.orry at gmail.com (Lionel Orry)
Date: Mon, 27 Mar 2006 11:15:53 +0200
Subject: [Nitro] Why should I use Nitro instead of Rails - missing proof
In-Reply-To: <4b6f054f0603241548l6ba7d1e8w6c45dd38ee1b9542@mail.gmail.com>
References: <20060324003908.GA10705@mlabdial.hit.bme.hu>	<bd93db0d0603231734p20721962ya058bde1e37378ea@mail.gmail.com>	<1cdb6083d10033a00d45f045131653d7@oggu.de>
	<4b6f054f0603241548l6ba7d1e8w6c45dd38ee1b9542@mail.gmail.com>
Message-ID: <4427AD49.1080302@gmail.com>



TRANS wrote:
>> I made a small research for html2pdf, but could only find pdf-writer
>> for ruby, anyone who knows more?

I guess Rails devs could help a bit :

http://wiki.rubyonrails.com/rails/pages/HowtoGeneratePDFs

> 
> Search for Austin Ziegler's PDF project. I forget what it's called.
> But it shouldn't be too hard to track down. You can always ask on
> RubyForge too if need be.
In fact, Austin's PDF project is PDF::Writer, which is referred to in RAA by 'pdf-writer'. The homepage is 
http://ruby-pdf.rubyforge.org/pdf-writer/


> 
> T.
> 
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
> 

cheers,

Lionel



From guillaume.pierronnet at gmail.com  Mon Mar 27 05:51:05 2006
From: guillaume.pierronnet at gmail.com (guillaume pierronnet)
Date: Mon, 27 Mar 2006 12:51:05 +0200
Subject: [Nitro] nested templates calls and shared data
In-Reply-To: <bd93db0d0603221622y66313a58mf363b22c6bb4e1f5@mail.gmail.com>
References: <6a7d49ca0603150715o7a13c22br@mail.gmail.com>
	<bd93db0d0603221622y66313a58mf363b22c6bb4e1f5@mail.gmail.com>
Message-ID: <6a7d49ca0603270251i26b502f9n@mail.gmail.com>

thanks, i was stuck on the idea that script must go just before the
</body> tag, but your solution looks good!

i implemented it as a patch.




2006/3/23, Bryan Soto <bryan.a.soto at gmail.com>:
> On 3/15/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > i spent all my week-end on trying to implement a sort of 'global
> > compiling pipeline', with no success.
> >
> > does anybody have a idea on how to fix it ?
>
> Hi Guill,
>
> Check out nitro/lib/nitro/compiler/script.rb:90
>
> Currently, the script compiler looks for a </body> tag and writes the
> script out at the end of it. As a "fugly hack", you can add
>
> <!-- </body> -->
>
> to the end of your album_list.xhtml file.
>
> Perhaps a real fix would be to check if text =~ /<\/body>/ and,
> failing that, just appending the the script at the end of text? What
> do you think?
>
> Bryan
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.gz
Type: application/x-gzip
Size: 12713 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060327/09b444ec/attachment.gz 

From james_b at neurogami.com  Mon Mar 27 10:14:04 2006
From: james_b at neurogami.com (James Britt)
Date: Mon, 27 Mar 2006 08:14:04 -0700
Subject: [Nitro] Dynamic Settings
In-Reply-To: <509369bc93828575beeb9fea6fab30e7@oggu.de>
References: <44260863.5080501@neurogami.com>
	<509369bc93828575beeb9fea6fab30e7@oggu.de>
Message-ID: <4428013C.8040309@neurogami.com>

Fabian Buch wrote:
> Am 26.03.2006 um 05:20 schrieb James Britt:
> 
>>I can think of various ways to stop writing the gnarly conditionals 
>>that
>>have begun littering run.rb, but before I head off on something, what 
>>do
>>other people do?  Is there a Nitro Way to assign Og and Server settings
>>based on some environment setting or command line argument?
> 
> 
> The blog example in the nitro-examples package uses a conf directory 
> for various settings both in ruby and yaml. Dunno if that does exactly 
> what you want. I didn't use it myself yet, but wanted to have a closer 
> look at this somewhen.

It's similar to what I've put together for myself.  My version checks 
for ENV[ 'NITRO_ENV' ] and loads different settings  based on the value.



Thanks.
-- 
James Britt

http://www.ruby-doc.org       - Ruby Help & Documentation
http://www.artima.com/rubycs/ - The Journal By & For Rubyists
http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://www.jamesbritt.com     - Playing with Better Toys
http://www.30secondrule.com   - Building Better Tools


From george.moschovitis at gmail.com  Mon Mar 27 10:58:07 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 18:58:07 +0300
Subject: [Nitro] Dynamic Settings
In-Reply-To: <4428013C.8040309@neurogami.com>
References: <44260863.5080501@neurogami.com>
	<509369bc93828575beeb9fea6fab30e7@oggu.de>
	<4428013C.8040309@neurogami.com>
Message-ID: <afd80b540603270758h446af406rcea21c7ce0fe33ff@mail.gmail.com>

> It's similar to what I've put together for myself.  My version checks
> for ENV[ 'NITRO_ENV' ] and loads different settings  based on the value.

you can have different settings for debug, stage and live/production. Run with

ruby run.rb --debug/--stage/--production etc.

have a look at runner.rb for details.

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar 27 11:54:25 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 19:54:25 +0300
Subject: [Nitro] Glue Undo
In-Reply-To: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
References: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
Message-ID: <afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>

>   glue/lib/glue  -> nitro/lib/nitro/glue

how about removing glue altogether? and use something like:

lib/nitro/util

instead?


> (glue/lib/html will also have to be moved but I'm not sure why that's
> out in lib space anyway rather then in a lib directory)

this is an external library (ie not coded by me)

> * I've moved aspect.rb to Facets (I assume this is completely generic,
> please tell me if not).

it tries to be generic. BTW I would like to use some of your aop ideas
in the future.

> Some notes about the things in Glue:
>
> * Flexob isn't needed. It is essentially an OpenStruct with a few

flexob was supposed to be an optimized version of OpenStruct. I think
there i such a utility in facets (called OpenObject or something?)

> The lib template.rb, I recall, is too tightly tied to Nitro. Not sure
> about configuration.rb.

Configuration is general, wait for my new configuration patch though.


-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From james_b at neurogami.com  Mon Mar 27 11:59:58 2006
From: james_b at neurogami.com (James Britt)
Date: Mon, 27 Mar 2006 09:59:58 -0700
Subject: [Nitro] Dynamic Settings
In-Reply-To: <afd80b540603270758h446af406rcea21c7ce0fe33ff@mail.gmail.com>
References: <44260863.5080501@neurogami.com>	<509369bc93828575beeb9fea6fab30e7@oggu.de>	<4428013C.8040309@neurogami.com>
	<afd80b540603270758h446af406rcea21c7ce0fe33ff@mail.gmail.com>
Message-ID: <44281A0E.9020201@neurogami.com>

George Moschovitis wrote:
>>It's similar to what I've put together for myself.  My version checks
>>for ENV[ 'NITRO_ENV' ] and loads different settings  based on the value.
> 
> 
> you can have different settings for debug, stage and live/production. Run with
> 
> ruby run.rb --debug/--stage/--production etc.
> 
> have a look at runner.rb for details.

Well, this is not encouraging:

#--
# FIXME: Rename/Reimplement this class.
#++


:)

But, more important, what I don't see is how I tell the runner that I 
want --foo_mode or --development2 or any arbitrary environment setting 
and have it load a set of configuration values specific to that environment.

The method load_external_configuration will load a config file, but the 
setup_options method seems to be restricted to a particular set of 
run-mode options.

So I added an option to runner.rb

   opts.on('--config CONF', 'Load  config file') do |conf|
     load_external_configuration conf
   end


I noticed that load_external_configuration just quietly falls through if 
there is no matching config file.  Is this useful?  I see, for example, 
that, setup_debug does some configuring, then calls out to 
load_external_configuration; I gather in that case a missing debug.* 
config file is OK.  But, for my purposes, I'd want an exception if I 
specifically asked for a config file and it was not found.

Right now I've added this to 'load_external_configuration'

   unless self.class.mode.to_s =~ /^(debug|stage|live)$/
     raise "No matching config file!" unless  have_config?( mode )
   end

and this helper method:

def have_config? mode
  File.exist?( "conf/#{mode}.rb" ) || File.exist?( "conf/#{mode}.yml" )
end

Not happy with the hard-coded set of special-case modes, though.

Thanks,




-- 
James Britt

http://www.rubystuff.com      - The Ruby Store for Ruby Stuff
http://www.jamesbritt.com     - Playing with Better Toys
http://www.30secondrule.com   - Building Better Tools


From george.moschovitis at gmail.com  Mon Mar 27 12:44:45 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 20:44:45 +0300
Subject: [Nitro] Repo optimization
Message-ID: <afd80b540603270944v7a02b906qfc75c3b42e7b62e4@mail.gmail.com>

Dear devs,

I just did a darcs get and had to wait for downloading >500 patches.
Does anyone know how we can optimize the darcs repo (ie, merge the
patches or something).

thanks,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar 27 12:47:48 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 20:47:48 +0300
Subject: [Nitro] nested templates calls and shared data
In-Reply-To: <6a7d49ca0603270251i26b502f9n@mail.gmail.com>
References: <6a7d49ca0603150715o7a13c22br@mail.gmail.com>
	<bd93db0d0603221622y66313a58mf363b22c6bb4e1f5@mail.gmail.com>
	<6a7d49ca0603270251i26b502f9n@mail.gmail.com>
Message-ID: <afd80b540603270947i3866d13aobab1120c99b83a55@mail.gmail.com>

I think we could work out a better implementation, that keeps all
javascript in one block.

-g.

On 3/27/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> thanks, i was stuck on the idea that script must go just before the
> </body> tag, but your solution looks good!
>
> i implemented it as a patch.
>
>
>
>
> 2006/3/23, Bryan Soto <bryan.a.soto at gmail.com>:
> > On 3/15/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > > i spent all my week-end on trying to implement a sort of 'global
> > > compiling pipeline', with no success.
> > >
> > > does anybody have a idea on how to fix it ?
> >
> > Hi Guill,
> >
> > Check out nitro/lib/nitro/compiler/script.rb:90
> >
> > Currently, the script compiler looks for a </body> tag and writes the
> > script out at the end of it. As a "fugly hack", you can add
> >
> > <!-- </body> -->
> >
> > to the end of your album_list.xhtml file.
> >
> > Perhaps a real fix would be to check if text =~ /<\/body>/ and,
> > failing that, just appending the the script at the end of text? What
> > do you think?
> >
> > Bryan
> >
> > --
> > "Never tell people how to do things. Tell them what to do and they
> > will surprise you with their ingenuity." ?General George S. Patton
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From kashia at vfemail.net  Mon Mar 27 12:51:52 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Mon, 27 Mar 2006 19:51:52 +0200
Subject: [Nitro] Repo optimization
In-Reply-To: <afd80b540603270944v7a02b906qfc75c3b42e7b62e4@mail.gmail.com>
References: <afd80b540603270944v7a02b906qfc75c3b42e7b62e4@mail.gmail.com>
Message-ID: <op.s622kqkvg3glhz@jo.local>

Hi,

> I just did a darcs get and had to wait for downloading >500 patches.
> Does anyone know how we can optimize the darcs repo (ie, merge the
> patches or something).

http://darcs.net/DarcsWiki/BestPractices

$ darcs --tag 0.30 (for example)
$ darcs optimize --checkpoint

and then everyone can do:

$ darcs get --partial

which then gets only the patches since the checkpoint

"darcs get" will still get the whole history though

hope that helps

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From george.moschovitis at gmail.com  Mon Mar 27 12:56:24 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 20:56:24 +0300
Subject: [Nitro] Repo optimization
In-Reply-To: <op.s622kqkvg3glhz@jo.local>
References: <afd80b540603270944v7a02b906qfc75c3b42e7b62e4@mail.gmail.com>
	<op.s622kqkvg3glhz@jo.local>
Message-ID: <afd80b540603270956x45b4afe9w5960368af807b2a0@mail.gmail.com>

> $ darcs get --partial

Ah, you need that --partial thing.

I remember I tagged 0.28.0

thanks,
George.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From kgergely at mcl.hu  Mon Mar 27 13:25:55 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Mon, 27 Mar 2006 20:25:55 +0200
Subject: [Nitro] Server.map and paths?
Message-ID: <20060327182555.GA11490@mlabdial.hit.bme.hu>

Hi!

I'd like to have a simple authentication.
I tried this:

.. (up a directory)
/home/kgergely/nitro/ {{{
|-- auth/ {{{
|   |-- controller.rb
|   |-- model.rb
|   `-- template/ {{{
|       `-- index.xhtml }}} }}}

|-- controller.rb
|-- model.rb
|-- run.rb
`-- template/ }}}

In auth/controller.rb there is:

class AuthController < Nitro::Controller
  e= File.join File.dirname(__FILE__), 'template'
  puts "Auth template is: #{e}"
  Template.root = e
 # ...
end

It prints the correct answer, and should set Template.root

Auth template is: ./auth/template

But, as I navigate to myserver/auth

I get:
D, [2006-03-27T20:26:36.944828 #17065] DEBUG -- : Rendering '/auth'.
E, [2006-03-27T20:26:36.945743 #17065] ERROR -- : No action for path '/auth' on 'AuthController'
D, [2006-03-27T20:26:36.945901 #17065] DEBUG -- : Rendering '/error'.
D, [2006-03-27T20:26:36.946601 #17065] DEBUG -- : Compiling action 'MyController#error'
D, [2006-03-27T20:26:36.947835 #17065] DEBUG -- : Compiling template 'MyController: /usr/lib/ruby/gems/1.8/gems/nitro-0.29.0/lib/../proto/public/error.xhtml'
D, [2006-03-27T20:26:36.989252 #17065] DEBUG -- : Rendering '/error'.

However my run.rb looks like:
Nitro::Server.map = {
  '/auth' => AuthController,
  '/' => MyController # keep at end!
}
Nitro.start

If I define a method index in my AuthController, the result of that function is rendered, but the template is never found :(

What am I doing wrong?

thx
Gergo
--
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060327/ac72f924/attachment.bin 

From transfire at gmail.com  Mon Mar 27 13:54:14 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 13:54:14 -0500
Subject: [Nitro] Glue Undo
In-Reply-To: <afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>
References: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
	<afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>
Message-ID: <4b6f054f0603271054k427b9427v622c46509c6fb21f@mail.gmail.com>

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> >   glue/lib/glue  -> nitro/lib/nitro/glue
>
> how about removing glue altogether? and use something like:
>
> lib/nitro/util
>
> instead?

That's a great idea!

> > (glue/lib/html will also have to be moved but I'm not sure why that's
> > out in lib space anyway rather then in a lib directory)
>
> this is an external library (ie not coded by me)

I see.

> > * I've moved aspect.rb to Facets (I assume this is completely generic,
> > please tell me if not).
>
> it tries to be generic. BTW I would like to use some of your aop ideas
> in the future.

That would be great. I have an implementation of cuts in pure Ruby
btw. I just never realeased it b/c it not as ideal as a tru
implementation.

> > Some notes about the things in Glue:
> >
> > * Flexob isn't needed. It is essentially an OpenStruct with a few
>
> flexob was supposed to be an optimized version of OpenStruct. I think
> there i such a utility in facets (called OpenObject or something?)

Yes. OpenObject would do the trick if you prefer that to OpenStruct
--and in fact in my limited test OpenObject was faseter thean
OpenStruct (to my surprise).

> > The lib template.rb, I recall, is too tightly tied to Nitro. Not sure
> > about configuration.rb.
>
> Configuration is general, wait for my new configuration patch though.

Well, I'm completely waiting ;-) b/c I think you will have to move the
glue files over and figure out how that will sync with the other repo.
Once you do that then I can transition certain libs over to Facets.

T.



From george.moschovitis at gmail.com  Mon Mar 27 14:04:40 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 22:04:40 +0300
Subject: [Nitro] Glue Undo
In-Reply-To: <4b6f054f0603271054k427b9427v622c46509c6fb21f@mail.gmail.com>
References: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
	<afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>
	<4b6f054f0603271054k427b9427v622c46509c6fb21f@mail.gmail.com>
Message-ID: <afd80b540603271104x37fdcb40nad2505023db733e5@mail.gmail.com>

> That would be great. I have an implementation of cuts in pure Ruby
> btw. I just never realeased it b/c it not as ideal as a tru
> implementation.

yeah, i ve seen that in the tarball you have sent me some time ago.

> Yes. OpenObject would do the trick if you prefer that to OpenStruct
> --and in fact in my limited test OpenObject was faseter thean
> OpenStruct (to my surprise).

ok perhaps you can replace flexob with OpenObject.



-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From bryan.a.soto at gmail.com  Mon Mar 27 14:04:47 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 11:04:47 -0800
Subject: [Nitro] Repo optimization
In-Reply-To: <op.s622kqkvg3glhz@jo.local>
References: <afd80b540603270944v7a02b906qfc75c3b42e7b62e4@mail.gmail.com>
	<op.s622kqkvg3glhz@jo.local>
Message-ID: <bd93db0d0603271104y535a8c61w202b274d846715a5@mail.gmail.com>

That's funny. I was going to ask this one myself.

On 3/27/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hi,
>
> > I just did a darcs get and had to wait for downloading >500 patches.
> > Does anyone know how we can optimize the darcs repo (ie, merge the
> > patches or something).
>
> http://darcs.net/DarcsWiki/BestPractices
>
> $ darcs --tag 0.30 (for example)
> $ darcs optimize --checkpoint
>
> and then everyone can do:
>
> $ darcs get --partial
>
> which then gets only the patches since the checkpoint
>
> "darcs get" will still get the whole history though
>
> hope that helps

Definitely. Thanks!

>
> Kashia
>
> --
> Feel the love
> http://pinkjuice.com/pics/ruby.png
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From george.moschovitis at gmail.com  Mon Mar 27 14:06:42 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 22:06:42 +0300
Subject: [Nitro] [PATCH] Configuration + Og startup speedup
Message-ID: <afd80b540603271106k635ebab6r44dd6e5b3b29378e@mail.gmail.com>

Bryan, can you add these to the stable repo?

thanks,
-g.


Mon Mar 27 21:59:51 EEST 2006  George Moschovitis <gm at navel.gr>
  * Startup optimization, don't search in ObjectSpace for managed classes.

Mon Mar 27 21:43:44 EEST 2006  George Moschovitis <gm at navel.gr>
  * Further improved configuration system, even added some ruby magic,
check out the rdocs and the updated test case.



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Mon Mar 27 14:11:17 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 22:11:17 +0300
Subject: [Nitro] properties
Message-ID: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>

Bryan,

I saw you removed the og_properties patch. Perhaps we could include
this but please do NOT name the variable og_properties. Properties are
a general feature, NOT Og specific. Perhaps __properties would do? (Of
course we should keep the properties alias).

Perhaps there is a better alternative to avoid the GTK2 nameclash
though. Any ideas?

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From transfire at gmail.com  Mon Mar 27 14:22:31 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 14:22:31 -0500
Subject: [Nitro] properties
In-Reply-To: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
Message-ID: <4b6f054f0603271122u71886b05w6777d8be745f9869@mail.gmail.com>

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Bryan,
>
> I saw you removed the og_properties patch. Perhaps we could include
> this but please do NOT name the variable og_properties. Properties are
> a general feature, NOT Og specific. Perhaps __properties would do? (Of
> course we should keep the properties alias).
>
> Perhaps there is a better alternative to avoid the GTK2 nameclash
> though. Any ideas?

  propertys

;-) Just kidding, how about:

  instance_properties

similar to instance_variables.

T.



From george.moschovitis at gmail.com  Mon Mar 27 14:27:44 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Mon, 27 Mar 2006 22:27:44 +0300
Subject: [Nitro] properties
In-Reply-To: <4b6f054f0603271122u71886b05w6777d8be745f9869@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<4b6f054f0603271122u71886b05w6777d8be745f9869@mail.gmail.com>
Message-ID: <afd80b540603271127u5b0ed58ai89a283358ccfe33c@mail.gmail.com>

>   instance_properties

but they are class properties not instance properties.

-g.

--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From aglarond at gmail.com  Mon Mar 27 14:30:04 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Mon, 27 Mar 2006 21:30:04 +0200
Subject: [Nitro] properties
In-Reply-To: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
Message-ID: <55c107bf0603271130y5510bdedya3b7b492aa0d3cf5@mail.gmail.com>

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:

> Perhaps there is a better alternative to avoid the GTK2 nameclash
> though. Any ideas?

What about "facet"? (def 4. on dictionary.com:  One of numerous
aspects, as of a subject.)

- Dimitri



From transfire at gmail.com  Mon Mar 27 14:56:13 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 14:56:13 -0500
Subject: [Nitro] properties
In-Reply-To: <afd80b540603271127u5b0ed58ai89a283358ccfe33c@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<4b6f054f0603271122u71886b05w6777d8be745f9869@mail.gmail.com>
	<afd80b540603271127u5b0ed58ai89a283358ccfe33c@mail.gmail.com>
Message-ID: <4b6f054f0603271156q31ad1badt40797577ad02a9fa@mail.gmail.com>

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> >   instance_properties
>
> but they are class properties not instance properties.

Arn't they attributes? (ALbeit with annotations). To lookup attributes
one uses 'instance_variables'.

T.



From bryan.a.soto at gmail.com  Mon Mar 27 15:14:15 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 12:14:15 -0800
Subject: [Nitro] [PATCH] Configuration + Og startup speedup
In-Reply-To: <afd80b540603271106k635ebab6r44dd6e5b3b29378e@mail.gmail.com>
References: <afd80b540603271106k635ebab6r44dd6e5b3b29378e@mail.gmail.com>
Message-ID: <bd93db0d0603271214i229f66a8w11b1d00ca07162e1@mail.gmail.com>

Hey g,

You forgot the patch. :)

Thanks

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Bryan, can you add these to the stable repo?
>
> thanks,
> -g.
>
>
> Mon Mar 27 21:59:51 EEST 2006  George Moschovitis <gm at navel.gr>
>   * Startup optimization, don't search in ObjectSpace for managed classes.
>
> Mon Mar 27 21:43:44 EEST 2006  George Moschovitis <gm at navel.gr>
>   * Further improved configuration system, even added some ruby magic,
> check out the rdocs and the updated test case.
>
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Mon Mar 27 15:15:19 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 15:15:19 -0500
Subject: [Nitro] cattr
Message-ID: <4b6f054f0603271215u8b608c0l8f338f3aebeddea7@mail.gmail.com>

Currently glue/attribute defines the cattr_... mtheods to take default
values as a second parameter. Eg.

  cattr_accessor :foo, "default"

This mean only one attribute can be defined per call. Facets works
witha hash instead.

  cattr_accessor :foo => "default"

Which allows multiple attributes to be defined with one call.

