From gthiesfeld at gmail.com  Tue Apr  1 12:04:59 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Tue, 1 Apr 2008 11:04:59 -0500
Subject: [Rubyinstaller-devel] Gem based Mingw build environment
Message-ID: <e9ece6840804010904p44d5dba8v5da9331cc302cd99@mail.gmail.com>

I've been rereading the threads from the last few months, and in
"Windows Compilation madness"[1], a gem based Windows development
environment was discussed.  I think that being able to gem install a
mingw development environment is a really great idea.  In that thread
there were three reasons against this approach, and I want to discsuss
them:

Luis Lavena wrote:
> Roger Pack suggested me to bundle MinGW in a gem...:

> 1) a 8MB gem (!!!)

Ok, I understand that 'gem instal'l would have a problem with this,
but I think if we (or I, if no one else is interested) abstract out
the necessary bits from Luis'  installer3 rake recipes, we could just
distribute all ruby code.  We just need to make it fairly simple for
the user to get msys and mingw set up.

> 2) because mingw is inside the gem, will not be easy hook it into PATH

We could have a command in the gem's bin dir that would help us set
this up.  It would ask the user if he has msys and mingw installed,
and if so prompt for the path.  If not, it could download it into the
location of the user's choosing.  Then it could create a config file
in $HOME or $USERPROFILE, similar to what Hoe or Autotest or Rubyforge
uses.   This file would contain the paths to msys and mingw and any
other global configuration needed to make this work.  From there, the
users could require this gem into their rakefile, and have all the
tools they need.

> 3) that didn't solve the rbconfig issues.

This issue is one I'm not sure I understand, and it may be the
showstopper.  Anyway, those are my ideas I'm interested to hear what
people think.

Thanks,

Gordon


[1]  http://groups.google.com/group/comp.lang.ruby/browse_frm/thread/6b299623e53edbd9/71760b19769067e6?lnk=gst&q=lavena+mingw+gem#71760b19769067e6

From luislavena at gmail.com  Tue Apr  1 12:46:00 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Tue, 1 Apr 2008 13:46:00 -0300
Subject: [Rubyinstaller-devel] Gem based Mingw build environment
In-Reply-To: <e9ece6840804010904p44d5dba8v5da9331cc302cd99@mail.gmail.com>
References: <e9ece6840804010904p44d5dba8v5da9331cc302cd99@mail.gmail.com>
Message-ID: <71166b3b0804010946i77635af0w397cdb1ec5c14106@mail.gmail.com>

On Tue, Apr 1, 2008 at 1:04 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> I've been rereading the threads from the last few months, and in
>  "Windows Compilation madness"[1], a gem based Windows development
>  environment was discussed.  I think that being able to gem install a
>  mingw development environment is a really great idea.  In that thread
>  there were three reasons against this approach, and I want to discsuss
>  them:
>

hey Gordon, great feedback!


>  Luis Lavena wrote:
>  > Roger Pack suggested me to bundle MinGW in a gem...:
>
>  > 1) a 8MB gem (!!!)
>
>  Ok, I understand that 'gem instal'l would have a problem with this,
>  but I think if we (or I, if no one else is interested) abstract out
>  the necessary bits from Luis'  installer3 rake recipes, we could just
>  distribute all ruby code.  We just need to make it fairly simple for
>  the user to get msys and mingw set up.
>

At that time I was negative to that idea, but after thinking it a bit,
it started to sound good. Is something like apt-get build-essentials
or similar.

What I didn't liked at that time was that rubygems use open-uri and is
not possible keep track of partial downloads or similar, something we
achieved in installer3 adapting some code from Buildr project [1].

In any case, our next step is work on the MSI scripts (Microsoft
Installer build with WiX) to generate the Ruby 1.8 module and also
some modules for MinGW and MSYS.

Then create a Developer Kit that could be installed and hooked into
the $PATH, as I commented in my blog post [2] a few days back.