I would like to change this to the Facets way in Nitro.

Other's take?

T.



From aidan at yoyo.org  Mon Mar 27 15:41:22 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Tue, 28 Mar 2006 07:41:22 +1100
Subject: [Nitro] cattr
In-Reply-To: <4b6f054f0603271215u8b608c0l8f338f3aebeddea7@mail.gmail.com>
References: <4b6f054f0603271215u8b608c0l8f338f3aebeddea7@mail.gmail.com>
Message-ID: <AA987F17-4027-4939-9454-76DFCCD516B9@yoyo.org>

What is glue/attribute used for?

Aidan

On 28/03/2006, at 7:15 AM, TRANS wrote:

> Currently glue/attribute defines the cattr_... mtheods to take default
> values as a second parameter. Eg.
>
>   cattr_accessor :foo, "default"
>
> This mean only one attribute can be defined per call. Facets works
> witha hash instead.
>
>   cattr_accessor :foo => "default"
>
> Which allows multiple attributes to be defined with one call.
>
> I would like to change this to the Facets way in Nitro.
>
> Other's take?
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From transfire at gmail.com  Mon Mar 27 15:54:21 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 15:54:21 -0500
Subject: [Nitro] cattr
In-Reply-To: <AA987F17-4027-4939-9454-76DFCCD516B9@yoyo.org>
References: <4b6f054f0603271215u8b608c0l8f338f3aebeddea7@mail.gmail.com>
	<AA987F17-4027-4939-9454-76DFCCD516B9@yoyo.org>
Message-ID: <4b6f054f0603271254r66375be0y35c6482650493b7e@mail.gmail.com>

On 3/27/06, Aidan Rogers <aidan at yoyo.org> wrote:
> What is glue/attribute used for?

It defines a class variable (@@foo) with accessor both at the class
level and the instance level. So:

  class X
    cattr_reader :foo
  end

does the same as:

  class X

    def self.foo
      @@foo
    end

    def foo
      @@foo
    end

  end

T.

P.S. Although one wonders if maybe it should do this instead:

  class X

    def self.foo
      @foo
    end

    def foo
      self.class.foo
    end

  end

which would make the attribute local to this particular class rather
then shared with all subclasses.



From aidan at yoyo.org  Mon Mar 27 15:54:32 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Tue, 28 Mar 2006 07:54:32 +1100
Subject: [Nitro] properties
In-Reply-To: <4b6f054f0603271156q31ad1badt40797577ad02a9fa@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<4b6f054f0603271122u71886b05w6777d8be745f9869@mail.gmail.com>
	<afd80b540603271127u5b0ed58ai89a283358ccfe33c@mail.gmail.com>
	<4b6f054f0603271156q31ad1badt40797577ad02a9fa@mail.gmail.com>
Message-ID: <A1BCBC2B-DF2E-4568-B431-042C9131401E@yoyo.org>


On 28/03/2006, at 6:56 AM, TRANS wrote:

> On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
>>>   instance_properties
>>
>> but they are class properties not instance properties.
>
> Arn't they attributes? (ALbeit with annotations). To lookup attributes
> one uses 'instance_variables'.

I had a few thoughts about this.

They are essentially instance variables, much like attr_accessor sets  
up in regular ruby, correct?

It would be very cool if we could just use attr_accessor/reader/ 
writer.  It could work much like property, but just take an extra  
class name on the end.  Something like this:

attr_accessor :first_name, :last_name, :middle_name, String

The downside of this is of course clashes if you _actually_ want to  
use the existing attr_accessor.

Personally, I think "property" is a very generic term - one we'll  
probably run foul of as Ruby's libraries mature and grow.

Right now, we can use prop_accessor/reader/writer and not use the  
keyword 'property' at all - that would solve the namespace clash for  
those who have it, no?

I realise this was a bit of a ramble - sorry - kind of a stream of  
consciousness thing :-)

Aidan


From transfire at gmail.com  Mon Mar 27 16:21:14 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 16:21:14 -0500
Subject: [Nitro] Two files changed w/o flexob
Message-ID: <4b6f054f0603271321r77bcc059ha0d2ce0b207332f0@mail.gmail.com>

Here are the two files that changed to get rid of Flexob. Both are
from glue. I ran the glue tests and they checked out.

T.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: template.rb
Type: application/x-extension-rb
Size: 5712 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060327/798bc93f/attachment.bin 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: configuration.rb
Type: application/x-extension-rb
Size: 3081 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060327/798bc93f/attachment-0001.bin 

From transfire at gmail.com  Mon Mar 27 16:36:34 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 16:36:34 -0500
Subject: [Nitro] Glue Undo
In-Reply-To: <afd80b540603271104x37fdcb40nad2505023db733e5@mail.gmail.com>
References: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
	<afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>
	<4b6f054f0603271054k427b9427v622c46509c6fb21f@mail.gmail.com>
	<afd80b540603271104x37fdcb40nad2505023db733e5@mail.gmail.com>
Message-ID: <4b6f054f0603271336s57a3b861g8ba2765e0945aa82@mail.gmail.com>

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > That would be great. I have an implementation of cuts in pure Ruby
> > btw. I just never realeased it b/c it not as ideal as a tru
> > implementation.
>
> yeah, i ve seen that in the tarball you have sent me some time ago.
>
> > Yes. OpenObject would do the trick if you prefer that to OpenStruct
> > --and in fact in my limited test OpenObject was faseter thean
> > OpenStruct (to my surprise).
>
> ok perhaps you can replace flexob with OpenObject.

Done. I sent the files to the mailing list. Would you prefer a patch. I did:

  darcs record
  darcs send -o patchcfile

Do I just email 'patchfile' now?

Thanks,
T.



From transfire at gmail.com  Mon Mar 27 16:39:17 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 16:39:17 -0500
Subject: [Nitro] REST
Message-ID: <4b6f054f0603271339p48da5fb0v50062577399a396f@mail.gmail.com>

interesting...

  http://jamis.jamisbuck.org/articles/2006/03/27/web-services-rails-style

T.



From bryan.a.soto at gmail.com  Mon Mar 27 16:45:55 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 13:45:55 -0800
Subject: [Nitro] Glue Undo
In-Reply-To: <4b6f054f0603271336s57a3b861g8ba2765e0945aa82@mail.gmail.com>
References: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
	<afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>
	<4b6f054f0603271054k427b9427v622c46509c6fb21f@mail.gmail.com>
	<afd80b540603271104x37fdcb40nad2505023db733e5@mail.gmail.com>
	<4b6f054f0603271336s57a3b861g8ba2765e0945aa82@mail.gmail.com>
Message-ID: <bd93db0d0603271345r28b4df4fkbe70ff39c6f081c8@mail.gmail.com>

On 3/27/06, TRANS <transfire at gmail.com> wrote:
> On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Done. I sent the files to the mailing list. Would you prefer a patch. I did:
>
>   darcs record
>   darcs send -o patchcfile
>
> Do I just email 'patchfile' now?
>

You might need to compress. The mailing list can be picky about sizes. :)

Other than that, just email to the list.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Mon Mar 27 16:54:16 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 16:54:16 -0500
Subject: [Nitro] Glue Undo
In-Reply-To: <bd93db0d0603271345r28b4df4fkbe70ff39c6f081c8@mail.gmail.com>
References: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
	<afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>
	<4b6f054f0603271054k427b9427v622c46509c6fb21f@mail.gmail.com>
	<afd80b540603271104x37fdcb40nad2505023db733e5@mail.gmail.com>
	<4b6f054f0603271336s57a3b861g8ba2765e0945aa82@mail.gmail.com>
	<bd93db0d0603271345r28b4df4fkbe70ff39c6f081c8@mail.gmail.com>
Message-ID: <4b6f054f0603271354o6531aad6qab0ca79b7ceac7cf@mail.gmail.com>

On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> You might need to compress. The mailing list can be picky about sizes. :)
>
> Other than that, just email to the list.

Thanks.

Well here's my first attempt at a darcs patch :-)

T.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patchfile.tar.bz2
Type: application/x-bzip2
Size: 10451 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060327/8fffa3d8/attachment.bz2 

From bryan.a.soto at gmail.com  Mon Mar 27 17:13:35 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 14:13:35 -0800
Subject: [Nitro] Glue Undo
In-Reply-To: <4b6f054f0603271354o6531aad6qab0ca79b7ceac7cf@mail.gmail.com>
References: <4b6f054f0603250513h26c973dr4c47a8a20960ec4e@mail.gmail.com>
	<afd80b540603270854w388d8c10wf38e77ed01ad6c96@mail.gmail.com>
	<4b6f054f0603271054k427b9427v622c46509c6fb21f@mail.gmail.com>
	<afd80b540603271104x37fdcb40nad2505023db733e5@mail.gmail.com>
	<4b6f054f0603271336s57a3b861g8ba2765e0945aa82@mail.gmail.com>
	<bd93db0d0603271345r28b4df4fkbe70ff39c6f081c8@mail.gmail.com>
	<4b6f054f0603271354o6531aad6qab0ca79b7ceac7cf@mail.gmail.com>
Message-ID: <bd93db0d0603271413l383ef575ldb2447ee961e1c0a@mail.gmail.com>

On 3/27/06, TRANS <transfire at gmail.com> wrote:
> On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> > You might need to compress. The mailing list can be picky about sizes. :)
> >
> > Other than that, just email to the list.
>
> Thanks.
>
> Well here's my first attempt at a darcs patch :-)

I'll check it out if no one beats me too it. Just got to finish up
with Dylan's Kirby patch. Oh, we usually send as a new thread with a
PATCH prefix in the subject. It helps ensure we don't miss patches.
Not that anyone ever does that... ;)

Thanks for the patch. :)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 27 17:16:54 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 14:16:54 -0800
Subject: [Nitro] Two files changed w/o flexob
In-Reply-To: <4b6f054f0603271321r77bcc059ha0d2ce0b207332f0@mail.gmail.com>
References: <4b6f054f0603271321r77bcc059ha0d2ce0b207332f0@mail.gmail.com>
Message-ID: <bd93db0d0603271416v10713a77k710d74b6d3f20aa9@mail.gmail.com>

Hi, I just want to confirm even though I'm fairly sure, this should be
ignored in favor of your patch, right?

On 3/27/06, TRANS <transfire at gmail.com> wrote:
> Here are the two files that changed to get rid of Flexob. Both are
> from glue. I ran the glue tests and they checked out.
>
> T.
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 27 17:29:56 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 14:29:56 -0800
Subject: [Nitro] REST
In-Reply-To: <4b6f054f0603271339p48da5fb0v50062577399a396f@mail.gmail.com>
References: <4b6f054f0603271339p48da5fb0v50062577399a396f@mail.gmail.com>
Message-ID: <bd93db0d0603271429q684fc2f6i8ad8426ec9372144@mail.gmail.com>

On 3/27/06, TRANS <transfire at gmail.com> wrote:
> interesting...
>
>   http://jamis.jamisbuck.org/articles/2006/03/27/web-services-rails-style
>

Yeah, I've been pondering that one myself. Though I was thinking that
you'd want to have different views and compiler pipelines (I've
actually just combined them both in my head and started calling them
transformations). Then, it'd just be delegated, i.e. if the client
requests xml, the xml transformation would be called after the
controller sets things up based off of the ordering the accept
headers.

I haven't really had much time to play with this idea, though. It'd
need zimba's per-controller dynamic pipeline in place, which is on my
list to integrate into stable. I guess my idea though, is that the
compiler pipeline should really belong to the view rather than the
controller.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 27 17:37:17 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 14:37:17 -0800
Subject: [Nitro] Glue move
In-Reply-To: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
References: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
Message-ID: <bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>

On 3/26/06, TRANS <transfire at gmail.com> wrote:
> I did a little more work on working Glue out. But as a proceeded I
> realized I probably would not be able to just submit a patch for this
> given that I pulled down George's repo then applied all the patches
> from the other development repo; plus the fact the it is such a major
> change.
>
> It seems like it will difficult to incorporate such a major change
> into two seperate repos. How should we proceed on this?
>
> Outline of what I'm doing:
>
>   move glue/lib/glue to nitro/lib/nitro/glue
>   move nitro/lib/glue/* to nitro/lib/nitro/glue
>   move glue/lib/html to nitro/lib/html
>   merge glue/lib/glue.rb into nitro/lib/nitro.rb (with some imporvements)
>   update all realted requires
>

Well, from what George has said previously, there is only one repo. He
was going to change the dns entry for repo.nitrohq.com to point to
devlab.

As for the file moves, as far as I know, darcs mv works fine in
patches. At least I've used it once in a patch with no problems. I
should probably double check that it retains the history though.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 27 17:55:08 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 14:55:08 -0800
Subject: [Nitro] AJAX: of interest
Message-ID: <bd93db0d0603271455m261a6b81hc88147cf4868d090@mail.gmail.com>

Since we're posting of interest links...

http://www.ajaxwrite.com/

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Mon Mar 27 18:46:20 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 18:46:20 -0500
Subject: [Nitro] Two files changed w/o flexob
In-Reply-To: <bd93db0d0603271416v10713a77k710d74b6d3f20aa9@mail.gmail.com>
References: <4b6f054f0603271321r77bcc059ha0d2ce0b207332f0@mail.gmail.com>
	<bd93db0d0603271416v10713a77k710d74b6d3f20aa9@mail.gmail.com>
Message-ID: <4b6f054f0603271546q7f57d064gd10a5bdd592498fa@mail.gmail.com>

On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Hi, I just want to confirm even though I'm fairly sure, this should be
> ignored in favor of your patch, right?

Right.



From transfire at gmail.com  Mon Mar 27 18:50:51 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 18:50:51 -0500
Subject: [Nitro] Glue move
In-Reply-To: <bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>
References: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
	<bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>
Message-ID: <4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>

On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:

> Well, from what George has said previously, there is only one repo. He
> was going to change the dns entry for repo.nitrohq.com to point to
> devlab.

Oh. Is that change going to happen then then?

> As for the file moves, as far as I know, darcs mv works fine in
> patches. At least I've used it once in a patch with no problems. I
> should probably double check that it retains the history though.

Well, if darcs can handle it then I can do it and submit a patch if you want.

T.



From transfire at gmail.com  Mon Mar 27 18:52:01 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 18:52:01 -0500
Subject: [Nitro] AJAX: of interest
In-Reply-To: <bd93db0d0603271455m261a6b81hc88147cf4868d090@mail.gmail.com>
References: <bd93db0d0603271455m261a6b81hc88147cf4868d090@mail.gmail.com>
Message-ID: <4b6f054f0603271552n5702154bpf64ba36d475428b1@mail.gmail.com>

While iteresting, I think this kind of thing starts to run into the
problem of running inside a browser.

T.



From bryan.a.soto at gmail.com  Mon Mar 27 19:25:15 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 16:25:15 -0800
Subject: [Nitro] Server.map and paths?
In-Reply-To: <20060327182555.GA11490@mlabdial.hit.bme.hu>
References: <20060327182555.GA11490@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>

On 3/27/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Hi!
> In auth/controller.rb there is:
>
> class AuthController < Nitro::Controller
>   e= File.join File.dirname(__FILE__), 'template'
>   puts "Auth template is: #{e}"

I think what you want to do is

mount_at(e) # mount_at seems assume e is a relative path from public so
                    # e should be '../auth/template'. I'm not sure if
that's a bug or correct functionality.

instead of
# >   Template.root = e

>  # ...
> end
>
> It prints the correct answer, and should set Template.root
>

George made a change to that before 0.29 was released. Hopefully he'll
be able to confirm my suggestion or steer you right. :)

Hope that helps.

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Mon Mar 27 19:33:40 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Mon, 27 Mar 2006 16:33:40 -0800
Subject: [Nitro] Glue move
In-Reply-To: <4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>
References: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
	<bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>
	<4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>
Message-ID: <bd93db0d0603271633i4c58c719s23e31f65b86f6f18@mail.gmail.com>

On 3/27/06, TRANS <transfire at gmail.com> wrote:
> On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
>
> > Well, from what George has said previously, there is only one repo. He
> > was going to change the dns entry for repo.nitrohq.com to point to
> > devlab.
>
> Oh. Is that change going to happen then then?
>

I'll leave that for George o answer. :)

> > As for the file moves, as far as I know, darcs mv works fine in
> > patches. At least I've used it once in a patch with no problems. I
> > should probably double check that it retains the history though.
>
> Well, if darcs can handle it then I can do it and submit a patch if you want.

Hmm... Were we still going to add facets to the repo? If so, why don't
we do that first. Especially if doing the darcs mv retains history. I
still need to double check that.

Either way, we should probably figure out where everything will be
moving or if it's even going to be kept. I'll put your sketch up on
the devlab wiki and open a ticket. We can hash it out from there.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Mon Mar 27 19:49:59 2006
From: transfire at gmail.com (TRANS)
Date: Mon, 27 Mar 2006 19:49:59 -0500
Subject: [Nitro] Glue move
In-Reply-To: <bd93db0d0603271633i4c58c719s23e31f65b86f6f18@mail.gmail.com>
References: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
	<bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>
	<4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>
	<bd93db0d0603271633i4c58c719s23e31f65b86f6f18@mail.gmail.com>
Message-ID: <4b6f054f0603271649jef98c83le0288970b55783e5@mail.gmail.com>

On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/27/06, TRANS <transfire at gmail.com> wrote:
> > On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> >
> > > Well, from what George has said previously, there is only one repo. He
> > > was going to change the dns entry for repo.nitrohq.com to point to
> > > devlab.
> >
> > Oh. Is that change going to happen then then?
> >
>
> I'll leave that for George o answer. :)
>
> > > As for the file moves, as far as I know, darcs mv works fine in
> > > patches. At least I've used it once in a patch with no problems. I
> > > should probably double check that it retains the history though.
> >
> > Well, if darcs can handle it then I can do it and submit a patch if you want.
>
> Hmm... Were we still going to add facets to the repo? If so, why don't
> we do that first.

I don't think so actually. I want a repo but I think you are right
that nitro's should be spilt up, do lets put facets in a separte repo.
There is also another reason for this. Facets has some special
characteristics. For instance there are a couple of rake tasks that
must be run before a release to generate redirection require files.
Also it can not really be tested in the normal way b/c there are a few
facets can interfere with others. That is why I use reap test, b/c
reap test every test program in a fully separate process.

Anyway, if you see no difficulties in managing these differences then
I'll send you my latest version of Facets and we can proceed in that
regard.

> Especially if doing the darcs mv retains history. I
> still need to double check that.

Check that and let me know. If it's okay I can do it in pretty short order.

> Either way, we should probably figure out where everything will be
> moving or if it's even going to be kept. I'll put your sketch up on
> the devlab wiki and open a ticket. We can hash it out from there.

I looked through it and I think the best thing to do is just get it
over there first --with everytihg working. Then we can start widdling
away at it. If we try to do it the other way around it could be a
while before we ever get past the Glue dependency and I would like to
at least see that part done since we've been talking about it for so
long and it is the major change involved.

T.



From james_b at neurogami.com  Mon Mar 27 22:48:25 2006
From: james_b at neurogami.com (James Britt)
Date: Mon, 27 Mar 2006 20:48:25 -0700
Subject: [Nitro] REST
In-Reply-To: <4b6f054f0603271339p48da5fb0v50062577399a396f@mail.gmail.com>
References: <4b6f054f0603271339p48da5fb0v50062577399a396f@mail.gmail.com>
Message-ID: <4428B209.6040601@neurogami.com>

TRANS wrote:
> interesting...
> 
>   http://jamis.jamisbuck.org/articles/2006/03/27/web-services-rails-style
> 

This may prove enlightening , regarding REST and Web services:

http://lists.xml.org/archives/xml-dev/200503/msg00468.html


I believe the REST ideas in Rails had their genesis via this discussion 
list:


http://microformats.org/discuss/mail/microformats-rest/2005-November/000014.html

Some interesting points are made, independent of any specific 
implementation.


-- 
James Britt

"The greatest obstacle to discovery is not ignorance, but the illusion 
of knowledge."
  - D. Boorstin


From transfire at gmail.com  Tue Mar 28 02:25:15 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 28 Mar 2006 02:25:15 -0500
Subject: [Nitro] More Great Reap News
Message-ID: <4b6f054f0603272325p96968cdt222b3c6f5f942bac@mail.gmail.com>

I just added another feature to Reap. I had almost forgotten all about
this idea from many months ago when I first implemented test
extraction....

Now Reap includes a bin file called 'rubytest'. It will automatically
run the embedded test in a source file, so there's no longer any need
to manually remark out the '=begin test' and '=end' to do a local
development test. Eg.

  # foo.rb

  def x ; 1 ; end

  =begin test

    class TCFoo < Test::Unit::TestCase

       def test_x
         assert_equal( 1, x )
       end

    end

  =end

then on the command line:

  % rubytest foo.rb
  Loaded suite /usr/bin/rubytest
  Started
  .
  Finished in 0.010354 seconds.

  1 tests, 1 assertions, 0 failures, 0 errors


T.



From bryan.a.soto at gmail.com  Tue Mar 28 03:04:53 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 00:04:53 -0800
Subject: [Nitro] PATCH: bs-kirby-fixes
Message-ID: <bd93db0d0603280004w50a2e5ffuaa21ecd5c54bc411@mail.gmail.com>

Hey list,

I had to rework Dylan's Kirby fixes patch. It didn't apply cleanly
since things had changed in the devlab repo since his submission.
Obviously not his fault. :)

It's almost all his code so he get's the credit. I just changed some
of his camelCase variable names. :)

Kirby now goes from not running to:

43 tests, 130 assertions, 4 failures, 7 errors

Could someone give this a lookover? It's late for me.