>  > 2) because mingw is inside the gem, will not be easy hook it into PATH
>
>  We could have a command in the gem's bin dir that would help us set
>  this up.  It would ask the user if he has msys and mingw installed,
>  and if so prompt for the path.  If not, it could download it into the
>  location of the user's choosing.  Then it could create a config file
>  in $HOME or $USERPROFILE, similar to what Hoe or Autotest or Rubyforge
>  uses.   This file would contain the paths to msys and mingw and any
>  other global configuration needed to make this work.  From there, the
>  users could require this gem into their rakefile, and have all the
>  tools they need.

I have played a bit under that road, but there is a small problem
cannot easily fix, see the following IRB and command line output:

D:\Users\Luis>set FOOBAR
Environment variable FOOBAR not defined

D:\Users\Luis>irb
irb(main):001:0> ENV['FOOBAR']
=> nil
irb(main):002:0> ENV['FOOBAR'] = "Something"
=> "Something"
irb(main):003:0> ENV['FOOBAR']
=> "Something"
irb(main):004:0> exit

D:\Users\Luis>set FOOBAR
Environment variable FOOBAR not defined

D:\Users\Luis>SET FOOBAR=Something

D:\Users\Luis>irb
irb(main):001:0> ENV['FOOBAR']
=> "Something"
irb(main):002:0> ENV.delete('FOOBAR')
=> "Something"
irb(main):003:0> ENV['FOOBAR']
=> nil
irb(main):004:0> exit

D:\Users\Luis>SET FOOBAR
FOOBAR=Something

===

As you can see, we cannot alter any environment variable inside Ruby
and make the change "persist" outside the Ruby VM. What we could do is
wrap around and inherit the new environment, that is just nasty and
feel dirty...

What Curt Hibbs and Andy Hunt ended doing with installer-win2 on
subversion repository is wrap the environment variable changes into a
batch file that was the responsible to call nmake.

Will be better if we install MinGW+MSYS outside the controller Ruby VM
and environment and provide a windows shortcut like "MinGW command
prompt" or something like that.

>  > 3) that didn't solve the rbconfig issues.
>
>  This issue is one I'm not sure I understand, and it may be the
>  showstopper.  Anyway, those are my ideas I'm interested to hear what
>  people think.

The rbconfig hack was related to the subject of keep using VC6 builds
and use MinGW as compiler for the extensions, since mkmf, the library
used to generate the makefiles for ruby extension is highly coupled
with the compiler used to generate the ruby installation, example:

Ruby 1.8 built with VC6 (require rbconfig):

RbConfig::CONFIG['CC']
=> "cl -nologo"
RbConfig::CONFIG['CFLAGS']
=> "-MD -Zi -O2b2xg- -G6"
RbConfig::CONFIG['AR']
=> "lib -nologo"

Now, Ruby 1.8 with MinGW

irb(main):001:0> require 'rbconfig'
=> true
irb(main):002:0> RbConfig::CONFIG['CC']
=> "gcc"
irb(main):003:0> RbConfig::CONFIG['CFLAGS']
=> "-g -O2 "
irb(main):004:0> RbConfig::CONFIG['AR']
=> "ar"

As you can see, we need to tweak the RbConfig options to make them
work with MinGW command tools, but there are other issues, like the
platform RubyGems uses (RUBY_PLATFORM) and it will try to fire 'nmake'
instead of 'make' when building the extensions bundled into a gem...