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bs-kirby-fixes.zip
Type: application/zip
Size: 18462 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060328/a7c3f8d0/attachment.zip 

From transfire at gmail.com  Tue Mar 28 09:01:31 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 28 Mar 2006 09:01:31 -0500
Subject: [Nitro] REST
In-Reply-To: <4428B209.6040601@neurogami.com>
References: <4b6f054f0603271339p48da5fb0v50062577399a396f@mail.gmail.com>
	<4428B209.6040601@neurogami.com>
Message-ID: <4b6f054f0603280601g350d5aa4o4322fc2729bfe4b7@mail.gmail.com>

On 3/27/06, James Britt <james_b at neurogami.com> wrote:
> This may prove enlightening , regarding REST and Web services:
>
> http://lists.xml.org/archives/xml-dev/200503/msg00468.html
>
>
> I believe the REST ideas in Rails had their genesis via this discussion
> list:
>
>
> http://microformats.org/discuss/mail/microformats-rest/2005-November/000014.html
>
> Some interesting points are made, independent of any specific
> implementation.

Very interesting!

T.



From vseguip at gmail.com  Tue Mar 28 09:54:21 2006
From: vseguip at gmail.com (vseguip at gmail.com)
Date: Tue, 28 Mar 2006 16:54:21 +0200
Subject: [Nitro] properties
In-Reply-To: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
Message-ID: <dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>

 I was the author of the original patch. I have been out of office for
some days and I have some how missed (I admit I havent read through
all the ML) why the patch has been removed (I remember it had been
commited), but I would like to work out the reasons for it not being
applied and re submit a new patch if necessary as this is quite
important for me.

Thanks,
  V. Segu?

P.S: How about is "class_properties" or as suggested "facet_properties"?



From bryan.a.soto at gmail.com  Tue Mar 28 15:35:29 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 12:35:29 -0800
Subject: [Nitro] properties
In-Reply-To: <dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
Message-ID: <bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>

On 3/28/06, vseguip at gmail.com <vseguip at gmail.com> wrote:
>  I was the author of the original patch. I have been out of office for
> some days and I have some how missed (I admit I havent read through
> all the ML) why the patch has been removed (I remember it had been
> commited), but I would like to work out the reasons for it not being
> applied and re submit a new patch if necessary as this is quite
> important for me.
>

Hi,

I (un)pulled the patch for a couple of reasons. Partly because it was
causing some errors with some missed renames of properties to
og_properties, but mostly because I was sure George would object to
the name. :)

By the way, I just looked and saw that property is just an alias for
prop_accessor. Does GTK2 use that method name as well? If not, perhaps
that's a temporary solution for you until we can get a good name or
alternate implementation nailed down?

Bryan

> Thanks,
>   V. Segu?
>
> P.S: How about is "class_properties" or as suggested "facet_properties"?
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Tue Mar 28 15:36:31 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 12:36:31 -0800
Subject: [Nitro] properties
In-Reply-To: <bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
Message-ID: <bd93db0d0603281236g5afd75e8udf57439a9039ddd9@mail.gmail.com>

On 3/28/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> On 3/28/06, vseguip at gmail.com <vseguip at gmail.com> wrote:
> >  I was the author of the original patch. I have been out of office for
> > some days and I have some how missed (I admit I havent read through
> > all the ML) why the patch has been removed (I remember it had been
> > commited), but I would like to work out the reasons for it not being
> > applied and re submit a new patch if necessary as this is quite
> > important for me.
> >
>
> Hi,
>
> I (un)pulled the patch for a couple of reasons. Partly because it was
> causing some errors with some missed renames of properties to
> og_properties, but mostly because I was sure George would object to
> the name. :)
>
> By the way, I just looked and saw that property is just an alias for
> prop_accessor. Does GTK2 use that method name as well? If not, perhaps
> that's a temporary solution for you until we can get a good name or
> alternate implementation nailed down?
>

Sorry, I got it stuck in my head that your problem was something
different. My mistake.

> > P.S: How about is "class_properties" or as suggested "facet_properties"?
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From news at stephan.walter.name  Tue Mar 28 15:57:16 2006
From: news at stephan.walter.name (Stephan Walter)
Date: Tue, 28 Mar 2006 22:57:16 +0200
Subject: [Nitro] og relations strangeness
Message-ID: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>

Hi,

consider these two Og classes:

class Widget
        property :name, String
        belongs_to :owner, User
end

class User
	property :name
	has_many :widgets, Widget
end


Og will create an owner_oid row for the widgets table. IMHO this should be
user_oid instead. Currently Og or Nitro (I'm not sure...) is looking for a
user_oid field in the widgets table when looking up the widgets for a user
(in the admin interface for example). Which obviously won't work and give
you an Og::StoreException.

Or maybe what I want to do is not supported? It would be nice to be able
to describe an object's roles in relationships differently. E.g. a
user can be a sender, an editor, a receiver, a maintainer, ...etc.

I know that I can specify :field explicitly, but the default way should
work out-of-the-box.

-Stephan




From kgergely at mcl.hu  Tue Mar 28 16:57:43 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Tue, 28 Mar 2006 23:57:43 +0200
Subject: [Nitro] table helper example?
Message-ID: <20060328215743.GA27000@mlabdial.hit.bme.hu>

Hi!

I've found a nice tip a bout table helper's oxyliquit, but a bit 
stucked.

It mentions about :values in :order hash of the table method, and I can 
also have a nice asc_pic and desc_pic, but how can my controller use 
that info? So when I click on the images, something should happen in the 
controller... (sg. has to be changed in the controller)

thx
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060328/ecb6c24d/attachment.bin 

From kgergely at mcl.hu  Tue Mar 28 17:05:22 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Wed, 29 Mar 2006 00:05:22 +0200
Subject: [Nitro] automatic validation
Message-ID: <20060328220522.GB27000@mlabdial.hit.bme.hu>

Hi!

Come across in nitrohq.com:
  class User
    prop_accessor :name, :password, String
    validate_confirmation :password
    validate_length :name, :range => 2..12
  end

  u = User.new(...)
  unless u.valid?
    p u.errors.on(:name)
    p u.errors[:password]
  end

What does this code fragment do?

thx
Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060328/c54fbf37/attachment.bin 

From kgergely at mcl.hu  Tue Mar 28 17:09:59 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Wed, 29 Mar 2006 00:09:59 +0200
Subject: [Nitro] Server.map and paths?
In-Reply-To: <bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>
References: <20060327182555.GA11490@mlabdial.hit.bme.hu>
	<bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>
Message-ID: <20060328220959.GC27000@mlabdial.hit.bme.hu>

Bryan Soto azt ?rta, hogy: 
> On 3/27/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> > Hi!
> > In auth/controller.rb there is:
> >
> > class AuthController < Nitro::Controller
> >   e= File.join File.dirname(__FILE__), 'template'
> >   puts "Auth template is: #{e}"
> I think what you want to do is
> 
> mount_at(e) # mount_at seems assume e is a relative path from public so
>                     # e should be '../auth/template'. I'm not sure if
> that's a bug or correct functionality.

Figured out, that manipulating @template_root also can help...

Can I somehow get this automatic?

I plan to have many "components", which are sitting in  separate 
directories, and don't like the idea to put 5-6 garbage lines in each 
source file.
However, when using __FILE__ module inclusion won't do, what I'd like to 
achieve (putting this garbage in a module, and including it...)

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060328/d1f13faa/attachment.bin 

From bryan.a.soto at gmail.com  Tue Mar 28 17:19:41 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 14:19:41 -0800
Subject: [Nitro] kirby patches and a test fix
In-Reply-To: <441A1162.1070707@digitalvalence.com>
References: <44132BC3.9080905@digitalvalence.com>
	<bd93db0d0603131407j20fdd1dbjdf05b412201fa0e1@mail.gmail.com>
	<44161AFB.8030801@digitalvalence.com>
	<bd93db0d0603160025t2d26a88bi82ff24de022c69bd@mail.gmail.com>
	<441A1162.1070707@digitalvalence.com>
Message-ID: <bd93db0d0603281419p1c0dc7c2pebffe41d1cba4eab@mail.gmail.com>

Hi Dylan,

Sorry for the long delay. I just wanted to let you know your patch has
been applied. Thanks for submitting. :)

On 3/16/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
> Let me know if there are any bugs you don't want/have time to just patch
> up; as for the style tip, I could swear that I read that method names
> were underscored and variable names were camel case; but I see now that
> the latter is totally incorrect.  I must have been in a code induced
> haze when I hallucinated that.  I read so much Java code during my day
> job that I didn't notice the disparity.  Thanks for bringing it to my
> attention.
>
> I'll have to earmark some time to reread the unofficial style guidelines
> at caliban.
>
> Bryan Soto wrote:
> > On 3/13/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
> >
> >> You're quite welcome; I only wish I had more time to contribute to
> >> things like this.  The story of us all ;)
> >>
> >> Dylan
> >>
> >>
> >
> > I certainly hope you find some more time. :)
> >
> > Before patch:
> > 43 tests,  78 assertions, 6 failures, 14 errors
> >
> > After patch:
> > 43 tests, 126 assertions, 5 failures, 6 errors
> >
> > Some conflicts with the recently applied og_properties fix in the GTK2 patch.
> >
> > I like your idea of storing ogtype in the SchemaInheritanceBase
> > module. Perhaps that should be applied to the other stores. Makes
> > things simpler if it's just another Og managed property.
> >
> > One style tip, though you seem to know it. Ruby programs typically
> > don't use camel case for variables, i.e. resultSet would be result_set
> > and testCondition would be test_condition. Not a big deal at all, of
> > course. Just thought I'd mention it not knowing your familiarity with
> > Ruby. :)
> >
> > I'll review this a bit more as it's a fairly big change, but I'm
> > liking what I see so far. :)
> >
> > Bryan
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
> >
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Tue Mar 28 18:53:19 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 28 Mar 2006 18:53:19 -0500
Subject: [Nitro] Glue Patch Away!
Message-ID: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>

Here's the patch for removing the glue dependency.

There still remains a few details to attend. I merged lib/glue.rb into
lib/nitro.rb. That seemed like the right thing to do (although maybe
there's a reason to keep some separation?) In a couple of places in
the system there's

   require 'glue'

I just remarked these out. They may need to be replaced with

  require 'nitro'

It will take a little bit of additional work yet to get all the kinks
out, but I'm having issues running the tests right now. So hopefully
someone with more expeirence in the patch "end game" can work these
thru without much trouble.

One error I was getting that worries me is with nitro/util/paramix.rb
--a stack level too deep error. Of course ultimately this lib should
be replaced with facet/paramix.rb, but George mentioned some problem
in that regard --I'm not sure what it is. Anyway, at this point it may
just be a problem with running the tests on my end and not a real
error at all.

One last simple thing. I didn't remove the glue package dependency in
the ProjectInfo file. That will need to be done, but that's an easy
one.

Oh, and this patch includes the removal of Flexob too.

T.

P.S. I recorded four times in the proccess, if that matters.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patchfile.tar.bz2
Type: application/x-bzip2
Size: 27919 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060328/d8b9e004/attachment.bz2 

From dylanb at digitalvalence.com  Tue Mar 28 19:19:33 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Tue, 28 Mar 2006 18:19:33 -0600
Subject: [Nitro] kirby patches and a test fix
In-Reply-To: <bd93db0d0603281419p1c0dc7c2pebffe41d1cba4eab@mail.gmail.com>
References: <44132BC3.9080905@digitalvalence.com>	<bd93db0d0603131407j20fdd1dbjdf05b412201fa0e1@mail.gmail.com>	<44161AFB.8030801@digitalvalence.com>	<bd93db0d0603160025t2d26a88bi82ff24de022c69bd@mail.gmail.com>	<441A1162.1070707@digitalvalence.com>
	<bd93db0d0603281419p1c0dc7c2pebffe41d1cba4eab@mail.gmail.com>
Message-ID: <4429D295.5030203@digitalvalence.com>

Not a problem, Bryan.  We are all busy folk, it seems.

Bryan Soto wrote:
> Hi Dylan,
>
> Sorry for the long delay. I just wanted to let you know your patch has
> been applied. Thanks for submitting. :)
>
> On 3/16/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
>   
>> Let me know if there are any bugs you don't want/have time to just patch
>> up; as for the style tip, I could swear that I read that method names
>> were underscored and variable names were camel case; but I see now that
>> the latter is totally incorrect.  I must have been in a code induced
>> haze when I hallucinated that.  I read so much Java code during my day
>> job that I didn't notice the disparity.  Thanks for bringing it to my
>> attention.
>>
>> I'll have to earmark some time to reread the unofficial style guidelines
>> at caliban.
>>
>> Bryan Soto wrote:
>>     
>>> On 3/13/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
>>>
>>>       
>>>> You're quite welcome; I only wish I had more time to contribute to
>>>> things like this.  The story of us all ;)
>>>>
>>>> Dylan
>>>>
>>>>
>>>>         
>>> I certainly hope you find some more time. :)
>>>
>>> Before patch:
>>> 43 tests,  78 assertions, 6 failures, 14 errors
>>>
>>> After patch:
>>> 43 tests, 126 assertions, 5 failures, 6 errors
>>>
>>> Some conflicts with the recently applied og_properties fix in the GTK2 patch.
>>>
>>> I like your idea of storing ogtype in the SchemaInheritanceBase
>>> module. Perhaps that should be applied to the other stores. Makes
>>> things simpler if it's just another Og managed property.
>>>
>>> One style tip, though you seem to know it. Ruby programs typically
>>> don't use camel case for variables, i.e. resultSet would be result_set
>>> and testCondition would be test_condition. Not a big deal at all, of
>>> course. Just thought I'd mention it not knowing your familiarity with
>>> Ruby. :)
>>>
>>> I'll review this a bit more as it's a fairly big change, but I'm
>>> liking what I see so far. :)
>>>
>>> Bryan
>>>
>>> _______________________________________________
>>> Nitro-general mailing list
>>> Nitro-general at rubyforge.org
>>> http://rubyforge.org/mailman/listinfo/nitro-general
>>>
>>>
>>>       
>> _______________________________________________
>> Nitro-general mailing list
>> Nitro-general at rubyforge.org
>> http://rubyforge.org/mailman/listinfo/nitro-general
>>
>>     
>
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>   



From aidan at yoyo.org  Tue Mar 28 19:20:50 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Wed, 29 Mar 2006 11:20:50 +1100
Subject: [Nitro] og relations strangeness
In-Reply-To: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
References: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
Message-ID: <AFB2F8D2-D690-40B8-9E54-6F8A16C8FEF8@yoyo.org>

Stephan,

You can specify the relationship to have a different foreign name  
like so:

has_many :widgets, Widget, :foreign_name => :owner

See the Og tutorial on http://oxyliquit.de/tutorial/1 for more info.

Aidan

On 29/03/2006, at 7:57 AM, Stephan Walter wrote:

> Hi,
>
> consider these two Og classes:
>
> class Widget
>         property :name, String
>         belongs_to :owner, User
> end
>
> class User
> 	property :name
> 	has_many :widgets, Widget
> end
>
>
> Og will create an owner_oid row for the widgets table. IMHO this  
> should be
> user_oid instead. Currently Og or Nitro (I'm not sure...) is  
> looking for a
> user_oid field in the widgets table when looking up the widgets for  
> a user
> (in the admin interface for example). Which obviously won't work  
> and give
> you an Og::StoreException.
>
> Or maybe what I want to do is not supported? It would be nice to be  
> able
> to describe an object's roles in relationships differently. E.g. a
> user can be a sender, an editor, a receiver, a maintainer, ...etc.
>
> I know that I can specify :field explicitly, but the default way  
> should
> work out-of-the-box.
>
> -Stephan
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From james_b at neurogami.com  Tue Mar 28 19:35:05 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 28 Mar 2006 17:35:05 -0700
Subject: [Nitro] Server.map and paths?
In-Reply-To: <20060328220959.GC27000@mlabdial.hit.bme.hu>
References: <20060327182555.GA11490@mlabdial.hit.bme.hu>	<bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>
	<20060328220959.GC27000@mlabdial.hit.bme.hu>
Message-ID: <4429D639.4020504@neurogami.com>

Gergely Kontra wrote:
> Bryan Soto azt ?rta, hogy: 
> 
>>On 3/27/06, Gergely Kontra <kgergely at mcl.hu> wrote:
>>
>>>Hi!
>>>In auth/controller.rb there is:
>>>
>>>class AuthController < Nitro::Controller
>>>  e= File.join File.dirname(__FILE__), 'template'
>>>  puts "Auth template is: #{e}"
>>
>>I think what you want to do is
>>
>>mount_at(e) # mount_at seems assume e is a relative path from public so
>>                    # e should be '../auth/template'. I'm not sure if
>>that's a bug or correct functionality.
> 
> 
> Figured out, that manipulating @template_root also can help...

Can you explain  bit more what your did?


Thanks,

-- 
James Britt

"In Ruby, no one cares who your parents were, all they care
  about is if you know  what you are talking about."
   - Logan Capaldo


From transfire at gmail.com  Tue Mar 28 20:18:43 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 28 Mar 2006 20:18:43 -0500
Subject: [Nitro] og relations strangeness
In-Reply-To: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
References: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
Message-ID: <4b6f054f0603281718o16a5ac70vf8ce02645835b729@mail.gmail.com>

On 3/28/06, Stephan Walter <news at stephan.walter.name> wrote:
> Hi,
>
> consider these two Og classes:
>
> class Widget
>         property :name, String
>         belongs_to :owner, User
> end
>
> class User
>         property :name
>         has_many :widgets, Widget
> end
>
>
> Og will create an owner_oid row for the widgets table. IMHO this should be
> user_oid instead. Currently Og or Nitro (I'm not sure...) is looking for a
> user_oid field in the widgets table when looking up the widgets for a user
> (in the admin interface for example). Which obviously won't work and give
> you an Og::StoreException.

Hmm... well that must be a bug. Which is it "supposed" to be?

T.



From transfire at gmail.com  Tue Mar 28 22:03:02 2006
From: transfire at gmail.com (TRANS)
Date: Tue, 28 Mar 2006 22:03:02 -0500
Subject: [Nitro] how classes are enchanted
Message-ID: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>

Was talking to manveru more about how classes get enchanted. If you'll
recall he had suggested this clever device:

  def require_og_models(*models)
    snapshot_pre = []
    snapshot_post = []
    ObjectSpace.each_object{|o| snapshot_pre << o}
    models.each{|m| require(m)}
    ObjectSpace.each_object{|o| snapshot_post << o}
    enchant(snapshot_post-snapshot_pre)
  end

  require_og_models "model.rb"

This is interesting, but traversing the ObjectSpace like this is too
slow. I thought of another way something like this (untested):

  def require_og_models(*models)
    def ::Object.inherited( klass )
      klass.class_eval { include Og::EntityMixin }
    end
    models.each{|m| require(m)}
    class << ::Object
      remove_method :inherited
    end
  end
  require_og_models "model.rb"

This manages to pull off the same cool effect but in an efficient way.
The only downside to this whole idea though is that it forces us to
keep our models in special separate files. Is that too stiff a
requirement? Or maybe there is way around that?

So that's something to think about.

But that aside, at the very least I would like to rename
Og::EntityMixin to Og::Enchanted. That way one can specify

  class Foo
    is Og::Enchanted

to manually enchanted a class. (One can still of course use 'class Foo
< Og::Entity' if one prefers).

T.


P.S. There's also the longer range idea of dividing enchantment into
two parts. The first would be give notation relation capabilities to
all objects:

  module Kernel
    include Enchantments

Only later would the actual persistance layer be injected into the
classes as perscibed by their usage of the enchantments.



From james_b at neurogami.com  Tue Mar 28 23:40:09 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 28 Mar 2006 21:40:09 -0700
Subject: [Nitro] how classes are enchanted
In-Reply-To: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
Message-ID: <442A0FA9.8030107@neurogami.com>

TRANS wrote:
...

> This manages to pull off the same cool effect but in an efficient way.
> The only downside to this whole idea though is that it forces us to
> keep our models in special separate files. Is that too stiff a
> requirement? Or maybe there is way around that?

One of the things that made Og to appealing to me is that I can just 
write Ruby code with without tracking possible persistence needs.  If 
and when I decide some object needs saving, I can just change a few 
lines and let the magic happen.

-- 
James Britt

"A principle or axiom is of no value without the rules for applying it."
   - Len Bullard


From transfire at gmail.com  Wed Mar 29 00:07:17 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 00:07:17 -0500
Subject: [Nitro] how classes are enchanted
In-Reply-To: <442A0FA9.8030107@neurogami.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
	<442A0FA9.8030107@neurogami.com>
Message-ID: <4b6f054f0603282107v6563549ap7263f3aa1c4f871b@mail.gmail.com>

On 3/28/06, James Britt <james_b at neurogami.com> wrote:
> > This manages to pull off the same cool effect but in an efficient way.
> > The only downside to this whole idea though is that it forces us to
> > keep our models in special separate files. Is that too stiff a
> > requirement? Or maybe there is way around that?
>
> One of the things that made Og to appealing to me is that I can just
> write Ruby code with without tracking possible persistence needs.  If
> and when I decide some object needs saving, I can just change a few
> lines and let the magic happen.

How's that? You still have to change all those attributes to properties.

T.



From transfire at gmail.com  Wed Mar 29 00:20:19 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 00:20:19 -0500
Subject: [Nitro] how classes are enchanted
In-Reply-To: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
Message-ID: <4b6f054f0603282120s7769030g98c3684144d21d6a@mail.gmail.com>

Some more consideration:

  def Og.enchant(*models)

    model_files, models = models.partition { |m| String === m }

    # load model files, enchanting classes as loaded

    def ::Object.inherited( klass )
      klass.class_eval { include Og::EntityMixin }
    end
    model_files.each{|m| require(m)}
    class << ::Object
      remove_method :inherited
    end

    # enchant model classes

    models.each { |mod|
      case mod
      when Class
        klass.class_eval { include Og::EntityMixin }
      when Module
         # ... enchant all classes in module ...
      else
        # error or singleon enchant?
      end
    }

  end

This gives a number of ways to enchant a class, besides the manual way.

  Og.enchant( FooClass )      # directly on an external class
  Og.enchant( FooModule )   # classes in FooModule
  Og.enchant( "models.rb" )  # classes in a separate file

T.



From kashia at vfemail.net  Wed Mar 29 00:24:43 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Wed, 29 Mar 2006 07:24:43 +0200
Subject: [Nitro] table helper example?
In-Reply-To: <20060328215743.GA27000@mlabdial.hit.bme.hu>
References: <20060328215743.GA27000@mlabdial.hit.bme.hu>
Message-ID: <op.s65tbhlhg3glhz@jo.local>

On Tue, 28 Mar 2006 23:57:43 +0200, Gergely Kontra <kgergely at mcl.hu> wrote:

> Hi!
>
> I've found a nice tip a bout table helper's oxyliquit, but a bit
> stucked.
>
> It mentions about :values in :order hash of the table method, and I can
> also have a nice asc_pic and desc_pic, but how can my controller use
> that info? So when I click on the images, something should happen in the
> controller... (sg. has to be changed in the controller)
>
> thx

got no time right now

order_by = request.params[TableHelper.order_by_key] || 'oid'
     order_direction = request.params[TableHelper.order_direction_key] || "ASC"
     order = "#{order_by} #{order_direction}"

     entries, @pager = paginate(ContainersView, :condition => "oid IN (SELECT * FROM tag_or_filetype('#{tags.join(', ')}'))", :order => order, :per_page => 10)
     @headers = ['ID', 'Movie', 'Dist', 'Tags']
     @data = entries.map {|c| [c.oid, c.title, c.distribution, c.tags] }
     @order_options = { :right => true,
             :values => ['oid', 'title'],
             :asc_pic => "/images/upc.png",
             :desc_pic => "/images/downc.png" }

messy, but I hopt it helps, digest slowly

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From james_b at neurogami.com  Wed Mar 29 01:19:10 2006
From: james_b at neurogami.com (James Britt)
Date: Tue, 28 Mar 2006 23:19:10 -0700
Subject: [Nitro] how classes are enchanted
In-Reply-To: <4b6f054f0603282107v6563549ap7263f3aa1c4f871b@mail.gmail.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>	<442A0FA9.8030107@neurogami.com>
	<4b6f054f0603282107v6563549ap7263f3aa1c4f871b@mail.gmail.com>
Message-ID: <442A26DE.4070002@neurogami.com>

TRANS wrote:
> On 3/28/06, James Britt <james_b at neurogami.com> wrote:
> 
>>>This manages to pull off the same cool effect but in an efficient way.
>>>The only downside to this whole idea though is that it forces us to
>>>keep our models in special separate files. Is that too stiff a
>>>requirement? Or maybe there is way around that?
>>
>>One of the things that made Og to appealing to me is that I can just
>>write Ruby code with without tracking possible persistence needs.  If
>>and when I decide some object needs saving, I can just change a few
>>lines and let the magic happen.
> 
> 
> How's that? You still have to change all those attributes to properties.

True.  But very simple, and doesn't break my flow.
-- 
James Britt

http://web2.0validator.com    - We're the Dot in Web 2.0
http://refreshingcities.org   - Design, technology, usability
http://yourelevatorpitch.com  - Finding Business Focus
http://www.jamesbritt.com     - Playing with Better Toys


From bryan.a.soto at gmail.com  Wed Mar 29 01:26:24 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 22:26:24 -0800
Subject: [Nitro] automatic validation
In-Reply-To: <20060328220522.GB27000@mlabdial.hit.bme.hu>
References: <20060328220522.GB27000@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603282226q29c8ce55pb54d8e0325a1ef9e@mail.gmail.com>

On 3/28/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Hi!
>
> Come across in nitrohq.com:
>   class User
>     prop_accessor :name, :password, String

>     validate_confirmation :password
       # compares password to password_confirmation. I'm not sure how
this one works.
       # It seems to assume there is a method for user named
password_confirmation.
       # You probably need an attr_accessor :password_confirmation in
the class that
       # you'd set appropriately.

>     validate_length :name, :range => 2..12
       # ensures that the string in name is between 2 and 12 characters long.
>   end
>
>   u = User.new(...)
>   unless u.valid? # this method is automatically called by u.save
>     p u.errors.on(:name)
       # displays any errors for the field name
>     p u.errors[:password]
       # displays any errors for the field password
>   end
>

Assuming you installed nitro via gems, try running gem_server at your
command prompt then go to http://localhost:8808/ in your web browser.

Scroll through the installed gems to glue, click on RDoc link. The
class you are interested in is Glue::Validation,
Glue::Validation::ClassMethods and Glue::Validation::Errors. It's not
obvious, but the method signatures are links that will pop up a new
window containing the source code for that method.

Hope that helps.

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From aglarond at gmail.com  Wed Mar 29 01:30:23 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Wed, 29 Mar 2006 08:30:23 +0200
Subject: [Nitro] og relations strangeness
In-Reply-To: <4b6f054f0603281718o16a5ac70vf8ce02645835b729@mail.gmail.com>
References: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
	<4b6f054f0603281718o16a5ac70vf8ce02645835b729@mail.gmail.com>
Message-ID: <55c107bf0603282230s6871d56fpca7ac63044722048@mail.gmail.com>

On 3/29/06, TRANS <transfire at gmail.com> wrote:
> Hmm... well that must be a bug. Which is it "supposed" to be?

Doesn't look like a bug to me.  Re: Aidan's response, User naturally
looks for a field called :user_oid, unless you override it with the
:foreign_name declaration.  To avoid this problem in the first place,
the initial class declaration for Widget could read

 class Widget
       property :name, String
       belongs_to :user, User
end

 or even

class Widget
       property :name, String
       belongs_to :user
end

Right?  Or, just do as Aidan suggested.

- Dimitri



From vseguip at gmail.com  Wed Mar 29 01:33:52 2006
From: vseguip at gmail.com (vseguip at gmail.com)
Date: Wed, 29 Mar 2006 08:33:52 +0200
Subject: [Nitro] properties
In-Reply-To: <bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
Message-ID: <dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>

>
> I (un)pulled the patch for a couple of reasons. Partly because it was
> causing some errors with some missed renames of properties to
> og_properties, but mostly because I was sure George would object to
> the name. :)
>

Hi,
  Do you remember where I missed renaming so I can add it? It would be
very helpful to have a test case to demo it (I'm pretty sure Og as
such passed all the test cases when I wrote the patch) and integrate
it with the current test suites.
 OTOH I hope we can agree on a name in this thread so I can put my ass
to work asap.

  Cheers,
  V.Segu?



From aglarond at gmail.com  Wed Mar 29 01:36:20 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Wed, 29 Mar 2006 08:36:20 +0200
Subject: [Nitro] how classes are enchanted
In-Reply-To: <4b6f054f0603282120s7769030g98c3684144d21d6a@mail.gmail.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
	<4b6f054f0603282120s7769030g98c3684144d21d6a@mail.gmail.com>
Message-ID: <55c107bf0603282236i1b36f775l81bc5a2e7afae42d@mail.gmail.com>

On 3/29/06, TRANS <transfire at gmail.com> wrote:

>   Og.enchant( FooClass )      # directly on an external class
>   Og.enchant( FooModule )   # classes in FooModule
>   Og.enchant( "models.rb" )  # classes in a separate file

I like this.  I like being able to control the magic.

- Dimitri



From aglarond at gmail.com  Wed Mar 29 01:41:11 2006
From: aglarond at gmail.com (Dimitri Aivaliotis)
Date: Wed, 29 Mar 2006 08:41:11 +0200
Subject: [Nitro] properties
In-Reply-To: <dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
Message-ID: <55c107bf0603282241y49021194gfbc427e635b7fabe@mail.gmail.com>

On 3/28/06, vseguip at gmail.com <vseguip at gmail.com> wrote:

> P.S: How about is "class_properties" or as suggested "facet_properties"?

It wasn't "facet_properties" that I suggested, just "facet", as in

class Person
  facet :firstname, :lastname
  facet :birthdate
...
end

I don't think it clashes with anything.  Might cause a little
confusion with "Ruby Facets" though, i.e. facet/core and facet/more.

- Dimitri



From bryan.a.soto at gmail.com  Wed Mar 29 01:52:53 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 22:52:53 -0800
Subject: [Nitro] properties
In-Reply-To: <dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
	<dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>
Message-ID: <bd93db0d0603282252q64792558la42fc38b7885f409@mail.gmail.com>

On 3/28/06, vseguip at gmail.com <vseguip at gmail.com> wrote:
> >
> > I (un)pulled the patch for a couple of reasons. Partly because it was
> > causing some errors with some missed renames of properties to
> > og_properties, but mostly because I was sure George would object to
> > the name. :)
> >
>
> Hi,
>   Do you remember where I missed renaming so I can add it? It would be
> very helpful to have a test case to demo it (I'm pretty sure Og as
> such passed all the test cases when I wrote the patch) and integrate
> it with the current test suites.
>  OTOH I hope we can agree on a name in this thread so I can put my ass
> to work asap.

It showed up when I ran the test suite for glue, probably where the
difference was. :)

I'm not sure I like any of the names I've seen so far. Of course, it's
not just up to me. :)  Just curious, but what is properties used for
in GTK? I'm just trying to get some ideas for a better name. I'll be
poking through Og and Glue code for inspiration, but you know GTK so
hopefully you could provide a quick summary?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Wed Mar 29 02:06:23 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 23:06:23 -0800
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
Message-ID: <bd93db0d0603282306q66a385e4hb0d6c0c80fd2d242@mail.gmail.com>

On 3/28/06, TRANS <transfire at gmail.com> wrote:
> Here's the patch for removing the glue dependency.
>

Thanks. :)

> There still remains a few details to attend. I merged lib/glue.rb into
> lib/nitro.rb. That seemed like the right thing to do (although maybe
> there's a reason to keep some separation?) In a couple of places in
> the system there's
>
>    require 'glue'
>
> I just remarked these out. They may need to be replaced with
>
>   require 'nitro'
>

We'll have to see how you managed to split things out. :)

> It will take a little bit of additional work yet to get all the kinks
> out, but I'm having issues running the tests right now. So hopefully
> someone with more expeirence in the patch "end game" can work these
> thru without much trouble.
>

What type of trouble are you having with the tests? Oh, speaking of
tests, how does one specify the load path for requires in reap for
test running? For example, in nitro, running rake test will generate
library load paths based off of the location of the rake file. Can you
do something similar in reap? I rather liked the way you handled
output, but I had problems with it loading the installed gems...

> One error I was getting that worries me is with nitro/util/paramix.rb
> --a stack level too deep error. Of course ultimately this lib should
> be replaced with facet/paramix.rb, but George mentioned some problem
> in that regard --I'm not sure what it is. Anyway, at this point it may
> just be a problem with running the tests on my end and not a real
> error at all.
>

Stack level, eh? So there's a recursive call somewhere...

> One last simple thing. I didn't remove the glue package dependency in
> the ProjectInfo file. That will need to be done, but that's an easy
> one.
>
> Oh, and this patch includes the removal of Flexob too.
>
> T.
>
> P.S. I recorded four times in the proccess, if that matters.

We'll find out. :)

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From vseguip at gmail.com  Wed Mar 29 02:12:02 2006
From: vseguip at gmail.com (vseguip at gmail.com)
Date: Wed, 29 Mar 2006 09:12:02 +0200
Subject: [Nitro] properties
In-Reply-To: <bd93db0d0603282252q64792558la42fc38b7885f409@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
	<dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>
	<bd93db0d0603282252q64792558la42fc38b7885f409@mail.gmail.com>
Message-ID: <dda3b98e0603282312i5db883b8h72a22445aef98157@mail.gmail.com>

Hi,

> I'm not sure I like any of the names I've seen so far. Of course, it's
> not just up to me. :)  Just curious, but what is properties used for
> in GTK?

It's actually used as a method to access the GLib properties of a
GObject (useful for implementing observer pattern, introspection, etc,
when in C).


Cheers,
  V. Segu?



From bryan.a.soto at gmail.com  Wed Mar 29 02:35:57 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Tue, 28 Mar 2006 23:35:57 -0800
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
Message-ID: <bd93db0d0603282335u62b3f55veeed8d0f5c2d138f@mail.gmail.com>

On 3/28/06, TRANS <transfire at gmail.com> wrote:
> Here's the patch for removing the glue dependency.
>

Hey Trans, this patch seems to only consist of the flexob removal.
Could you double check?

Thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From news at stephan.walter.name  Wed Mar 29 03:33:07 2006
From: news at stephan.walter.name (Stephan Walter)
Date: Wed, 29 Mar 2006 10:33:07 +0200
Subject: [Nitro] og relations strangeness
References: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
	<AFB2F8D2-D690-40B8-9E54-6F8A16C8FEF8@yoyo.org>
Message-ID: <pan.2006.03.29.08.33.06.885006@stephan.walter.name>

Hi Aidan,

On Wed, 29 Mar 2006 11:20:50 +1100, Aidan Rogers wrote:

> You can specify the relationship to have a different foreign name like so:
> 
> has_many :widgets, Widget, :foreign_name => :owner

I am aware that there are ways to solve that problem, but in my opinion it
should work without having to specify the :foreign_name. After all, if I
write
   belongs_to :owner, User
it is clear that I am referring to the User class.

-Stephan



From aidan at yoyo.org  Wed Mar 29 05:52:22 2006
From: aidan at yoyo.org (Aidan Rogers)
Date: Wed, 29 Mar 2006 21:52:22 +1100
Subject: [Nitro] og relations strangeness
In-Reply-To: <pan.2006.03.29.08.33.06.885006@stephan.walter.name>
References: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
	<AFB2F8D2-D690-40B8-9E54-6F8A16C8FEF8@yoyo.org>
	<pan.2006.03.29.08.33.06.885006@stephan.walter.name>
Message-ID: <01601888-D29E-48FE-A95C-C926DD15647E@yoyo.org>

Sure, but what if your class looked like this:

belongs_to :owner, User
has_one :boss, User
has_many :friends, User
has_many :siblings, User

How would the reciprocating (User) class be able to determine which  
relationship was which?  What you are suggesting only works if there  
is a single relationship between the two classes (often not the  
case).  However, if you can come up with a way to implement it such  
that there is no need for the foreign_name, code it and send a  
patch! :-)

Aidan

On 29/03/2006, at 7:33 PM, Stephan Walter wrote:

> Hi Aidan,
>
> On Wed, 29 Mar 2006 11:20:50 +1100, Aidan Rogers wrote:
>
>> You can specify the relationship to have a different foreign name  
>> like so:
>>
>> has_many :widgets, Widget, :foreign_name => :owner
>
> I am aware that there are ways to solve that problem, but in my  
> opinion it
> should work without having to specify the :foreign_name. After all,  
> if I
> write
>    belongs_to :owner, User
> it is clear that I am referring to the User class.
>
> -Stephan
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>



From riku.raisanen at walkingwoods.com  Wed Mar 29 07:07:46 2006
From: riku.raisanen at walkingwoods.com (Riku =?iso-8859-1?q?R=E4is=E4nen?=)
Date: Wed, 29 Mar 2006 15:07:46 +0300
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <mailman.971.1143613106.25833.nitro-general@rubyforge.org>
References: <mailman.971.1143613106.25833.nitro-general@rubyforge.org>
Message-ID: <200603291507.47110.riku.raisanen@walkingwoods.com>

>Stack level, eh? So there's a recursive call somewhere...

umm. umm.

glue -> nitro
require 'glue' -> require 'nitro'

nitro requires glue, which is now nitro -> recursive call

ehm.

That's it. Maybe pick a better name next time. :f


-Riku R?is?nen



From transfire at gmail.com  Wed Mar 29 07:05:46 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 07:05:46 -0500
Subject: [Nitro] og relations strangeness
In-Reply-To: <55c107bf0603282230s6871d56fpca7ac63044722048@mail.gmail.com>
References: <pan.2006.03.28.20.57.16.210439@stephan.walter.name>
	<4b6f054f0603281718o16a5ac70vf8ce02645835b729@mail.gmail.com>
	<55c107bf0603282230s6871d56fpca7ac63044722048@mail.gmail.com>
Message-ID: <4b6f054f0603290405i3685b8d2p5c945b65cc923ba1@mail.gmail.com>

On 3/29/06, Dimitri Aivaliotis <aglarond at gmail.com> wrote:

> Doesn't look like a bug to me.  Re: Aidan's response, User naturally
> looks for a field called :user_oid, unless you override it with the
> :foreign_name declaration.  To avoid this problem in the first place,
> the initial class declaration for Widget could read

Perhaps I miss understtod Stephan when he said:

"Currently Og or Nitro (I'm not sure...) is looking for a
user_oid field in the widgets table when looking up the widgets for a user
(in the admin interface for example). Which obviously won't work and give
you an Og::StoreException."

T.



From transfire at gmail.com  Wed Mar 29 07:13:19 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 07:13:19 -0500
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <bd93db0d0603282335u62b3f55veeed8d0f5c2d138f@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
	<bd93db0d0603282335u62b3f55veeed8d0f5c2d138f@mail.gmail.com>
Message-ID: <4b6f054f0603290413g74d8538erf6edfe7bd193e4b1@mail.gmail.com>

On 3/29/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Hey Trans, this patch seems to only consist of the flexob removal.
> Could you double check?

Ok. Try agan...
-------------- next part --------------
A non-text attachment was scrubbed...
Name: patchfile.tar.bz2
Type: application/x-bzip2
Size: 27919 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/12706b0b/attachment.bz2 

From transfire at gmail.com  Wed Mar 29 07:16:14 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 07:16:14 -0500
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <bd93db0d0603282306q66a385e4hb0d6c0c80fd2d242@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
	<bd93db0d0603282306q66a385e4hb0d6c0c80fd2d242@mail.gmail.com>
Message-ID: <4b6f054f0603290416o1b1f513eq131325b31d3a2d71@mail.gmail.com>

On 3/29/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:

> What type of trouble are you having with the tests? Oh, speaking of
> tests, how does one specify the load path for requires in reap for
> test running? For example, in nitro, running rake test will generate
> library load paths based off of the location of the rake file. Can you
> do something similar in reap? I rather liked the way you handled
> output, but I had problems with it loading the installed gems...

Same kind of troubles. Gems getting in the way, Hmmm.. yes I think I
can add some features to Reap to address this. I work on that today.

Is there a way to deactive gems after the fact?

T.



From transfire at gmail.com  Wed Mar 29 07:18:15 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 07:18:15 -0500
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <200603291507.47110.riku.raisanen@walkingwoods.com>
References: <mailman.971.1143613106.25833.nitro-general@rubyforge.org>
	<200603291507.47110.riku.raisanen@walkingwoods.com>
Message-ID: <4b6f054f0603290418h5afd2fdbnbd87e668d785b0be@mail.gmail.com>

On 3/29/06, Riku R?is?nen <riku.raisanen at walkingwoods.com> wrote:
> >Stack level, eh? So there's a recursive call somewhere...
>
> umm. umm.
>
> glue -> nitro
> require 'glue' -> require 'nitro'
>
> nitro requires glue, which is now nitro -> recursive call
>
> ehm.
>
> That's it. Maybe pick a better name next time. :f

Unless I'm misunderstading, I don't think so. The whole idea is to
merge glue into nitro. Not create another "nitro". The actual files
have been moved to /nitro/util.

Thanks,
T.



From transfire at gmail.com  Wed Mar 29 07:29:47 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 07:29:47 -0500
Subject: [Nitro] properties
In-Reply-To: <dda3b98e0603282312i5db883b8h72a22445aef98157@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
	<dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>
	<bd93db0d0603282252q64792558la42fc38b7885f409@mail.gmail.com>
	<dda3b98e0603282312i5db883b8h72a22445aef98157@mail.gmail.com>
Message-ID: <4b6f054f0603290429r70749027id928d7dcfa0771e3@mail.gmail.com>

Personally I always it might be better to specify persistance separate.

  class Person

    attr_accessor :firstname, :lastname, :birthdate

    persist :firstname, :lastname, String,
                :birthdate, Date
   ...
  end

But besides that, if 'property' is a problem, what's wrong with just 'prop'?

  class Person
    prop :firstname, :lastname
    prop :birthdate
    ...
  end

Then maybe, 'graph_properties'?

T.



From james_b at neurogami.com  Wed Mar 29 08:47:38 2006
From: james_b at neurogami.com (James Britt)
Date: Wed, 29 Mar 2006 06:47:38 -0700
Subject: [Nitro] how classes are enchanted
In-Reply-To: <55c107bf0603282236i1b36f775l81bc5a2e7afae42d@mail.gmail.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>	<4b6f054f0603282120s7769030g98c3684144d21d6a@mail.gmail.com>
	<55c107bf0603282236i1b36f775l81bc5a2e7afae42d@mail.gmail.com>
Message-ID: <442A8FFA.3080204@neurogami.com>

Dimitri Aivaliotis wrote:
> On 3/29/06, TRANS <transfire at gmail.com> wrote:
> 
> 
>>  Og.enchant( FooClass )      # directly on an external class
>>  Og.enchant( FooModule )   # classes in FooModule
>>  Og.enchant( "models.rb" )  # classes in a separate file
> 
> 
> I like this.  I like being able to control the magic.

Yes, very nice.


-- 
James Britt

"In physics the truth is rarely perfectly clear, and that is certainly
  universally the case in human affairs. Hence, what is not surrounded by
  uncertainty cannot be the truth."
  - R. Feynman


From transfire at gmail.com  Wed Mar 29 08:57:02 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 08:57:02 -0500
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <4b6f054f0603290416o1b1f513eq131325b31d3a2d71@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
	<bd93db0d0603282306q66a385e4hb0d6c0c80fd2d242@mail.gmail.com>
	<4b6f054f0603290416o1b1f513eq131325b31d3a2d71@mail.gmail.com>
Message-ID: <4b6f054f0603290557t24f2df7dgab4092a5ea319a85@mail.gmail.com>

On 3/29/06, TRANS <transfire at gmail.com> wrote:
> On 3/29/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
>
> > What type of trouble are you having with the tests? Oh, speaking of
> > tests, how does one specify the load path for requires in reap for
> > test running? For example, in nitro, running rake test will generate
> > library load paths based off of the location of the rake file. Can you
> > do something similar in reap? I rather liked the way you handled
> > output, but I had problems with it loading the installed gems...
>
> Same kind of troubles. Gems getting in the way, Hmmm.. yes I think I
> can add some features to Reap to address this. I work on that today.

Hey hey. I got this working! Reap can not test Nitro. I may some
change to the test task:

> reap help test

test: Run unit-tests (each in a separate process).

  Attributes

    files       Test files (eg. test/tc_**/*.rb)
                Defaults to typcial selection.

    requires    List of any files to pre-require.

    libs        List of lookup directories to include in
                load path. './lib' is always included.

    live        Flag to quickly deactive use of local libs.
                Test against installed files instead.


And yes. paramix is bombing :(

T.



From kashia at vfemail.net  Wed Mar 29 08:57:50 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Wed, 29 Mar 2006 15:57:50 +0200
Subject: [Nitro] Hiding a column from Og?
Message-ID: <op.s66g2ocag3glhz@jo.local>

Hi,

how do I create a column in a database which is not noticed by Og?