It get too much complicated if we go down that road, so instead of
"patching and hacking" current build of ruby, jump into MinGW
completely sounds more logical :-D

>  Thanks,

Thanks to you Gordon, for taking the time to analyze this and provide
your feedback. I wish we could do some of these stuff without hack
around Ruby, but avoiding ruby sometimes is healthier :-)

[1] http://incubator.apache.org/buildr/
[2] http://blog.mmediasys.com/2008/03/29/progress-of-one-click-installer-rubyinstaller/

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Tue Apr  1 15:39:14 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Tue, 1 Apr 2008 14:39:14 -0500
Subject: [Rubyinstaller-devel] Gem based Mingw build environment
In-Reply-To: <71166b3b0804010946i77635af0w397cdb1ec5c14106@mail.gmail.com>
References: <e9ece6840804010904p44d5dba8v5da9331cc302cd99@mail.gmail.com>
	<71166b3b0804010946i77635af0w397cdb1ec5c14106@mail.gmail.com>
Message-ID: <e9ece6840804011239s340ce66cs7264b42e09ac0756@mail.gmail.com>

On Tue, Apr 1, 2008 at 11:46 AM, Luis Lavena <luislavena at gmail.com> wrote:
>
>  Thanks to you Gordon, for taking the time to analyze this and provide
>  your feedback. I wish we could do some of these stuff without hack
>  around Ruby, but avoiding ruby sometimes is healthier :-)
>

I appreciate you taking the time to explain all of this to me.  So,
the package size issue and the rbconfig issues are resolved.  That
leaves us with the issue of setting env variables.  Then, it's just a
matter of which packaging mechanism to use (msi, or rubygems).

I don't have a lot of experience with compiling extensions.  I have
compiled the win32console extension with VC6, and with MinGW.  With
VC6, I had to download the Windows platform SDK, run two batch files
to set env vars, and finally I got it to build.  It took me hours to
get it working.  Maybe that's because I'd never done it before - for
someone with a lot of C/C++ Windows experience it may have been
trivial.  With Mingw, I hacked into your installer project's rake
recipes, used the msys_sh method, and I was done.  It took me 20
minutes tops.  I guess my question is, since you can provide a package
that builds ruby and all of it's stdlib extensions via a rakefile
without setting env vars, is it something that we need to worry about?

Maybe you're just making it look too easy, I don't know :-)

Thanks,

Gordon

From builder at mmediasys.com  Mon Apr  7 17:10:57 2008
From: builder at mmediasys.com (Integration Builder)
Date: Mon, 7 Apr 2008 18:10:57 -0300
Subject: [Rubyinstaller-devel] [CruiseControl] mongrel-stable_1-2-mingw32
	build 1002 failed
Message-ID: <47fa8de126160_83c1462a8815e@edicion01.tmail>

The build failed.

  Note: if you set Configuration.dashboard_url in config/site_config.rb, you'd see a link to the build page here.

CHANGES
-------
New revision 1002 detected
Revision 1002 committed by luislavena on 2008-04-06 00:14:45
GemPlugin was looking for 'init.rb' on every gem, that's nasty.

  M /branches/stable_1-2/projects/gem_plugin/lib/gem_plugin.rb
  M /branches/stable_1-2



BUILD LOG
---------
D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"  
D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/bin/ruby extconf.rb
(in D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work)
[CruiseControl] Invoking Rake task "clobber"
Cleaning
- ext/http11/http11.o
- ext/http11/http11_parser.o
- ext/http11/Makefile
- lib/http11.so
[CruiseControl] Invoking Rake task "compile"
checking for main() in -lc... no
creating Makefile
gcc -I. -I. -ID:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/1.8/i386-mingw32 -I.  -g -O2   -c http11.c
gcc -I. -I. -ID:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/1.8/i386-mingw32 -I.  -g -O2   -c http11_parser.c
gcc -shared -s -o http11.so http11.o http11_parser.o -L. -LD:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib -L.  -Wl,--enable-auto-image-base,--enable-auto-import,--export-all   -lmsvcrt-ruby18  -lshell32 -lws2_32  
D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/bin/ruby -Ilib;ext;bin;test "D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader.rb" "test/test_helper.rb" "test/unit/test_cgi_wrapper.rb" "test/unit/test_command.rb" "test/unit/test_conditional.rb" "test/unit/test_configurator.rb" "test/unit/test_debug.rb" "test/unit/test_handlers.rb" "test/unit/test_http_parser.rb" "test/unit/test_redirect_handler.rb" "test/unit/test_request_progress.rb" "test/unit/test_response.rb" "test/unit/test_stats.rb" "test/unit/test_uriclassifier.rb" "test/unit/test_ws.rb" 
[CruiseControl] Invoking Rake task "test"
./test/test_helper.rb:8: warning: already initialized constant HERE
Loaded suite D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader
Started
.......................F..E................................
Finished in 32.781 seconds.