I want to add an additional column to a table, and want to avoid Og messing with it or even noticing it, is that possible?

http://oxyliquit.de/question/23

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From george.moschovitis at gmail.com  Wed Mar 29 10:49:30 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 18:49:30 +0300
Subject: [Nitro] [PATCH] Configuration + Og startup speedup
In-Reply-To: <bd93db0d0603271214i229f66a8w11b1d00ca07162e1@mail.gmail.com>
References: <afd80b540603271106k635ebab6r44dd6e5b3b29378e@mail.gmail.com>
	<bd93db0d0603271214i229f66a8w11b1d00ca07162e1@mail.gmail.com>
Message-ID: <afd80b540603290749y68401afbr60f030bfaa10ed94@mail.gmail.com>

Yeah,

here is the patch, please submit it ;-)

-g.

On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Hey g,
>
> You forgot the patch. :)
>
> Thanks
>
> On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Bryan, can you add these to the stable repo?
> >
> > thanks,
> > -g.
> >
> >
> > Mon Mar 27 21:59:51 EEST 2006  George Moschovitis <gm at navel.gr>
> >   * Startup optimization, don't search in ObjectSpace for managed classes.
> >
> > Mon Mar 27 21:43:44 EEST 2006  George Moschovitis <gm at navel.gr>
> >   * Further improved configuration system, even added some ruby magic,
> > check out the rdocs and the updated test case.
> >
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle
Type: application/octet-stream
Size: 45207 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/58b305ba/attachment.obj 

From george.moschovitis at gmail.com  Wed Mar 29 11:36:32 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 19:36:32 +0300
Subject: [Nitro] cattr
In-Reply-To: <4b6f054f0603271215u8b608c0l8f338f3aebeddea7@mail.gmail.com>
References: <4b6f054f0603271215u8b608c0l8f338f3aebeddea7@mail.gmail.com>
Message-ID: <afd80b540603290836l4eead600j20d159935f02f1d4@mail.gmail.com>

On 3/27/06, TRANS <transfire at gmail.com> wrote:
> Currently glue/attribute defines the cattr_... mtheods to take default
> values as a second parameter. Eg.
>
>   cattr_accessor :foo, "default"
>
> This mean only one attribute can be defined per call. Facets works
> witha hash instead.
>
>   cattr_accessor :foo => "default"
>
> Which allows multiple attributes to be defined with one call.
>
> I would like to change this to the Facets way in Nitro.
>
> Other's take?


prabably allow bot ways ;-) ie either a hash or the current method.
what do you think?

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar 29 11:38:48 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 19:38:48 +0300
Subject: [Nitro] Glue move
In-Reply-To: <4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>
References: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
	<bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>
	<4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>
Message-ID: <afd80b540603290838t79b68f3es88145b0b0571c639@mail.gmail.com>

> Oh. Is that change going to happen then then?

yeah, for the moment send the patch to bryan. I would like to publicly
ask Zimba to give you ssh access to the repo though.

-g.



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar 29 11:39:28 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 19:39:28 +0300
Subject: [Nitro] More Great Reap News
In-Reply-To: <4b6f054f0603272325p96968cdt222b3c6f5f942bac@mail.gmail.com>
References: <4b6f054f0603272325p96968cdt222b3c6f5f942bac@mail.gmail.com>
Message-ID: <afd80b540603290839y55c43d01gfb16aa99f1ef5579@mail.gmail.com>

nice, but where is that rubyforge feature ? ;-) ;-) ;-)

-g.

On 3/28/06, TRANS <transfire at gmail.com> wrote:
> I just added another feature to Reap. I had almost forgotten all about
> this idea from many months ago when I first implemented test
> extraction....
>
> Now Reap includes a bin file called 'rubytest'. It will automatically
> run the embedded test in a source file, so there's no longer any need
> to manually remark out the '=begin test' and '=end' to do a local
> development test. Eg.
>
>   # foo.rb
>
>   def x ; 1 ; end
>
>   =begin test
>
>     class TCFoo < Test::Unit::TestCase
>
>        def test_x
>          assert_equal( 1, x )
>        end
>
>     end
>
>   =end
>
> then on the command line:
>
>   % rubytest foo.rb
>   Loaded suite /usr/bin/rubytest
>   Started
>   .
>   Finished in 0.010354 seconds.
>
>   1 tests, 1 assertions, 0 failures, 0 errors
>
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar 29 11:52:32 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 19:52:32 +0300
Subject: [Nitro] properties
In-Reply-To: <4b6f054f0603290429r70749027id928d7dcfa0771e3@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
	<dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>
	<bd93db0d0603282252q64792558la42fc38b7885f409@mail.gmail.com>
	<dda3b98e0603282312i5db883b8h72a22445aef98157@mail.gmail.com>
	<4b6f054f0603290429r70749027id928d7dcfa0771e3@mail.gmail.com>
Message-ID: <afd80b540603290852r70b9bafar63cc5ade1c23bdca@mail.gmail.com>

On 3/29/06, TRANS <transfire at gmail.com> wrote:
> Personally I always it might be better to specify persistance separate.
>
>   class Person
>
>     attr_accessor :firstname, :lastname, :birthdate
>
>     persist :firstname, :lastname, String,
>                 :birthdate, Date
>    ...
>   end


you can do that...

class Person
  attr_accessor ...

  ann ...
end

-g.
--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar 29 11:56:50 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 19:56:50 +0300
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <4b6f054f0603290557t24f2df7dgab4092a5ea319a85@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
	<bd93db0d0603282306q66a385e4hb0d6c0c80fd2d242@mail.gmail.com>
	<4b6f054f0603290416o1b1f513eq131325b31d3a2d71@mail.gmail.com>
	<4b6f054f0603290557t24f2df7dgab4092a5ea319a85@mail.gmail.com>
Message-ID: <afd80b540603290856g68c44d4bwaa8da85d9328fc3@mail.gmail.com>

> And yes. paramix is bombing :(

yeah :(

-g.


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From george.moschovitis at gmail.com  Wed Mar 29 12:33:09 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 20:33:09 +0300
Subject: [Nitro] Bounties
Message-ID: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>

Dear devs,

i am wondering if anyone has the time to contribute 2 useful methods to facets:

obj.to_xml
obj.to_json

both methods are obvious and quite useful.

anyone up to the job? ;-)

regards,
George.



--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From transfire at gmail.com  Wed Mar 29 12:44:52 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 12:44:52 -0500
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <4b6f054f0603290557t24f2df7dgab4092a5ea319a85@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
	<bd93db0d0603282306q66a385e4hb0d6c0c80fd2d242@mail.gmail.com>
	<4b6f054f0603290416o1b1f513eq131325b31d3a2d71@mail.gmail.com>
	<4b6f054f0603290557t24f2df7dgab4092a5ea319a85@mail.gmail.com>
Message-ID: <4b6f054f0603290944u5b98baebwdcbb28255b937f05@mail.gmail.com>

On 3/29/06, TRANS <transfire at gmail.com> wrote:
> Hey hey. I got this working! Reap can not test Nitro. I may some
> change to the test task:

  s/not/now/
  s/may/made/

Really wish I had a proof reader What's this computer good for if it
can't read my mind anyway? ;-)



From transfire at gmail.com  Wed Mar 29 12:47:30 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 12:47:30 -0500
Subject: [Nitro] Glue move
In-Reply-To: <afd80b540603290838t79b68f3es88145b0b0571c639@mail.gmail.com>
References: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
	<bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>
	<4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>
	<afd80b540603290838t79b68f3es88145b0b0571c639@mail.gmail.com>
Message-ID: <4b6f054f0603290947m20c41f49mde32b898d5fc27fa@mail.gmail.com>

On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Oh. Is that change going to happen then then?
>
> yeah, for the moment send the patch to bryan. I would like to publicly
> ask Zimba to give you ssh access to the repo though.

He already have. He didin't specif ssh, only the repository. But might it work?



From transfire at gmail.com  Wed Mar 29 12:48:41 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 12:48:41 -0500
Subject: [Nitro] Glue move
In-Reply-To: <4b6f054f0603290947m20c41f49mde32b898d5fc27fa@mail.gmail.com>
References: <4b6f054f0603260908h6ecec886re9f5422abf75dc63@mail.gmail.com>
	<bd93db0d0603271437x233858eci33865bd0fb126842@mail.gmail.com>
	<4b6f054f0603271550x603b52feie75dc4eedd664a22@mail.gmail.com>
	<afd80b540603290838t79b68f3es88145b0b0571c639@mail.gmail.com>
	<4b6f054f0603290947m20c41f49mde32b898d5fc27fa@mail.gmail.com>
Message-ID: <4b6f054f0603290948h25026013k8e92e1be8db09cc0@mail.gmail.com>

On 3/29/06, TRANS <transfire at gmail.com> wrote:
> On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > > Oh. Is that change going to happen then then?
> >
> > yeah, for the moment send the patch to bryan. I would like to publicly
> > ask Zimba to give you ssh access to the repo though.
>
> He already have. He didin't specif ssh, only the repository. But might it work?

He *may* already have.

anyway I'll try it shortly if I don't hear from anything otherwise.

T.



From transfire at gmail.com  Wed Mar 29 12:50:57 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 12:50:57 -0500
Subject: [Nitro] cattr
In-Reply-To: <afd80b540603290836l4eead600j20d159935f02f1d4@mail.gmail.com>
References: <4b6f054f0603271215u8b608c0l8f338f3aebeddea7@mail.gmail.com>
	<afd80b540603290836l4eead600j20d159935f02f1d4@mail.gmail.com>
Message-ID: <4b6f054f0603290950q64d0413ag58600c422a800d70@mail.gmail.com>

On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:

> prabably allow bot ways ;-) ie either a hash or the current method.
> what do you think?

Thank that's reasonably feasible? Well, I can give it a go and see how
it works out.

T.



From transfire at gmail.com  Wed Mar 29 12:54:13 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 12:54:13 -0500
Subject: [Nitro] properties
In-Reply-To: <afd80b540603290852r70b9bafar63cc5ade1c23bdca@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
	<dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>
	<bd93db0d0603282252q64792558la42fc38b7885f409@mail.gmail.com>
	<dda3b98e0603282312i5db883b8h72a22445aef98157@mail.gmail.com>
	<4b6f054f0603290429r70749027id928d7dcfa0771e3@mail.gmail.com>
	<afd80b540603290852r70b9bafar63cc5ade1c23bdca@mail.gmail.com>
Message-ID: <4b6f054f0603290954u72ab3d25n7b9ddc27bc162e61@mail.gmail.com>

On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
>
> class Person
>   attr_accessor ...
>
>   ann ...
> end

Right. But that won't enchant the class, will it? You'd still had to
use #property instead of #ann or enchant the class manually --which of
coure is fine by me, I'm just saying.

T.



From transfire at gmail.com  Wed Mar 29 12:58:00 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 12:58:00 -0500
Subject: [Nitro] Bounties
In-Reply-To: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
References: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
Message-ID: <4b6f054f0603290958peea5cf3hb831b74291899cb2@mail.gmail.com>

On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Dear devs,
>
> i am wondering if anyone has the time to contribute 2 useful methods to facets:
>
> obj.to_xml
> obj.to_json

I may do it soon.... but feel free to beat me too it.

There's plenty of open source code out there to use as a jumping
point. But if you port be sure the source is Ruby or GPL license and
include the original copyright.

BTW, doesn't REXML have a #to_xml method?

T.



From george.moschovitis at gmail.com  Wed Mar 29 13:00:07 2006
From: george.moschovitis at gmail.com (George Moschovitis)
Date: Wed, 29 Mar 2006 21:00:07 +0300
Subject: [Nitro] properties
In-Reply-To: <4b6f054f0603290954u72ab3d25n7b9ddc27bc162e61@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<dda3b98e0603280654t2c207fc2la57b20f2ccc3fdab@mail.gmail.com>
	<bd93db0d0603281235t34078e77s2d7d661d301e9b98@mail.gmail.com>
	<dda3b98e0603282233t2d1b1d5eu17b133aad2cc9b17@mail.gmail.com>
	<bd93db0d0603282252q64792558la42fc38b7885f409@mail.gmail.com>
	<dda3b98e0603282312i5db883b8h72a22445aef98157@mail.gmail.com>
	<4b6f054f0603290429r70749027id928d7dcfa0771e3@mail.gmail.com>
	<afd80b540603290852r70b9bafar63cc5ade1c23bdca@mail.gmail.com>
	<4b6f054f0603290954u72ab3d25n7b9ddc27bc162e61@mail.gmail.com>
Message-ID: <afd80b540603291000q6879019cy4a4f5e82aa431f3@mail.gmail.com>

prop uses ann under the hood ;-) hey you can even use prop

class Person
  attr_accessor ...

  property ...
end


On 3/29/06, TRANS <transfire at gmail.com> wrote:
> On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> >
> > class Person
> >   attr_accessor ...
> >
> >   ann ...
> > end
>
> Right. But that won't enchant the class, will it? You'd still had to
> use #property instead of #ann or enchant the class manually --which of
> coure is fine by me, I'm just saying.
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
http://www.gmosx.com
http://www.navel.gr
http://www.nitrohq.com



From kgergely at mcl.hu  Wed Mar 29 13:48:17 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Wed, 29 Mar 2006 20:48:17 +0200
Subject: [Nitro] form controls
In-Reply-To: <20060326222039.GC3704@mlabdial.hit.bme.hu>
References: <pan.2006.03.26.14.17.53.96319@stephan.walter.name>
	<20060326222039.GC3704@mlabdial.hit.bme.hu>
Message-ID: <20060329184817.GA7997@mlabdial.hit.bme.hu>

Gergely Kontra azt ?rta, hogy: 

found lib/nitro/helper/form/controls.rb

Still don't know what the heck are the parameters...

Form::FixnumControl.new( ????? )

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/3fa14061/attachment.bin 

From kgergely at mcl.hu  Wed Mar 29 13:52:13 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Wed, 29 Mar 2006 20:52:13 +0200
Subject: [Nitro] Server.map and paths?
In-Reply-To: <4429D639.4020504@neurogami.com>
References: <20060327182555.GA11490@mlabdial.hit.bme.hu>
	<bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>
	<20060328220959.GC27000@mlabdial.hit.bme.hu>
	<4429D639.4020504@neurogami.com>
Message-ID: <20060329185213.GB7997@mlabdial.hit.bme.hu>

James Britt azt ?rta, hogy: 
> Can you explain  bit more what your did?

I'd like to develop componenets for my apps: like auth, admin, and so 
on.
So my directory structure looks like:
/
- admin
  - putlic #admin's templates
- auth
  - public #auth's template

the controllers are in the admin and auth dirs.
So, the controller will tell to nitro where are it's own templates, like 
this:

  def self.setup_template_root(path) # hack
    super
    @template_root <<  File.join(File.dirname(__FILE__), 'public')
  end

# don't know why is it so complicated, but works...

Sad to say, that images, and stuff like this cannot go into subdirs, 
because it is served by the web server :(

So, the basic idea, that installing a module is just copying ONE 
directory tree is broken...

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/80555fc5/attachment.bin 

From transfire at gmail.com  Wed Mar 29 13:59:48 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 13:59:48 -0500
Subject: [Nitro] Glue Patch Away!
In-Reply-To: <afd80b540603290856g68c44d4bwaa8da85d9328fc3@mail.gmail.com>
References: <4b6f054f0603281553s31bbccf1m8c227fcad0af1b0c@mail.gmail.com>
	<bd93db0d0603282306q66a385e4hb0d6c0c80fd2d242@mail.gmail.com>
	<4b6f054f0603290416o1b1f513eq131325b31d3a2d71@mail.gmail.com>
	<4b6f054f0603290557t24f2df7dgab4092a5ea319a85@mail.gmail.com>
	<afd80b540603290856g68c44d4bwaa8da85d9328fc3@mail.gmail.com>
Message-ID: <4b6f054f0603291059x23bd0fa7md9ce1e22a02aea48@mail.gmail.com>

On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > And yes. paramix is bombing :(
>
> yeah :(

Looks liek I figured it out. Problem was that the facet,
module/basename, was not taking into account anonymous modules and was
bombing when a module's name came up blank ("") --which was occuring
in paramix.rb.

This looks like the fix. Change 'facet/core/module/basename' to

class Module

  # Returns the root name of the module/class.
  #
  #   module Example
  #     class Demo
  #     end
  #   end
  #
  #   Demo.name       #=> Example::Demo
  #   Demo.basename   #=> Demo
  #
  # For anonymous modules this will provide a basename
  # based on Module#inspect.
  #
  #   m = Module.new
  #   m.inspect       #=> "#<Module:0xb7bb0434>"
  #   m.basename      #=> "Module_0xb7bb0434"
  #
  def basename
    if self.name
      self.name.gsub(/^.*::/, '')
    else
      self.inspect.gsub('#<','').gsub('>','').sub(':', '_')
    end
  end

end


T.



From james_b at neurogami.com  Wed Mar 29 14:00:45 2006
From: james_b at neurogami.com (James Britt)
Date: Wed, 29 Mar 2006 12:00:45 -0700
Subject: [Nitro] Bounties
In-Reply-To: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
References: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
Message-ID: <442AD95D.1050209@neurogami.com>

George Moschovitis wrote:
> Dear devs,
> 
> i am wondering if anyone has the time to contribute 2 useful methods to facets:
> 
> obj.to_xml
> obj.to_json
> 
> both methods are obvious and quite useful.
> 
> anyone up to the job? ;-)

As I recall, including the ruby-json gem grants the to_json method.

I used that for my OrbJSON library (which, oddly enough, I never 
actually use.)


The to_xml is trickier; XML is just syntax, so there are endless ways to 
express an object as XML.  It'd rather think about to_atom or to_rdf or 
some way to export objects in a generally understood XML format, 
something where other tools already understand the semantics.

Are there well-defined XML object serialization formats? Is there a 
front-runner?



-- 
James Britt

"You harmonize; then you customize."
  - Wilson Pickett


From transfire at gmail.com  Wed Mar 29 14:43:29 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 14:43:29 -0500
Subject: [Nitro] Bounties
In-Reply-To: <442AD95D.1050209@neurogami.com>
References: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
	<442AD95D.1050209@neurogami.com>
Message-ID: <4b6f054f0603291143h3f65e8e9ic8f3c59d7920d6bc@mail.gmail.com>

On 3/29/06, James Britt <james_b at neurogami.com> wrote:
> George Moschovitis wrote:
> > Dear devs,
> >
> > i am wondering if anyone has the time to contribute 2 useful methods to facets:
> >
> > obj.to_xml
> > obj.to_json
> >
> > both methods are obvious and quite useful.
> >
> > anyone up to the job? ;-)
>
> As I recall, including the ruby-json gem grants the to_json method.
>
> I used that for my OrbJSON library (which, oddly enough, I never
> actually use.)
>
>
> The to_xml is trickier; XML is just syntax, so there are endless ways to
> express an object as XML.  It'd rather think about to_atom or to_rdf or
> some way to export objects in a generally understood XML format,
> something where other tools already understand the semantics.
>
> Are there well-defined XML object serialization formats? Is there a
> front-runner?

SOAP? XOXO?

T.



From rob at motionpath.com  Wed Mar 29 14:50:52 2006
From: rob at motionpath.com (rob)
Date: Wed, 29 Mar 2006 20:50:52 +0100
Subject: [Nitro] Bounties
In-Reply-To: <4b6f054f0603291143h3f65e8e9ic8f3c59d7920d6bc@mail.gmail.com>
References: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
	<442AD95D.1050209@neurogami.com>
	<4b6f054f0603291143h3f65e8e9ic8f3c59d7920d6bc@mail.gmail.com>
Message-ID: <66B15D3A-CD42-4D1A-B566-5AEEB639891A@motionpath.com>

I added a method Model.from_json that makes records from a json input  
if there's nothing like that in Nitro already and anyone wanted it  
they could have it...

On 29 Mar 2006, at 20:43, TRANS wrote:

> On 3/29/06, James Britt <james_b at neurogami.com> wrote:
>> George Moschovitis wrote:
>>> Dear devs,
>>>
>>> i am wondering if anyone has the time to contribute 2 useful  
>>> methods to facets:
>>>
>>> obj.to_xml
>>> obj.to_json
>>>
>>> both methods are obvious and quite useful.
>>>
>>> anyone up to the job? ;-)
>>
>> As I recall, including the ruby-json gem grants the to_json method.
>>
>> I used that for my OrbJSON library (which, oddly enough, I never
>> actually use.)
>>
>>
>> The to_xml is trickier; XML is just syntax, so there are endless  
>> ways to
>> express an object as XML.  It'd rather think about to_atom or  
>> to_rdf or
>> some way to export objects in a generally understood XML format,
>> something where other tools already understand the semantics.
>>
>> Are there well-defined XML object serialization formats? Is there a
>> front-runner?
>
> SOAP? XOXO?
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From bryan.a.soto at gmail.com  Wed Mar 29 15:10:04 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 29 Mar 2006 12:10:04 -0800
Subject: [Nitro] Server.map and paths?
In-Reply-To: <20060329185213.GB7997@mlabdial.hit.bme.hu>
References: <20060327182555.GA11490@mlabdial.hit.bme.hu>
	<bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>
	<20060328220959.GC27000@mlabdial.hit.bme.hu>
	<4429D639.4020504@neurogami.com>
	<20060329185213.GB7997@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603291210p7c86449fyae1c9d83bf615c90@mail.gmail.com>

On 3/29/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> James Britt azt ?rta, hogy:
> > Can you explain  bit more what your did?
>
> I'd like to develop componenets for my apps: like auth, admin, and so
> on.
> So my directory structure looks like:
> /
> - admin
>   - putlic #admin's templates
> - auth
>   - public #auth's template
>
> the controllers are in the admin and auth dirs.
> So, the controller will tell to nitro where are it's own templates, like
> this:
>
>   def self.setup_template_root(path) # hack
>     super
>     @template_root <<  File.join(File.dirname(__FILE__), 'public')
>   end
>
> # don't know why is it so complicated, but works...
>

George, help! ;)

> Sad to say, that images, and stuff like this cannot go into subdirs,
> because it is served by the web server :(
>

Well, that's not exactly true. :)

The web server punts when it can't find a static file to serve. What
you return from Nitro is entirely up to you. Most everyone just
responds with xhtml, but that's not a requirement.

Take part/admin as an example. It has a nice little stylesheet to make
things pretty located in part/admin/system.css.