  1) Failure:
test_horrible_queries(HttpParserTest)
    [./test/unit/test_http_parser.rb:105:in `test_horrible_queries'
     ./test/unit/test_http_parser.rb:103:in `times'
     ./test/unit/test_http_parser.rb:103:in `test_horrible_queries']:
<Mongrel::HttpParserError> exception expected but none was thrown.

  2) Error:
test_parse_strange_headers(HttpParserTest):
Mongrel::HttpParserError: Invalid HTTP format, parsing fails.
    ./test/unit/test_http_parser.rb:49:in `execute'
    ./test/unit/test_http_parser.rb:49:in `test_parse_strange_headers'

59 tests, 458 assertions, 1 failures, 1 errors
rake aborted!
Command failed with status (1): [D:/Users/Luis/ruby/ruby-186-svn-MINGW-ruby...]

(See full trace by running task with --trace)

dir : D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work
command : ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"
executed command : echo D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/build-1002/build.log && ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:\Users\Luis\.cruise\projects\mongrel-stable_1-2-mingw32\build-1002\build.log 2>&1
exitstatus: 1
STDERR TAIL START

STDERR TAIL END


PROJECT SETTINGS
----------------
# Project-specific configuration for CruiseControl.rb

Project.configure do |project|

  project.email_notifier.emails = ['mongrel-development at rubyforge.org', 'rubyinstaller-devel at rubyforge.org']

  project.rake_task = 'clobber compile test package'
  
  project.scheduler.polling_interval = 5.minutes

end


From builder at mmediasys.com  Mon Apr  7 17:12:03 2008
From: builder at mmediasys.com (Integration Builder)
Date: Mon, 7 Apr 2008 18:12:03 -0300
Subject: [Rubyinstaller-devel] [CruiseControl] mongrel-stable_1-2-mswin32
	build 1002 failed
Message-ID: <47fa8e236e988_88013de3a6141@edicion01.tmail>

The build failed.

  Note: if you set Configuration.dashboard_url in config/site_config.rb, you'd see a link to the build page here.

CHANGES
-------
New revision 1002 detected
Revision 1002 committed by luislavena on 2008-04-06 00:14:45
GemPlugin was looking for 'init.rb' on every gem, that's nasty.

  M /branches/stable_1-2/projects/gem_plugin/lib/gem_plugin.rb
  M /branches/stable_1-2



BUILD LOG
---------
D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"  
D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/bin/ruby extconf.rb
(in D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work)
[CruiseControl] Invoking Rake task "clobber"
Cleaning
- ext/http11/http11-i386-mswin32.def
- ext/http11/http11-i386-mswin32.exp
- ext/http11/http11-i386-mswin32.lib
- ext/http11/http11-i386-mswin32.pdb
- ext/http11/http11.obj
- ext/http11/http11_parser.obj
- ext/http11/Makefile
- ext/http11/vc60.pdb
- lib/http11.so
- log
- test_stderr.log
- test_stdout.log
[CruiseControl] Invoking Rake task "compile"
checking for main() in c.lib... no
creating Makefile

Microsoft (R) Program Maintenance Utility   Version 6.00.9782.0
Copyright (C) Microsoft Corp 1988-1998. All rights reserved.

	D:\Users\Luis\ruby\ruby-186-p114-VC6-rubygems-110\bin\ruby -e "puts 'EXPORTS', 'Init_http11'"  > http11-i386-mswin32.def
	cl -nologo -I. -I. -ID:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/i386-mswin32 -I. -MD -Zi -O2b2xg- -G6  -c -Tchttp11.c
http11.c
	cl -nologo -I. -I. -ID:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/i386-mswin32 -I. -MD -Zi -O2b2xg- -G6  -c -Tchttp11_parser.c
http11_parser.c
	cl -nologo -LD -Fehttp11.so http11.obj http11_parser.obj msvcrt-ruby18.lib  oldnames.lib user32.lib advapi32.lib ws2_32.lib  -link -incremental:no -debug -opt:ref -opt:icf -dll -libpath:"." -libpath:"D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib" -def:http11-i386-mswin32.def -implib:http11-i386-mswin32.lib -pdb:http11-i386-mswin32.pdb
   Creating library http11-i386-mswin32.lib and object http11-i386-mswin32.exp
D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/bin/ruby -Ilib;ext;bin;test "D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader.rb" "test/test_helper.rb" "test/unit/test_cgi_wrapper.rb" "test/unit/test_command.rb" "test/unit/test_conditional.rb" "test/unit/test_configurator.rb" "test/unit/test_debug.rb" "test/unit/test_handlers.rb" "test/unit/test_http_parser.rb" "test/unit/test_redirect_handler.rb" "test/unit/test_request_progress.rb" "test/unit/test_response.rb" "test/unit/test_stats.rb" "test/unit/test_uriclassifier.rb" "test/unit/test_ws.rb" 
[CruiseControl] Invoking Rake task "test"
./test/test_helper.rb:8: warning: already initialized constant HERE
Loaded suite D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader
Started
.......................F..E................................
Finished in 34.609 seconds.

  1) Failure:
test_horrible_queries(HttpParserTest)
    [./test/unit/test_http_parser.rb:105:in `test_horrible_queries'
     ./test/unit/test_http_parser.rb:103:in `times'
     ./test/unit/test_http_parser.rb:103:in `test_horrible_queries']:
<Mongrel::HttpParserError> exception expected but none was thrown.

  2) Error:
test_parse_strange_headers(HttpParserTest):
Mongrel::HttpParserError: Invalid HTTP format, parsing fails.
    ./test/unit/test_http_parser.rb:49:in `execute'
    ./test/unit/test_http_parser.rb:49:in `test_parse_strange_headers'

59 tests, 458 assertions, 1 failures, 1 errors
rake aborted!
Command failed with status (1): [D:/Users/Luis/ruby/ruby-186-p114-VC6-rubyg...]

(See full trace by running task with --trace)

dir : D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work
command : ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"
executed command : echo D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/build-1002/build.log && ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:\Users\Luis\.cruise\projects\mongrel-stable_1-2-mswin32\build-1002\build.log 2>&1
exitstatus: 1
STDERR TAIL START

STDERR TAIL END


PROJECT SETTINGS
----------------
# Project-specific configuration for CruiseControl.rb

Project.configure do |project|

  project.email_notifier.emails = ['mongrel-development at rubyforge.org', 'rubyinstaller-devel at rubyforge.org']

  project.rake_task = 'clobber compile test package'
  
  project.scheduler.polling_interval = 5.minutes

end


From dgleal at gmail.com  Thu Apr 10 06:33:29 2008
From: dgleal at gmail.com (David Leal)
Date: Thu, 10 Apr 2008 11:33:29 +0100
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
Message-ID: <1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>

(Posting this again because I forgot to subscribe to the list, and I don't
know if the mail is delivered in those conditions. Sorry if this is a double
post.)

Hi all,

so I spoke to Luis a few hours ago on IRC, because I was having problems
building ruby on windows using the new installer scripts. While I'm happy to
report that it now compiles flawlessly since the update to the last MinGW
version, I still have a problem installing any gem, because it can't find
zlib: "No such file to load -- zlib". I checked and zlib1.dll is in the
ruby_mingw/bin directory, and I have that in my path.

Any suggestions, other than "ditch Vista?"