In the skin.rb file, there is the following line:

<link rel="stylesheet" href="stylesheet/system" type="text/css"
media="screen" />

This causes the web browser to load /admin/stylesheet/system. There
isn't an existing file with this name, so the web server punts to
nitro again. It's handled by this method in part/admin/controller.rb:

  # fetch the css for this controller
  # checks for a /system.css file and uses that if available
  # otherwise uses system.css from the part/admin dir

  def stylesheet(sheet='system')
    sheet << ".css"
    [Nitro::Server.public_root, File.dirname(__FILE__)].each do |path|
      css_file = File.join(path, sheet)
      if File.exists? css_file
        response.content_type = 'text/css'
        @out << File.read(css_file)
        raise Nitro::RenderExit
      end
    end
    raise RenderExit
  end

Combine this with caching and you'd have a pretty portable, low-cost
method of serving images.

Hope this helps,

Bryan

> So, the basic idea, that installing a module is just copying ONE
> directory tree is broken...
>
> Gergo

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Wed Mar 29 15:18:25 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 15:18:25 -0500
Subject: [Nitro] Bounties
In-Reply-To: <66B15D3A-CD42-4D1A-B566-5AEEB639891A@motionpath.com>
References: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
	<442AD95D.1050209@neurogami.com>
	<4b6f054f0603291143h3f65e8e9ic8f3c59d7920d6bc@mail.gmail.com>
	<66B15D3A-CD42-4D1A-B566-5AEEB639891A@motionpath.com>
Message-ID: <4b6f054f0603291218o4964b4c3oad1c64ec10a01dae@mail.gmail.com>

On 3/29/06, rob <rob at motionpath.com> wrote:
> I added a method Model.from_json that makes records from a json input
> if there's nothing like that in Nitro already and anyone wanted it
> they could have it...

send it on over. I'll have a look.

T.



From kashia at vfemail.net  Wed Mar 29 15:17:38 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Wed, 29 Mar 2006 22:17:38 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
Message-ID: <op.s66ynonqg3glhz@jo.local>

Hey all,

I've been working on a requested feature, a printing view for tutorials:

http://oxyliquit.de/cache/pdf_cache/A_small_project_with_Og.pdf

this has been automatically created, tell me what you think and weither
the quality is sufficient.

Other news are:

  * Working links on the index page
  * Neat little images on the index page (thx bro)

And thank you for making me aware of bugs, you know who you are ;)

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From bryan.a.soto at gmail.com  Wed Mar 29 15:44:46 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 29 Mar 2006 12:44:46 -0800
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <op.s66ynonqg3glhz@jo.local>
References: <op.s66ynonqg3glhz@jo.local>
Message-ID: <bd93db0d0603291244t185b900cx5422eb1150a7be98@mail.gmail.com>

On 3/29/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hey all,
>
> I've been working on a requested feature, a printing view for tutorials:
>

Wow, very cool! How did you do it? I've been wanting to generate pdfs
for some forms I have.

> http://oxyliquit.de/cache/pdf_cache/A_small_project_with_Og.pdf
>
> this has been automatically created, tell me what you think and weither
> the quality is sufficient.
>

The quality looks okay to me. The code is rather light in comparison
to the text though.

> Other news are:
>
>   * Working links on the index page
>   * Neat little images on the index page (thx bro)
>

It's coming along great! Congrats!

> And thank you for making me aware of bugs, you know who you are ;)
>
> Kashia

Oh, I answered your question about fields that Og doesn't touch with
the lame answer you already knew. Just rate it as a 0 and be done with
it! ;)

Seriously though, I don't think there is any way. Og just assumes you
want it to manage everything. Perhaps a new annotation like hidden or
ignore or something should be made? I wonder how common a need this
is?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Wed Mar 29 16:19:47 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 29 Mar 2006 13:19:47 -0800
Subject: [Nitro] Bounties
In-Reply-To: <4b6f054f0603291218o4964b4c3oad1c64ec10a01dae@mail.gmail.com>
References: <afd80b540603290933x504db509j5b41d577b58eea4e@mail.gmail.com>
	<442AD95D.1050209@neurogami.com>
	<4b6f054f0603291143h3f65e8e9ic8f3c59d7920d6bc@mail.gmail.com>
	<66B15D3A-CD42-4D1A-B566-5AEEB639891A@motionpath.com>
	<4b6f054f0603291218o4964b4c3oad1c64ec10a01dae@mail.gmail.com>
Message-ID: <bd93db0d0603291319i659da3aen9cb96ee5033a10e8@mail.gmail.com>

Just to let everyone know, I've added these and George's previous
Bounties to the devlab wiki.

http://devlab.oree.ch/trac/nitrohq/wiki/Bounties

On 3/29/06, TRANS <transfire at gmail.com> wrote:
> On 3/29/06, rob <rob at motionpath.com> wrote:
> > I added a method Model.from_json that makes records from a json input
> > if there's nothing like that in Nitro already and anyone wanted it
> > they could have it...
>
> send it on over. I'll have a look.
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From kgergely at mcl.hu  Wed Mar 29 17:05:36 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Thu, 30 Mar 2006 00:05:36 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <op.s66ynonqg3glhz@jo.local>
References: <op.s66ynonqg3glhz@jo.local>
Message-ID: <20060329220536.GA12019@mlabdial.hit.bme.hu>

> Hey all,
> 
> I've been working on a requested feature, a printing view for tutorials:
> 
> http://oxyliquit.de/cache/pdf_cache/A_small_project_with_Og.pdf
> 
> this has been automatically created, tell me what you think and weither
> the quality is sufficient.

Hi!

I assume you need some critism. Here you are :)
Anyway, I never produced pdf directly, only through LaTeX

First, headings appear at the end of the page. This is bad.

The image is not labeled, not referenced, so if you have more images, it 
will not be obvious, which image you are talking about.

Pdf generation is a great thing, but for a tutorial, I don't think you 
should work so much. A correct css stylesheet can do.

Anyway, if you can solve automagically the mentioned things, I'm also 
interested how can one do such thing :)

Gerg?
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/57358150/attachment.bin 

From kgergely at mcl.hu  Wed Mar 29 17:09:57 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Thu, 30 Mar 2006 00:09:57 +0200
Subject: [Nitro] Server.map and paths?
In-Reply-To: <bd93db0d0603291210p7c86449fyae1c9d83bf615c90@mail.gmail.com>
References: <20060327182555.GA11490@mlabdial.hit.bme.hu>
	<bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>
	<20060328220959.GC27000@mlabdial.hit.bme.hu>
	<4429D639.4020504@neurogami.com>
	<20060329185213.GB7997@mlabdial.hit.bme.hu>
	<bd93db0d0603291210p7c86449fyae1c9d83bf615c90@mail.gmail.com>
Message-ID: <20060329220957.GB12019@mlabdial.hit.bme.hu>

Bryan Soto azt ?rta, hogy: 
> > Sad to say, that images, and stuff like this cannot go into subdirs,
> > because it is served by the web server :(
> >
> 
> Well, that's not exactly true. :)
> 
> The web server punts when it can't find a static file to serve. What
> you return from Nitro is entirely up to you. Most everyone just
> responds with xhtml, but that's not a requirement.
> 
>   def stylesheet(sheet='system')
>     sheet << ".css"
>     [Nitro::Server.public_root, File.dirname(__FILE__)].each do |path|
>       css_file = File.join(path, sheet)
>       if File.exists? css_file
>         response.content_type = 'text/css'
>         @out << File.read(css_file)
>         raise Nitro::RenderExit
>       end
>     end
>     raise RenderExit
>   end
> 
> Combine this with caching and you'd have a pretty portable, low-cost
> method of serving images.

Well
1.) portable? Portable for ONE module. You cannot refactor it because of 
    the __FILE__ thing IMHO
2.) low cost? Piping an image through nitro? Are you sure?

Gergo
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/24c5f59c/attachment.bin 

From kgergely at mcl.hu  Wed Mar 29 17:22:40 2006
From: kgergely at mcl.hu (Gergely Kontra)
Date: Thu, 30 Mar 2006 00:22:40 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <op.s66ynonqg3glhz@jo.local>
References: <op.s66ynonqg3glhz@jo.local>
Message-ID: <20060329222240.GC12019@mlabdial.hit.bme.hu>

Kashia Buch azt ?rta, hogy: 
> Other news are:
> 
>   * Working links on the index page
>   * Neat little images on the index page (thx bro)
Hi all!

New bug :)
When you login, you are 'kicked' soon, so it forgets me in minutes
Cannot edit my own question
Cannot rerate an answer (maybe it's a feature)
-- 
+-[ Gergely Kontra <kgergely at mcl.hu> http://www.mcl.hu/~kgergely ]------+
| PhD Student (Room:IB113)  PGP ID: 0x66875624  Mobile:(+36 20)356 9656 |
| Budapest University of Technology and Economics        ICQ# 175564914 |
+------------- "Olyan langesz vagyok, hogy poroltoval kellene jarnom!" -+
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/8a7b1915/attachment.bin 

From bryan.a.soto at gmail.com  Wed Mar 29 17:30:16 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 29 Mar 2006 14:30:16 -0800
Subject: [Nitro] Server.map and paths?
In-Reply-To: <20060329220957.GB12019@mlabdial.hit.bme.hu>
References: <20060327182555.GA11490@mlabdial.hit.bme.hu>
	<bd93db0d0603271625s48bb7850v5e33a79d8d0c915e@mail.gmail.com>
	<20060328220959.GC27000@mlabdial.hit.bme.hu>
	<4429D639.4020504@neurogami.com>
	<20060329185213.GB7997@mlabdial.hit.bme.hu>
	<bd93db0d0603291210p7c86449fyae1c9d83bf615c90@mail.gmail.com>
	<20060329220957.GB12019@mlabdial.hit.bme.hu>
Message-ID: <bd93db0d0603291430m4a6acbdaod0073e02295488d2@mail.gmail.com>

On 3/29/06, Gergely Kontra <kgergely at mcl.hu> wrote:
> Bryan Soto azt ?rta, hogy:
> > > Sad to say, that images, and stuff like this cannot go into subdirs,
> > > because it is served by the web server :(
> > >
> >
> > Well, that's not exactly true. :)
> >
> > The web server punts when it can't find a static file to serve. What
> > you return from Nitro is entirely up to you. Most everyone just
> > responds with xhtml, but that's not a requirement.
> >
> >   def stylesheet(sheet='system')
> >     sheet << ".css"
> >     [Nitro::Server.public_root, File.dirname(__FILE__)].each do |path|
> >       css_file = File.join(path, sheet)
> >       if File.exists? css_file
> >         response.content_type = 'text/css'
> >         @out << File.read(css_file)
> >         raise Nitro::RenderExit
> >       end
> >     end
> >     raise RenderExit
> >   end
> >
> > Combine this with caching and you'd have a pretty portable, low-cost
> > method of serving images.
>
> Well
> 1.) portable? Portable for ONE module. You cannot refactor it because of
>     the __FILE__ thing IMHO

Yeah, I don't quite understand that template root system.

> 2.) low cost? Piping an image through nitro? Are you sure?

When combined with caching, it should be no different than serving a
static file, no? That's ultimately what caching does, writes output to
a static file in the public directory that the web server can then
serve on the next request. The first request would be a hit, yes, but
after that, Nitro wouldn't even enter the loop.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Wed Mar 29 18:56:37 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 18:56:37 -0500
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <op.s66ynonqg3glhz@jo.local>
References: <op.s66ynonqg3glhz@jo.local>
Message-ID: <4b6f054f0603291556h6bfab4e1s8692400fa72377c4@mail.gmail.com>

Very nice. Your 1/20th of the way to a book! ;-)



From transfire at gmail.com  Wed Mar 29 19:21:28 2006
From: transfire at gmail.com (TRANS)
Date: Wed, 29 Mar 2006 19:21:28 -0500
Subject: [Nitro] Patch bundle ... we're close now
Message-ID: <4b6f054f0603291621x725ddefdjb851fc1af4bc1741@mail.gmail.com>

Here's a new patch bundle. I've gotten very close to getting the to
Nitro pass all tests with Glue moved.

I've released a new version of Facets 1.2.1. This fixes a bug with
module/basename which was causing a problem with paramix.rb, namely
that it could not handle anonymous modules. I fixed it to create a
basename based on the object_id if the module is anonymous. That was
the best idea I could think of instead of throwing an error
--hopefully that'll work okay.

Also note that I have renamed kernel/meta to kernel/instance --for
those whol might not know about this method, it alows you to get
information about an object even if the methods normally used have
been overridden. For example:

  class X
    def class ; "OBSCURED" ; end
  end

  x = X.new
  x.class                #=> "OBSCURED"
  x.instance.class  #=> X

Let me know if anyont thinks there's a problem with this new name or
has a suggestion for a better one. Thanks.

Lastly, I'll be releasing a nw version of Reap tonight too, since I
got testing working better and it now plays nice with nitro. (Still
need to finish the deb and release tasks though.)

T.

P.S. Hey Zimba. How are we doing on getting a Facets repo? Don't want
to get too far a field here before that happens. Thnx.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bundle.tar.bz2
Type: application/x-bzip2
Size: 28184 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/nitro-general/attachments/20060329/20543f77/attachment.bz2 

From bryan.a.soto at gmail.com  Wed Mar 29 19:55:51 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Wed, 29 Mar 2006 16:55:51 -0800
Subject: [Nitro] Patch bundle ... we're close now
In-Reply-To: <4b6f054f0603291621x725ddefdjb851fc1af4bc1741@mail.gmail.com>
References: <4b6f054f0603291621x725ddefdjb851fc1af4bc1741@mail.gmail.com>
Message-ID: <bd93db0d0603291655i7815d5f2g1761dcd547500e37@mail.gmail.com>

On 3/29/06, TRANS <transfire at gmail.com> wrote:
> Here's a new patch bundle. I've gotten very close to getting the to
> Nitro pass all tests with Glue moved.
>

Okay, cool. Seems to include your previous patchset, so we can ignore that one.

> I've released a new version of Facets 1.2.1. This fixes a bug with
> module/basename which was causing a problem with paramix.rb, namely
> that it could not handle anonymous modules. I fixed it to create a
> basename based on the object_id if the module is anonymous. That was
> the best idea I could think of instead of throwing an error
> --hopefully that'll work okay.
>

I noticed your update of Facets. :)
Do you have any immediate plans for new releases of Facets? If not, we
should probably put out an update for Nitro, like 29.1 or something
for compatibility with your newer versions. I think the repo is fairly
compatible with 29.0, but has a few bugfixes (routing comes to mind).
We still have some tickets to get through though.

Does anyone have experience with beta gems? I'd like to put some up
before release and figured out how to serve them, but I'm curious how
versioning is handled. Should 29.1 be beta, 29.2 the release? Or would
gems consider 29.1b to be less than 29.1? Or are they handled some
other way?

> Also note that I have renamed kernel/meta to kernel/instance --for
> those whol might not know about this method, it alows you to get
> information about an object even if the methods normally used have
> been overridden. For example:
>
>   class X
>     def class ; "OBSCURED" ; end
>   end
>
>   x = X.new
>   x.class                #=> "OBSCURED"
>   x.instance.class  #=> X
>
> Let me know if anyont thinks there's a problem with this new name or
> has a suggestion for a better one. Thanks.
>

First thing that came to my mind was shadowed.

x.shadowed.class # => X

It makes me think of variable shadowing in [insert any lexically
scoped language here].
Although I suppose overriden, somewhat like C# would be pretty
understandable too.

x.overriden.class # => X

Bryan

> Lastly, I'll be releasing a nw version of Reap tonight too, since I
> got testing working better and it now plays nice with nitro. (Still
> need to finish the deb and release tasks though.)
>
> T.
>
> P.S. Hey Zimba. How are we doing on getting a Facets repo? Don't want
> to get too far a field here before that happens. Thnx.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From lionel.orry at gmail.com  Thu Mar 30 01:53:02 2006
From: lionel.orry at gmail.com (Lionel Orry)
Date: Thu, 30 Mar 2006 08:53:02 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <20060329220536.GA12019@mlabdial.hit.bme.hu>
References: <op.s66ynonqg3glhz@jo.local>
	<20060329220536.GA12019@mlabdial.hit.bme.hu>
Message-ID: <442B804E.2090901@gmail.com>



Gergely Kontra wrote:
>> Hey all,
>>
>> I've been working on a requested feature, a printing view for tutorials:
>>
>> http://oxyliquit.de/cache/pdf_cache/A_small_project_with_Og.pdf
>>
>> this has been automatically created, tell me what you think and weither
>> the quality is sufficient.
> 
> Hi!
> 
> I assume you need some critism. Here you are :)
> Anyway, I never produced pdf directly, only through LaTeX

In fact, after having a look at the rendered pdf, I think using LaTeX to generate it would be nice. You would have a better general result 
(mostly about readability and correct document layout: pictures referencing, correct headings placement, etc.).

I've seen that Instiki has a LaTeX exporter that generates the LaTeX equivalent for a selected wiki page. On the top of that, Instiki uses 
the Textile syntax. The result is far from perfect as far as I know, but it could be a base to work on.

If the Instiki implementation cannot be applied to Oxyliquit, I still think that using ruby templates to render LaTeX from Textile or html 
would not be that hard... Well these are my 2 cents.

Lionel

> 
> First, headings appear at the end of the page. This is bad.
> 
> The image is not labeled, not referenced, so if you have more images, it 
> will not be obvious, which image you are talking about.
> 
> Pdf generation is a great thing, but for a tutorial, I don't think you 
> should work so much. A correct css stylesheet can do.
> 
> Anyway, if you can solve automagically the mentioned things, I'm also 
> interested how can one do such thing :)
> 
> Gerg?
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From manveru at weez.co.jp  Thu Mar 30 01:59:29 2006
From: manveru at weez.co.jp (Michael Fellinger)
Date: Thu, 30 Mar 2006 15:59:29 +0900
Subject: [Nitro] Setting default-values for properties
Message-ID: <200603301559.29451.manveru@weez.co.jp>

Hey guys,

Like the title says, i want to request a feature that i've been missing way 
too long :)
The problem is, after chewing through the Og-code i still have no idea where 
or how exactly to add it so it makes sense and is applicable to most stores. 
Once that place is found it should be quite trivial to implement, so i beg 
your help for that.

class Bar
  property :foo, String, :default => "bar"
  property :date, Time, :default => Time.now
  property :blah, Float, ;default => 0.0
  property :hash, Hash, :default => {:x => 4, :y => 6}
end

I think that should say everything, i want the possibility to preset values.
Now, there are some approaches that we have discussed about in IRC, but I 
really need more opinions (and maybe even a working patch ;)

Me and TRANS discussed a while about how to solve it in the best possible way 
and here is what we came up with:
* giving static default-values, they could be saved in the db-specific 
default-handling (i know that most SQL-stores have that capability)
* one should be able to give a block to :default, maybe we could use lazy.rb 
for that (will be integrated in facets 1.2.3)
* The values have to be defined pre-initialize (this could be tricky, since 
entity.new doesn't insert anything into the DB, so the default-values don't 
kick in and might have to be simulated (which gives us the question if 
implementing it in the store native is such a good idea after all)...
* maybe we could use some form of Aspect?

And last but not least - is there something like that maybe already 
implemented? i haven't seen anything in the docs/code, so i'm a bit clueless.

~~~~manveru


From kashia at vfemail.net  Thu Mar 30 05:24:51 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 30 Mar 2006 12:24:51 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <bd93db0d0603291244t185b900cx5422eb1150a7be98@mail.gmail.com>
References: <op.s66ynonqg3glhz@jo.local>
	<bd93db0d0603291244t185b900cx5422eb1150a7be98@mail.gmail.com>
Message-ID: <op.s671vpy2g3glhz@jo.local>

Hi,

> Wow, very cool! How did you do it? I've been wanting to generate pdfs
> for some forms I have.

I'm using htmldoc to make the pdf atm.


> The quality looks okay to me. The code is rather light in comparison
> to the text though.

I'm not sure if I should invest the time digging through htmldoc commands
for that... I'm not really fond of the overall quality :)

> Oh, I answered your question about fields that Og doesn't touch with
> the lame answer you already knew. Just rate it as a 0 and be done with
> it! ;)

I rated it a 4, because it has to be marked as "finished", there is no
other way :)

Kashia


-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kashia at vfemail.net  Thu Mar 30 05:28:56 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 30 Mar 2006 12:28:56 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <20060329220536.GA12019@mlabdial.hit.bme.hu>
References: <op.s66ynonqg3glhz@jo.local>
	<20060329220536.GA12019@mlabdial.hit.bme.hu>
Message-ID: <op.s6712ichg3glhz@jo.local>

Hi,

> I assume you need some critism. Here you are :)

Yes yes, more more ;D

> First, headings appear at the end of the page. This is bad.
>
> The image is not labeled, not referenced, so if you have more images, it
> will not be obvious, which image you are talking about.

Both are faults of htmldoc, can't do anything about them.

> Pdf generation is a great thing, but for a tutorial, I don't think you
> should work so much. A correct css stylesheet can do.

So, you think just html suffices?

> Anyway, if you can solve automagically the mentioned things, I'm also
> interested how can one do such thing :)

I can, but I'm not sure if it's worth the work :P

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kashia at vfemail.net  Thu Mar 30 05:32:04 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 30 Mar 2006 12:32:04 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <20060329222240.GC12019@mlabdial.hit.bme.hu>
References: <op.s66ynonqg3glhz@jo.local>
	<20060329222240.GC12019@mlabdial.hit.bme.hu>
Message-ID: <op.s6717qu9g3glhz@jo.local>

Hi,

> New bug :)
> When you login, you are 'kicked' soon, so it forgets me in minutes

Yep... current user system is built on sessions, which are invalidated
so fast.....

I'm in the process of figuring out the best way to do authentication,
this is not easy...
nitro-auth / HTTP authentication..

> Cannot edit my own question
... huh O.O

> Cannot rerate an answer (maybe it's a feature)
You can rerate .... O_o

Both should work, but will investigate.

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From kashia at vfemail.net  Thu Mar 30 05:33:00 2006
From: kashia at vfemail.net (Kashia Buch)
Date: Thu, 30 Mar 2006 12:33:00 +0200
Subject: [Nitro] Oxyliquit tutorial pdfs
In-Reply-To: <4b6f054f0603291556h6bfab4e1s8692400fa72377c4@mail.gmail.com>
References: <op.s66ynonqg3glhz@jo.local>
	<4b6f054f0603291556h6bfab4e1s8692400fa72377c4@mail.gmail.com>
Message-ID: <op.s6719ay1g3glhz@jo.local>

On Thu, 30 Mar 2006 01:56:37 +0200, TRANS <transfire at gmail.com> wrote:

> Very nice. Your 1/20th of the way to a book! ;-)

Yes, all you lazy asses get to work and write more tutorials >:D

(just kidding ;) )

Kashia

-- 
Feel the love
http://pinkjuice.com/pics/ruby.png


From itsme213 at hotmail.com  Thu Mar 30 10:05:25 2006
From: itsme213 at hotmail.com (itsme213)
Date: Thu, 30 Mar 2006 09:05:25 -0600
Subject: [Nitro] Setting default-values for properties
References: <200603301559.29451.manveru@weez.co.jp>
Message-ID: <e0gs3l$vr4$1@sea.gmane.org>


"Michael Fellinger" <manveru at weez.co.jp> wrote in message

> And last but not least - is there something like that maybe already
> implemented? i haven't seen anything in the docs/code, so i'm a bit 
> clueless.

Ara's traits library does this in a fully Ruby-esque way (e.g. works across 
modules, includes, subclasses).






From transfire at gmail.com  Thu Mar 30 11:10:07 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 30 Mar 2006 11:10:07 -0500
Subject: [Nitro] Setting default-values for properties
In-Reply-To: <e0gs3l$vr4$1@sea.gmane.org>
References: <200603301559.29451.manveru@weez.co.jp>
	<e0gs3l$vr4$1@sea.gmane.org>
Message-ID: <4b6f054f0603300810o3f712a99k589c42440046e375@mail.gmail.com>

On 3/30/06, itsme213 <itsme213 at hotmail.com> wrote:
>
> "Michael Fellinger" <manveru at weez.co.jp> wrote in message
>
> > And last but not least - is there something like that maybe already
> > implemented? i haven't seen anything in the docs/code, so i'm a bit
> > clueless.
>
> Ara's traits library does this in a fully Ruby-esque way (e.g. works across
> modules, includes, subclasses).

hmm... but it does everything in it's own way. It's not integrated
with the annotation system --and that's really where this needs to be.
Also realize when you define a trait it creates a reader, an accessor
and a query? method --while maybe that is a good thing, it's not
general practice of ruby really.

I think the optimal solution is to add a callback system into
anotation.rb. Something like:

  annotation_action( :default ) do |name, value|
      preinitialize {
         instance_variable_set( "@#{name}", value )
      }
  end

So whenever one did this (note this applies to property too):

  ann :foo, :defualt => 10

Then the :default => 10 part would trigger the above callback. Btw,
preinitialize.rb is from facets --it works but is kind of hackish,
IMHO. I wish Ruby had a *real* aop system :-(

T.



From dylanb at digitalvalence.com  Thu Mar 30 13:05:44 2006
From: dylanb at digitalvalence.com (Dylan Bruzenak)
Date: Thu, 30 Mar 2006 12:05:44 -0600
Subject: [Nitro] PATCH: bs-kirby-fixes
In-Reply-To: <bd93db0d0603280004w50a2e5ffuaa21ecd5c54bc411@mail.gmail.com>
References: <bd93db0d0603280004w50a2e5ffuaa21ecd5c54bc411@mail.gmail.com>
Message-ID: <442C1DF8.7070505@digitalvalence.com>

Wahahaa! I knew if I waited long enough someone would fix my camelcasing.

Actually, just forgot to change that. When I get 5 mins(the end of 2006 
?) I'll pull a new version and take a look at the remaining errors. If I 
remember correctly they were mostly things that were very unkerbylike... 
sql statements and the like.

Dylan

Bryan Soto wrote:
> Hey list,
>
> I had to rework Dylan's Kirby fixes patch. It didn't apply cleanly
> since things had changed in the devlab repo since his submission.
> Obviously not his fault. :)
>
> It's almost all his code so he get's the credit. I just changed some
> of his camelCase variable names. :)
>
> Kirby now goes from not running to:
>
> 43 tests, 130 assertions, 4 failures, 7 errors
>
> Could someone give this a lookover? It's late for me.
>
> Thanks,
>
> Bryan
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general



From bryan.a.soto at gmail.com  Thu Mar 30 16:34:16 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 13:34:16 -0800
Subject: [Nitro] [PATCH] Configuration + Og startup speedup
In-Reply-To: <afd80b540603290749y68401afbr60f030bfaa10ed94@mail.gmail.com>
References: <afd80b540603271106k635ebab6r44dd6e5b3b29378e@mail.gmail.com>
	<bd93db0d0603271214i229f66a8w11b1d00ca07162e1@mail.gmail.com>
	<afd80b540603290749y68401afbr60f030bfaa10ed94@mail.gmail.com>
Message-ID: <bd93db0d0603301334k3a8d4b97k50b2c5b723f023be@mail.gmail.com>

Just as an fyi, this make some tests fail for some reason. I'll see if
I can fix them up before applying.

On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Yeah,
>
> here is the patch, please submit it ;-)
>
> -g.
>
> On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> > Hey g,
> >
> > You forgot the patch. :)
> >
> > Thanks
> >
> > On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > > Bryan, can you add these to the stable repo?
> > >
> > > thanks,
> > > -g.
> > >
> > >
> > > Mon Mar 27 21:59:51 EEST 2006  George Moschovitis <gm at navel.gr>
> > >   * Startup optimization, don't search in ObjectSpace for managed classes.
> > >
> > > Mon Mar 27 21:43:44 EEST 2006  George Moschovitis <gm at navel.gr>
> > >   * Further improved configuration system, even added some ruby magic,
> > > check out the rdocs and the updated test case.
> > >
> > >
> > >
> > > --
> > > http://www.gmosx.com
> > > http://www.navel.gr
> > > http://www.nitrohq.com
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> >
> >
> > --
> > "Never tell people how to do things. Tell them what to do and they
> > will surprise you with their ingenuity." ?General George S. Patton
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>
>
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 16:48:56 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 13:48:56 -0800
Subject: [Nitro] properties
In-Reply-To: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
Message-ID: <bd93db0d0603301348y3973fe8jd88f6a027c494e2f@mail.gmail.com>

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> Bryan,
>
> I saw you removed the og_properties patch. Perhaps we could include
> this but please do NOT name the variable og_properties. Properties are
> a general feature, NOT Og specific. Perhaps __properties would do? (Of
> course we should keep the properties alias).
>
> Perhaps there is a better alternative to avoid the GTK2 nameclash
> though. Any ideas?

I have no code written, but I have a vague idea.

Perhaps the problem is that properties are generic, but not reusable.
As an example, for an Og enchanted class, perhaps field and fields
would be a better name than property and properties.

My basic idea, is that you'd have a symbol parameter that would be
used to define the method names that are later used for a class.
Basically, these names would be user definable. The only problem I've
thought of is you'd need an external setting that would store the name
of the holding variable (currently properties so people understand
what I'm saying) so that all the internal code will now what method to
call.

Anyway, a vague notion I haven't had time to actually try playing with.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Thu Mar 30 17:21:21 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 30 Mar 2006 17:21:21 -0500
Subject: [Nitro] properties
In-Reply-To: <bd93db0d0603301348y3973fe8jd88f6a027c494e2f@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<bd93db0d0603301348y3973fe8jd88f6a027c494e2f@mail.gmail.com>
Message-ID: <4b6f054f0603301421n40aba2dbs1b72ea9f2ddba932@mail.gmail.com>

On 3/30/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:

> My basic idea, is that you'd have a symbol parameter that would be
> used to define the method names that are later used for a class.
> Basically, these names would be user definable. The only problem I've
> thought of is you'd need an external setting that would store the name
> of the holding variable (currently properties so people understand
> what I'm saying) so that all the internal code will now what method to
> call.

Please explain more.  I'm not sure what you mean.

BTW --could someone specifically define the actual problem? Is it just
the GTK clash?

Thanks,
T.



From bryan.a.soto at gmail.com  Thu Mar 30 17:32:05 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 14:32:05 -0800
Subject: [Nitro] how classes are enchanted
In-Reply-To: <4b6f054f0603282120s7769030g98c3684144d21d6a@mail.gmail.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
	<4b6f054f0603282120s7769030g98c3684144d21d6a@mail.gmail.com>
Message-ID: <bd93db0d0603301432s7c8627b3m618d0ce02c30bc77@mail.gmail.com>

Trans,

You might check out George's recent patch. He's gone away from the
ObjectSpace iteration approach to registering classes as enchanted in
a hash.

Oh, and mark me as in favor of explicit enchanting. :)

On 3/28/06, TRANS <transfire at gmail.com> wrote:
> Some more consideration:
>
>   def Og.enchant(*models)
>
>     model_files, models = models.partition { |m| String === m }
>
>     # load model files, enchanting classes as loaded
>
>     def ::Object.inherited( klass )
>       klass.class_eval { include Og::EntityMixin }
>     end
>     model_files.each{|m| require(m)}
>     class << ::Object
>       remove_method :inherited
>     end
>
>     # enchant model classes
>
>     models.each { |mod|
>       case mod
>       when Class
>         klass.class_eval { include Og::EntityMixin }
>       when Module
>          # ... enchant all classes in module ...
>       else
>         # error or singleon enchant?
>       end
>     }
>
>   end
>
> This gives a number of ways to enchant a class, besides the manual way.
>
>   Og.enchant( FooClass )      # directly on an external class
>   Og.enchant( FooModule )   # classes in FooModule
>   Og.enchant( "models.rb" )  # classes in a separate file
>
> T.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 17:58:46 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 14:58:46 -0800
Subject: [Nitro] properties
In-Reply-To: <4b6f054f0603301421n40aba2dbs1b72ea9f2ddba932@mail.gmail.com>
References: <afd80b540603271111t1f33c524oc2cbb1e01be43c1f@mail.gmail.com>
	<bd93db0d0603301348y3973fe8jd88f6a027c494e2f@mail.gmail.com>
	<4b6f054f0603301421n40aba2dbs1b72ea9f2ddba932@mail.gmail.com>
Message-ID: <bd93db0d0603301458n604fa903pe794a2cf8d664815@mail.gmail.com>

On 3/30/06, TRANS <transfire at gmail.com> wrote:
> On 3/30/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
>
> > My basic idea, is that you'd have a symbol parameter that would be
> > used to define the method names that are later used for a class.
> > Basically, these names would be user definable. The only problem I've
> > thought of is you'd need an external setting that would store the name
> > of the holding variable (currently properties so people understand
> > what I'm saying) so that all the internal code will now what method to
> > call.
>
> Please explain more.  I'm not sure what you mean.
>

Where to start? It's not even fleshed out... :)

I guess rather than using property and properties, you could supply a
name that would make more sense for your particular application.
glue/property.rb would use this name to generate all the needed
methods. Like in Og, rather than using property and properties, we
could supply a parameter that would generate all the method names as
field and fields (not a suggestion by the way, just an example :).
Basically, rather than hardcoding property and properties in the file,
they would be replaced with a variable name in the eval statements.

Hopefully, code snippets will make a bit more sense. As an example,
glue/property.rb:241,

inheritor(:properties, Dictionary.new, :merge) unless @properties

 would become

inheritor(#{storage_name}, Dictionary.new, :merge) unless @#{storage_name}

if that's even valid syntax.

But since it can be changed, code can't really be written to use it
directly. So, you'd have to have it stored some place. Where we now
use,

klass.properties

it'd have to be retrieved externally, like maybe

m= Glue::Properties.name_for(klass)
klass_props = klass.send(m)

or maybe stored via a setting,

Og.properties_name= :fields

I don't know if it's a good idea, but it's a different idea than
renaming the variable, so I thought I may as well mention it. :)

> BTW --could someone specifically define the actual problem? Is it just
> the GTK clash?
>

glue/property.rb uses a method named properties for retrieving
properties from a class. GTK uses a method named properties to access
the GLib properties of a GObject. Hence a collision. :)


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From m.fellinger at gmail.com  Thu Mar 30 18:41:32 2006
From: m.fellinger at gmail.com (Michael Fellinger)
Date: Fri, 31 Mar 2006 08:41:32 +0900
Subject: [Nitro] how classes are enchanted
In-Reply-To: <bd93db0d0603301432s7c8627b3m618d0ce02c30bc77@mail.gmail.com>
References: <4b6f054f0603281903y6dbdaeb4ta55a7d029f606c57@mail.gmail.com>
	<4b6f054f0603282120s7769030g98c3684144d21d6a@mail.gmail.com>
	<bd93db0d0603301432s7c8627b3m618d0ce02c30bc77@mail.gmail.com>
Message-ID: <9c00d3e00603301541p5ddca670t3841b3117d1849f9@mail.gmail.com>

Another vote for explicity...
On 3/31/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:> Trans,>> You might check out George's recent patch. He's gone away from the> ObjectSpace iteration approach to registering classes as enchanted in> a hash.>> Oh, and mark me as in favor of explicit enchanting. :)>> On 3/28/06, TRANS <transfire at gmail.com> wrote:> > Some more consideration:> >> >   def Og.enchant(*models)> >> >     model_files, models = models.partition { |m| String === m }> >> >     # load model files, enchanting classes as loaded> >> >     def ::Object.inherited( klass )> >       klass.class_eval { include Og::EntityMixin }> >     end> >     model_files.each{|m| require(m)}> >     class << ::Object> >       remove_method :inherited> >     end> >> >     # enchant model classes> >> >     models.each { |mod|> >       case mod> >       when Class> >         klass.class_eval { include Og::EntityMixin }> >       when Module> >          # ... enchant all classes in module ...> >       else> >         # error or singleon enchant?> >       end> >     }> >> >   end> >> > This gives a number of ways to enchant a class, besides the manual way.> >> >   Og.enchant( FooClass )      # directly on an external class> >   Og.enchant( FooModule )   # classes in FooModule> >   Og.enchant( "models.rb" )  # classes in a separate file> >> > T.> >> > _______________________________________________> > Nitro-general mailing list> > Nitro-general at rubyforge.org> > http://rubyforge.org/mailman/listinfo/nitro-general> >>>> --> "Never tell people how to do things. Tell them what to do and they> will surprise you with their ingenuity." ?General George S. Patton>> _______________________________________________> Nitro-general mailing list> Nitro-general at rubyforge.org> http://rubyforge.org/mailman/listinfo/nitro-general>


From bryan.a.soto at gmail.com  Thu Mar 30 19:21:41 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 16:21:41 -0800
Subject: [Nitro] nested templates calls and shared data
In-Reply-To: <afd80b540603270947i3866d13aobab1120c99b83a55@mail.gmail.com>
References: <6a7d49ca0603150715o7a13c22br@mail.gmail.com>
	<bd93db0d0603221622y66313a58mf363b22c6bb4e1f5@mail.gmail.com>
	<6a7d49ca0603270251i26b502f9n@mail.gmail.com>
	<afd80b540603270947i3866d13aobab1120c99b83a55@mail.gmail.com>
Message-ID: <bd93db0d0603301621h1579935ex3d6c95a97a0168f8@mail.gmail.com>

I've opened a ticket for this.

http://devlab.oree.ch/trac/nitrohq/ticket/24#preview

On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> I think we could work out a better implementation, that keeps all
> javascript in one block.
>
> -g.
>
> On 3/27/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > thanks, i was stuck on the idea that script must go just before the
> > </body> tag, but your solution looks good!
> >
> > i implemented it as a patch.
> >
> >
> >
> >
> > 2006/3/23, Bryan Soto <bryan.a.soto at gmail.com>:
> > > On 3/15/06, guillaume pierronnet <guillaume.pierronnet at gmail.com> wrote:
> > > > i spent all my week-end on trying to implement a sort of 'global
> > > > compiling pipeline', with no success.
> > > >
> > > > does anybody have a idea on how to fix it ?
> > >
> > > Hi Guill,
> > >
> > > Check out nitro/lib/nitro/compiler/script.rb:90
> > >
> > > Currently, the script compiler looks for a </body> tag and writes the
> > > script out at the end of it. As a "fugly hack", you can add
> > >
> > > <!-- </body> -->
> > >
> > > to the end of your album_list.xhtml file.
> > >
> > > Perhaps a real fix would be to check if text =~ /<\/body>/ and,
> > > failing that, just appending the the script at the end of text? What
> > > do you think?
> > >
> > > Bryan
> > >
> > > --
> > > "Never tell people how to do things. Tell them what to do and they
> > > will surprise you with their ingenuity." ?General George S. Patton
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
> >
> >
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 19:29:59 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 16:29:59 -0800
Subject: [Nitro] Dynamic Settings
In-Reply-To: <44281A0E.9020201@neurogami.com>
References: <44260863.5080501@neurogami.com>
	<509369bc93828575beeb9fea6fab30e7@oggu.de>
	<4428013C.8040309@neurogami.com>
	<afd80b540603270758h446af406rcea21c7ce0fe33ff@mail.gmail.com>
	<44281A0E.9020201@neurogami.com>
Message-ID: <bd93db0d0603301629p4bb5583foc6046207cf3df752@mail.gmail.com>

On 3/27/06, James Britt <james_b at neurogami.com> wrote:
> George Moschovitis wrote:
> >>It's similar to what I've put together for myself.  My version checks
> >>for ENV[ 'NITRO_ENV' ] and loads different settings  based on the value.
> >
> >
> > you can have different settings for debug, stage and live/production. Run with
> >
> > ruby run.rb --debug/--stage/--production etc.
> >
> > have a look at runner.rb for details.
>
> But, more important, what I don't see is how I tell the runner that I
> want --foo_mode or --development2 or any arbitrary environment setting
> and have it load a set of configuration values specific to that environment.
>
> The method load_external_configuration will load a config file, but the
> setup_options method seems to be restricted to a particular set of
> run-mode options.
>
> So I added an option to runner.rb
>
>    opts.on('--config CONF', 'Load  config file') do |conf|
>      load_external_configuration conf
>    end
>
>
> I noticed that load_external_configuration just quietly falls through if
> there is no matching config file.  Is this useful?  I see, for example,
> that, setup_debug does some configuring, then calls out to
> load_external_configuration; I gather in that case a missing debug.*
> config file is OK.  But, for my purposes, I'd want an exception if I
> specifically asked for a config file and it was not found.
>
> Right now I've added this to 'load_external_configuration'
>
>    unless self.class.mode.to_s =~ /^(debug|stage|live)$/
>      raise "No matching config file!" unless  have_config?( mode )
>    end
>
> and this helper method:
>
> def have_config? mode
>   File.exist?( "conf/#{mode}.rb" ) || File.exist?( "conf/#{mode}.yml" )
> end
>

Do you think that is good enough to be added to the repo? If so, I'd
be happy to write it up in darcs patch form. Although I could have
swore Rob mentioned he stored configs in Yaml form. I'll have to see
if I can dig that up in my inbox.

> Not happy with the hard-coded set of special-case modes, though.

I suppose a plugin system with registration or const_get magic would
be an option. Or did you have another idea perhaps?


Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 19:31:50 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 16:31:50 -0800
Subject: [Nitro] PATCH: bs-kirby-fixes
In-Reply-To: <442C1DF8.7070505@digitalvalence.com>
References: <bd93db0d0603280004w50a2e5ffuaa21ecd5c54bc411@mail.gmail.com>
	<442C1DF8.7070505@digitalvalence.com>
Message-ID: <bd93db0d0603301631s5fd44b33m4790fcacd37e1f4c@mail.gmail.com>

On 3/30/06, Dylan Bruzenak <dylanb at digitalvalence.com> wrote:
> Wahahaa! I knew if I waited long enough someone would fix my camelcasing.
>

Don't get used to it. Your work productivity will go _way_ down. ;)

> Actually, just forgot to change that. When I get 5 mins(the end of 2006
> ?) I'll pull a new version and take a look at the remaining errors. If I
> remember correctly they were mostly things that were very unkerbylike...
> sql statements and the like.
>

That would be great. :)

> Dylan
>

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 19:34:36 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 16:34:36 -0800
Subject: [Nitro] Hiding a column from Og?
In-Reply-To: <op.s66g2ocag3glhz@jo.local>
References: <op.s66g2ocag3glhz@jo.local>
Message-ID: <bd93db0d0603301634j4d8cee7bx98b8c217a81b625@mail.gmail.com>

On 3/29/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hi,
>
> how do I create a column in a database which is not noticed by Og?
>
> I want to add an additional column to a table, and want to avoid Og messing with it or even noticing it, is that possible?
>
> http://oxyliquit.de/question/23

Kashia, could you explain a liitle on why you need this? Perhaps it
would help people get a feel for whether something like this should be
added.

Or perhaps someone already has an idea on this? Or even an opinion on
whether this is a bug or a feature?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 19:44:48 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 16:44:48 -0800
Subject: [Nitro] Setting default-values for properties
In-Reply-To: <4b6f054f0603300810o3f712a99k589c42440046e375@mail.gmail.com>
References: <200603301559.29451.manveru@weez.co.jp>
	<e0gs3l$vr4$1@sea.gmane.org>
	<4b6f054f0603300810o3f712a99k589c42440046e375@mail.gmail.com>
Message-ID: <bd93db0d0603301644x5cf76972v38247d9be17b9542@mail.gmail.com>

On 3/30/06, TRANS <transfire at gmail.com> wrote:
> On 3/30/06, itsme213 <itsme213 at hotmail.com> wrote:
> >
> > "Michael Fellinger" <manveru at weez.co.jp> wrote in message
> >
> > > And last but not least - is there something like that maybe already
> > > implemented? i haven't seen anything in the docs/code, so i'm a bit
> > > clueless.
> >
> > Ara's traits library does this in a fully Ruby-esque way (e.g. works across
> > modules, includes, subclasses).
>
> hmm... but it does everything in it's own way. It's not integrated
> with the annotation system --and that's really where this needs to be.
> Also realize when you define a trait it creates a reader, an accessor
> and a query? method --while maybe that is a good thing, it's not
> general practice of ruby really.
>
> I think the optimal solution is to add a callback system into
> anotation.rb. Something like:
>
>   annotation_action( :default ) do |name, value|
>       preinitialize {
>          instance_variable_set( "@#{name}", value )
>       }
>   end
>
> So whenever one did this (note this applies to property too):
>
>   ann :foo, :defualt => 10
>
> Then the :default => 10 part would trigger the above callback. Btw,
> preinitialize.rb is from facets --it works but is kind of hackish,
> IMHO. I wish Ruby had a *real* aop system :-(
>

So your thinking this should be a part of Facets then rather than an
Og specific feature?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 19:49:54 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 16:49:54 -0800
Subject: [Nitro] graph handling (selfjoin)
In-Reply-To: <op.s6d3v3g1g3glhz@jo.local>
References: <20060311024005.GB26563@mlabdial.hit.bme.hu>
	<op.s58zvznbg3glhz@jo.local>
	<20060311135512.GA28699@mlabdial.hit.bme.hu>
	<op.s588d1x5g3glhz@jo.local>
	<20060312123125.GA1206@mlabdial.hit.bme.hu>
	<op.s6a1foqjg3glhz@jo.local>
	<afd80b540603131220j6e1b2b0cna96ba8b40dcc700a@mail.gmail.com>
	<op.s6d3v3g1g3glhz@jo.local>
Message-ID: <bd93db0d0603301649k7de03946ib19e216fb77cd7ea@mail.gmail.com>

On 3/13/06, Kashia Buch <kashia at vfemail.net> wrote:
> On Mon, 13 Mar 2006 21:20:22 +0100, George Moschovitis <george.moschovitis at gmail.com> wrote:
>
> >> Yes, as you can see at my other posting about self joins, this doesn't seem
> >> to work...
> >> I hope George or someone else can help here.
> >> Please, I need this!
> >
> > I will not be able to look into this befor the weekend :(
>
> Don't worry, I found an evil workaround, which does it for now.
>
> Kash

This was one of the threads I had marked as unread, so I'm wondering
if this was resolved or is still an issue?

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 19:56:06 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 16:56:06 -0800
Subject: [Nitro] Self-join :through =>
In-Reply-To: <op.s6clqhlag3glhz@jo.local>
References: <op.s59k6qy0g3glhz@jo.local>
	<dea8525a0603122213q28acf05u8bc66f4553d6d0f3@mail.gmail.com>
	<op.s6clqhlag3glhz@jo.local>
Message-ID: <bd93db0d0603301656t3857ccaaidbd04c26f4ece726@mail.gmail.com>

On 3/13/06, Kashia Buch <kashia at vfemail.net> wrote:
> Hey,
>
> thanks for trying to help, but:
>
> > Easy...
>
> not quite so much :)
>
> > class Foo
> >   many_to_many :foorelations, Foorelation
> > end
> >
> > class Foorelation
> >   many_to_many :foos, Foo
> > end
> >
> > Is that what you want to do?
>
> Sadly no.
>
> I want to make a self join through a second class.
>
> this is what I came finally up with: (it works)
> I think Robs patches could've helped me enormously.
>

I think he ran into some problems with trying to handle them
automatically and had to give up that approach. He went with a :table
option which I've applied.

Patch is here or in your repo if you're up to date. Maybe this well help?

http://rubyforge.org/pipermail/nitro-general/2006-March/003326.html

If not, could you open a ticket on devlab? Is your example earlier in
this thread suitable for converting to a test case?

Thanks,

Bryan

> class Foo
>    # ...
>    many_to_many :relations, Manga, :through => MangaRelation, :control => :none
> end
>
> class MangaRelation
>    property :type, String
>
>    property :foo, Fixnum, :field => 'foo_oid', :foreign_key => 'oid', :relation => true, :control => :none
>    property :foo2, Fixnum, :field => 'foo_oid2', :foreign_key => 'oid', :relation => true, :control => :none
> end
>
> There's the problem that I can't use controls for Foo in that case then...
> minor detail for now, at least it works :)
>
> for everyone who is curious how I use this:
> foo.relations(:select => 'ogfoo.*, type')
>
> And no, every single thing here is not nice, but atm I don't know any other way.
>
> Kashia
>
> --
> Feel the love
> http://pinkjuice.com/pics/ruby.png
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 20:00:29 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 17:00:29 -0800
Subject: [Nitro] [PATCH] Entity copying methods
In-Reply-To: <dda3b98e0603090547x14d3d2cfp3de4f55d45afd16c@mail.gmail.com>
References: <1138372566.2932.34.camel@robs-p4>
	<dda3b98e0601270758w1bfd2792v1935be08733031c8@mail.gmail.com>
	<1138380314.2932.40.camel@robs-p4>
	<afd80b540602010116k7032afc5j7095938b298489af@mail.gmail.com>
	<1138789134.4872.22.camel@robs-p4>
	<dda3b98e0603090547x14d3d2cfp3de4f55d45afd16c@mail.gmail.com>
Message-ID: <bd93db0d0603301700j13e89ee8j977838624495cb5c@mail.gmail.com>

On 3/9/06, vseguip at gmail.com <vseguip at gmail.com> wrote:
>  Hi Rob,
>
>    I was was just wondering if you finally got to implement the copy
> of "join through" data. I'm interested since my join through tables
> actually relate to other tables and it's actually a PITA the copy
> them.
>
> Thanks for you time,
>   V. Segu?

Would it be possible for you to open a ticket on devlab with a
description of what you're looking for? I wasn't able to make out
exactly what that was, but I want to make sure this didn't get
forgotten.

Thanks.

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 20:03:50 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 17:03:50 -0800
Subject: [Nitro] [Bug] Memory based session cache does not release memory
In-Reply-To: <716700c90603070434me5306f9ud807d4d4dc82c4ff@mail.gmail.com>
References: <716700c90603050552u673ab3abqa61f7e6da537627f@mail.gmail.com>
	<bd93db0d0603062245w51f55061j8747c1012ca23b85@mail.gmail.com>
	<716700c90603070434me5306f9ud807d4d4dc82c4ff@mail.gmail.com>
Message-ID: <bd93db0d0603301703s169a369ex610ea8d2fea1bf19@mail.gmail.com>

I've opened a ticket for this on devlab.

On 3/7/06, Fang Sun <nusgnaf at gmail.com> wrote:
> Hi, Bryan.
> I have tested og, file and memory backed session using proto app. The
> leak exists only with memory backend. In fact with og or file session
> backed, when I run Nitro::Session.cache.gc!, the session expired is
> deleted from database or filesystem, and the nitro program consumed
> memory stayed around 13+ MB.
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 20:06:12 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 17:06:12 -0800
Subject: [Nitro] Admin improvements
In-Reply-To: <200603061558.10912.zimba.tm@gmail.com>
References: <200603061558.10912.zimba.tm@gmail.com>
Message-ID: <bd93db0d0603301706r42375a81xf41a5a81f27f39d2@mail.gmail.com>

Hi everyone,

Any ideas here we should add to devlab as enhancements or that anyone
would like to work on?

On 3/6/06, zimba.tm <zimba.tm at gmail.com> wrote:
> Hi list,
>
> the next thing I would to work on, is to improve the admin. Before I go on,
> here are the ideas I would like to implement :
>
> = Make settings changeable and storable
>
>  * Use the caching system or Og to load/save settings
>  * Make the settings modifiable in the web interface
>  * Group the settings in different pages, according to their namespace to
> avoid a huge page
>
> = Use rubygems for components
>
> I don't know if it will be straignforward.
>
>  * Use a nitro-specific settings for rubygems (store location, download
> location, ...)
>  * Make rubygems (un)installable/uploadable from the web interface
>  * Add the ability to instanciate components on specific web paths
>  * Upgrade Og so that it supports table installing/upgrading on the fly
> (without relaunching nitro)
>
> = Standalone nitro
>
> Make a nitro executable that will launch nitro with the admin and custom
> options
>
> = Editable
>
>  * Make sources and templates editable from the web interface
>
> --
> Cheers,
>   zimba.tm
>
> weblog : http://zimba.oree.ch
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 20:07:16 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 17:07:16 -0800
Subject: [Nitro] About the new Control system
In-Reply-To: <afd80b540603050037k411841c9m590a7272279d3359@mail.gmail.com>
References: <200603021258.39943.zimba.tm@gmail.com>
	<200603031850.10543.zimba.tm@gmail.com>
	<afd80b540603040225n7958efedi455f0e38de1a67ca@mail.gmail.com>
	<200603041527.09635.zimba.tm@gmail.com>
	<4E9828BE-2D68-4592-809D-8035039CEB77@ntecs.de>
	<afd80b540603050037k411841c9m590a7272279d3359@mail.gmail.com>
Message-ID: <bd93db0d0603301707o76bae12dr9539514f78018a8a@mail.gmail.com>

Hi George,

Did you ever uncover anything on this? :)

On 3/5/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> >
> > Maybe
> >
> > "#{var :target}"
> >
>
> *Interesting* idea! let me investigate this...
>
> thanks,
> -g.
>
>
> --
> http://www.gmosx.com
> http://www.navel.gr
> http://www.nitrohq.com
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 20:11:00 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 17:11:00 -0800
Subject: [Nitro] Discussion about setting relationships with unsaved
	objects
In-Reply-To: <1140613941.2916.0.camel@robs-p4>
References: <1140603804.28264.43.camel@robs-p4>
	<afd80b540602220414s4080dbe0ra3f3798618a1975e@mail.gmail.com>
	<1140613941.2916.0.camel@robs-p4>
Message-ID: <bd93db0d0603301711l46fa99acm8387519bed6c5705@mail.gmail.com>

Did this one ever get resolved? If not, we should open a ticket.

Thanks.

Bryan

On 2/22/06, Rob Pitt <rob at motionpath.com> wrote:
> That's a better idea - OK I'll just patch my project to throw Exceptions
> until this is done.
>
> On Wed, 2006-02-22 at 14:14 +0200, George Moschovitis wrote:
> > Please don't make this change yet.I would like to add a 'build' mode
> > in collections.
> > Ie, You create an object, assign the relations and save in one go.
> > Will add this during next week, so be patient.
> >
> > -g.
> >
> >
> > On 2/22/06, Rob Pitt <rob at motionpath.com> wrote:
> > > Currently in Og if you try and assign has many or joins many
> > > relationships to an unsaved object, the object is silently saved in the
> > > background (this is needed to make the relationship work), but if you do
> > > the same with refers_to/belongs_to/has_one, this does not happen,
> > > instead Og proceeds as if everything is fine essentially ignoring your
> > > request (and losing the relationship).
> > >
> > > I think all of these behaviours are wrong because you may not expect
> > > your object to have been saved in the has_many/joins_many case and you
> > > will not know why your code mysteriously breaks in the belongs_to case
> > > (this actually bit me while writing a unit test the other day and it
> > > took me a good five minutes to work out what was wrong).
> > >
> > > I believe that all of these cases should throw and exception as you
> > > should not be trying to assign relationships to unsaved objects, but
> > > before I make this happen I want to confirm with the list you agree this
> > > is the appropriate course of action. This way informs you that your code
> > > is broken and if you did want automatic saving all you'd need to do is:
> > >
> > > begin
> > >  obj.relation = remote
> > > rescue UnsavedException
> > >  obj.save!
> > >  remote.save!
> > >  retry
> > > end
> > >
> > > Of course, this is pretty silly since you should be doing:
> > >
> > > remote.save!
> > > obj.relation = remote
> > > obj.save!
> > >
> > > In your code anyway.
> > >
> > > At the very least, the refers_to/has_one/belongs_to relationships should
> > > have the same behaviour as the other relationships, be that auto-saving
> > > as they currently do, or as I believe, throwing exceptions.
> > >
> > > Comments?
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
>
> _______________________________________________
> Nitro-general mailing list
> Nitro-general at rubyforge.org
> http://rubyforge.org/mailman/listinfo/nitro-general
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 20:27:29 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 17:27:29 -0800
Subject: [Nitro] My Apologies for the email flood.
Message-ID: <bd93db0d0603301727k4a36d9b0oc3f2ad394ef1b0bc@mail.gmail.com>

Alot of these were threads I've marked as unread to ensure they didn't
get forgotten. Seems silly since we have devlab and tickets, and in a
period of unexpected free time I just decided to go through them.

We have alot of tickets, and possibly quite a few more if my flood has
any responses, all available here.

http://devlab.oree.ch/trac/nitrohq/report/1

So, to borrow a phrase, patches are welcome. :)

Many thanks,

Bryan

--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From bryan.a.soto at gmail.com  Thu Mar 30 21:15:50 2006
From: bryan.a.soto at gmail.com (Bryan Soto)
Date: Thu, 30 Mar 2006 18:15:50 -0800
Subject: [Nitro] [PATCH] Configuration + Og startup speedup
In-Reply-To: <bd93db0d0603301334k3a8d4b97k50b2c5b723f023be@mail.gmail.com>
References: <afd80b540603271106k635ebab6r44dd6e5b3b29378e@mail.gmail.com>
	<bd93db0d0603271214i229f66a8w11b1d00ca07162e1@mail.gmail.com>
	<afd80b540603290749y68401afbr60f030bfaa10ed94@mail.gmail.com>
	<bd93db0d0603301334k3a8d4b97k50b2c5b723f023be@mail.gmail.com>
Message-ID: <bd93db0d0603301815y55a533e6o3c05b05f1acb416b@mail.gmail.com>

On 3/30/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> Just as an fyi, this make some tests fail for some reason. I'll see if
> I can fix them up before applying.
>

You added that evil clear_all_settings method back to
Glue::Configuration again in tc_configuration. Is there a reason for
that? It causes the test cases after that to fail when you run rake
test for glue. Or perhaps the problem is actually in tc_fixture since
it fails after the settings are cleared?

  1) Error:
test_all(TestFixture):
NoMethodError: undefined method `[]' for nil:NilClass
    (eval):3:in `root_dir'
    ./test/glue/../../lib/glue/fixture.rb:70:in `initialize'
    ./test/glue/tc_fixture.rb:20:in `test_all'

  2) Error:
test_global(TestFixture):
NoMethodError: undefined method `[]' for nil:NilClass
    (eval):3:in `root_dir'
    ./test/glue/../../lib/glue/fixture.rb:70:in `initialize'
    ./test/glue/../../lib/glue/fixture.rb:23:in `load'
    ./test/glue/../../lib/glue/fixture.rb:22:in `load'
    ./test/glue/tc_fixture.rb:38:in `test_global'

36 tests, 160 assertions, 0 failures, 2 errors

Also some interesting errors in og:

D, [2006-03-30T17:48:44.984370 #19873] DEBUG -- : Og manageable classes: [TC_OgI
nheritance::Document, TC_OgInheritance::User, TC_OgInheritance::Car, TC_OgInheri
tance::Article, TC_OgInheritance::Photo, TC_OgInheritance::Admin]
I, [2006-03-30T17:48:45.145249 #19873]  INFO -- : Created table 'ogtc_oginherita
nce_user'.
D, [2006-03-30T17:48:45.150689 #19873] DEBUG -- : Created jointable 'ogj_tc_ogin
heritance_car_tc_oginheritance_user'.
I, [2006-03-30T17:48:45.183862 #19873]  INFO -- : Created table 'ogtc_oginherita
nce_car'.
I, [2006-03-30T17:48:45.332084 #19873]  INFO -- : WARNING: Table 'ogtc_oginherit
ance_user' is missing field 'car_oid integer' and :evolve_schema is not set to t
rue!

Then later,

  1) Error:
test_all(TC_OgInheritance):
Mysql::Error: Unknown column 'car_oid' in 'field list'
    (eval):5:in `query'
    (eval):5:in `og_insert'
    /mnt/zip/nitro.repo/og/lib/og/store.rb:104:in `save'
    /mnt/zip/nitro.repo/og/lib/og/entity.rb:125:in `create'
    ./test/og/tc_inheritance.rb:105:in `test_all'

Also,

  9) Error:
test_related(TestFixes):
NoMethodError: undefined method `relations' for
#<UnitTestClasses::RelationValidation:0xb7636c9c>
    ./test/og/tc_validation2.rb:197:in `test_related'

 10) Error:
test_value_validation(TestFixes):
NoMethodError: undefined method `store' for nil:NilClass
    /mnt/zip/nitro.repo/og/lib/og/entity.rb:32:in `save'
    ./test/og/tc_validation2.rb:82:in `assert_saved_without_errors'
    ./test/og/tc_validation2.rb:81:in `assert_saved_without_errors'
    ./test/og/tc_validation2.rb:111:in `test_value_validation'

The above occurs seven more times for different classes defined in
this particular test case...

 11) Error:
test_all(TestOgRevisable):
NoMethodError: undefined method `article_oid=' for
#<TestOgRevisable::Article::Revision:0xb76f294c>
    (eval):26:in `add_revision'
    /mnt/zip/nitro.repo/og/lib/og/collection.rb:123:in `<<'
    /mnt/zip/nitro.repo/og/lib/glue/revisable.rb:87:in `revise'
    ./test/glue/tc_revisable.rb:24:in `test_all'

I'll see what I can figure out...

Bryan

> On 3/29/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > Yeah,
> >
> > here is the patch, please submit it ;-)
> >
> > -g.
> >
> > On 3/27/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:
> > > Hey g,
> > >
> > > You forgot the patch. :)
> > >
> > > Thanks
> > >
> > > On 3/27/06, George Moschovitis <george.moschovitis at gmail.com> wrote:
> > > > Bryan, can you add these to the stable repo?
> > > >
> > > > thanks,
> > > > -g.
> > > >
> > > >
> > > > Mon Mar 27 21:59:51 EEST 2006  George Moschovitis <gm at navel.gr>
> > > >   * Startup optimization, don't search in ObjectSpace for managed classes.
> > > >
> > > > Mon Mar 27 21:43:44 EEST 2006  George Moschovitis <gm at navel.gr>
> > > >   * Further improved configuration system, even added some ruby magic,
> > > > check out the rdocs and the updated test case.
> > > >
> > > >
> > > >
> > > > --
> > > > http://www.gmosx.com
> > > > http://www.navel.gr
> > > > http://www.nitrohq.com
> > > >
> > > > _______________________________________________
> > > > Nitro-general mailing list
> > > > Nitro-general at rubyforge.org
> > > > http://rubyforge.org/mailman/listinfo/nitro-general
> > > >
> > >
> > >
> > > --
> > > "Never tell people how to do things. Tell them what to do and they
> > > will surprise you with their ingenuity." ?General George S. Patton
> > >
> > > _______________________________________________
> > > Nitro-general mailing list
> > > Nitro-general at rubyforge.org
> > > http://rubyforge.org/mailman/listinfo/nitro-general
> > >
> >
> >
> > --
> > http://www.gmosx.com
> > http://www.navel.gr
> > http://www.nitrohq.com
> >
> >
> > _______________________________________________
> > Nitro-general mailing list
> > Nitro-general at rubyforge.org
> > http://rubyforge.org/mailman/listinfo/nitro-general
> >
> >
> >
>
>
> --
> "Never tell people how to do things. Tell them what to do and they
> will surprise you with their ingenuity." ?General George S. Patton
>


--
"Never tell people how to do things. Tell them what to do and they
will surprise you with their ingenuity." ?General George S. Patton



From transfire at gmail.com  Thu Mar 30 23:02:49 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 30 Mar 2006 23:02:49 -0500
Subject: [Nitro] Setting default-values for properties
In-Reply-To: <bd93db0d0603301644x5cf76972v38247d9be17b9542@mail.gmail.com>
References: <200603301559.29451.manveru@weez.co.jp>
	<e0gs3l$vr4$1@sea.gmane.org>
	<4b6f054f0603300810o3f712a99k589c42440046e375@mail.gmail.com>
	<bd93db0d0603301644x5cf76972v38247d9be17b9542@mail.gmail.com>
Message-ID: <4b6f054f0603302002w56d0d54ch725faad2a1a8be6e@mail.gmail.com>

On 3/30/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:

> So your thinking this should be a part of Facets then rather than an
> Og specific feature?

The foundation of it. Nitro/Og could then write it's own annotation
actions as needed. Users could write there own too. Of course some
actions are obvious like :default, so those probably could be built
in, or at least easily activated.

T.



From transfire at gmail.com  Thu Mar 30 23:02:49 2006
From: transfire at gmail.com (TRANS)
Date: Thu, 30 Mar 2006 23:02:49 -0500
Subject: [Nitro] Setting default-values for properties
In-Reply-To: <bd93db0d0603301644x5cf76972v38247d9be17b9542@mail.gmail.com>
References: <200603301559.29451.manveru@weez.co.jp>
	<e0gs3l$vr4$1@sea.gmane.org>
	<4b6f054f0603300810o3f712a99k589c42440046e375@mail.gmail.com>
	<bd93db0d0603301644x5cf76972v38247d9be17b9542@mail.gmail.com>
Message-ID: <4b6f054f0603302002w56d0d54ch725faad2a1a8be6e@mail.gmail.com>

On 3/30/06, Bryan Soto <bryan.a.soto at gmail.com> wrote:

> So your thinking this should be a part of Facets then rather than an
> Og specific feature?

The foundation of it. Nitro/Og could then write it's own annotation
actions as needed. Users could write there own too. Of course some
actions are obvious like :default, so those probably could be built
in, or at least easily activated.

T.



From james_b at neurogami.com  Fri Mar 31 11:20:21 2006
From: james_b at neurogami.com (James Britt)
Date: Fri, 31 Mar 2006 09:20:21 -0700
Subject: [Nitro] Dynamic Settings
In-Reply-To: <bd93db0d0603301629p4bb5583foc6046207cf3df752@mail.gmail.com>
References: <44260863.5080501@neurogami.com>	<509369bc93828575beeb9fea6fab30e7@oggu.de>	<4428013C.8040309@neurogami.com>	<afd80b540603270758h446af406rcea21c7ce0fe33ff@mail.gmail.com>	<44281A0E.9020201@neurogami.com>
	<bd93db0d0603301629p4bb5583foc6046207cf3df752@mail.gmail.com>
Message-ID: <442D56C5.8080407@neurogami.com>

Bryan Soto wrote:
....
> 
> Do you think that is good enough to be added to the repo? 

Well, it works for me, but I haven't really thought about possible side 
effects.  Still, I don't think it is little more than a extension to how 
Nitro loads configs for the hard-coded set of options (debug, stage, and 
live), plus compulsory exceptions when requesting a "non-standard" file.



> If so, I'd
> be happy to write it up in darcs patch form. Although I could have
> swore Rob mentioned he stored configs in Yaml form. I'll have to see
> if I can dig that up in my inbox.
> 

Well, this looks for both yaml and rb config files (though I've had no 
luck getting YAML configs to work).  If you think it looks OK, then 
please, by all means, add it to the repo.

> 
>>Not happy with the hard-coded set of special-case modes, though.
> 
> 
> I suppose a plugin system with registration or const_get magic would
> be an option. Or did you have another idea perhaps?
> 

Haven't really thought about it.  I believe those options do more than 
simply set the database options and server port and address, and while a 
*.rb config file could probably do the same stuff, a plain YAML file 
might not.

One possibly useful feature might be to allow the merging of config 
options.  Tell Nitro, "Run with the debug (or live or stage) settings, 
but use this other config file to override specific settings."

-- 
James Britt