Cheers,

David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rubyforge.org/pipermail/rubyinstaller-devel/attachments/20080410/c7d935c0/attachment-0001.html 

From luislavena at gmail.com  Thu Apr 10 08:13:53 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Thu, 10 Apr 2008 09:13:53 -0300
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
Message-ID: <71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>

On Thu, Apr 10, 2008 at 7:33 AM, David Leal <dgleal at gmail.com> wrote:
> (Posting this again because I forgot to subscribe to the list, and I don't
> know if the mail is delivered in those conditions. Sorry if this is a double
> post.)
>

No double post, it seems it didn't made through :-P

> Hi all,
>
> so I spoke to Luis a few hours ago on IRC, because I was having problems
> building ruby on windows using the new installer scripts. While I'm happy to
> report that it now compiles flawlessly since the update to the last MinGW
> version, I still have a problem installing any gem, because it can't find
> zlib: "No such file to load -- zlib". I checked and zlib1.dll is in the
> ruby_mingw/bin directory, and I have that in my path.
>

Fir of all, thank you for your interest and the time you took to
report this. I'm glad the update "partially" worked :-)

It seems that building of zlib extension failed. you can check if the
extension got build in ruby_build/ext/zlib (look for mkmf.log and
leftovers like .o files).

If none of these things exists (or even the folder), could be the root
of the problem.

This will sound silly, but can you try cleanup and try again?

>rake clean
(wait until sandbox is wiped)
>rake
(that should extract and compile everything from scratch, again).

You could see it also if it gets build during this phase, the
extension is the last to be build, check of the output after
"compiling zlib" (expected output):

compiling zlib
make[1]: Entering directory
`/d/Users/Luis/projects/oss/oci/installer3/dev/sandbox/ruby_build/ext/zlib'
gcc -I. -I../.. -I../../../ruby_1_8 -I../../../ruby_1_8/ext/zlib -DHAVE_ZLIB_H
-DOS_CODE=OS_WIN32   -g -O2   -c ../../../ruby_1_8/ext/zlib/zlib.c
gcc -shared -s -o ../../.ext/i386-mingw32/zlib.so zlib.o -L. -L../..
-L.  -Wl,--enable-auto-image-base,--enable-auto-import,--export-all
-lmsvcrt-ruby18 -lzdll  -lshell32 -lws2_32
make[1]: Leaving directory
`/d/Users/Luis/projects/oss/oci/installer3/dev/sandbox/ruby_build/ext/zlib'

> Any suggestions, other than "ditch Vista?"
>

You wouldn't hear me say something like that, never :-P

> Cheers,
>
> David
>

Please let me know if that helped and report any other issues, I'll
happily hack in solutions for this :-)

Regards,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From dgleal at gmail.com  Thu Apr 10 11:35:27 2008
From: dgleal at gmail.com (David Leal)
Date: Thu, 10 Apr 2008 16:35:27 +0100
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
Message-ID: <1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>

On Thu, Apr 10, 2008 at 1:13 PM, Luis Lavena <luislavena at gmail.com> wrote:
> Fir of all, thank you for your interest and the time you took to
> report this. I'm glad the update "partially" worked :-)

:) Actually, thank you for all your work and patience. Working with
Ruby in Windows is a nightmare right now, I'm glad you're set to
change that.

> It seems that building of zlib extension failed. you can check if the
> extension got build in ruby_build/ext/zlib (look for mkmf.log and
> leftovers like .o files).

Ok, I recompiled everything and got the same thing. There is a
mkmf.log, which I'm pasting below. I hope it helps.

have_library: checking for deflateReset() in -lz... -------------------- no

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -lz  -lws2_32  "
conftest.c: In function `t':
conftest.c:6: error: `deflateReset' undeclared (first use in this function)
conftest.c:6: error: (Each undeclared identifier is reported only once
conftest.c:6: error: for each function it appears in.)
checked program was:
/* begin */
1: #include <windows.h>
2: #include <winsock.h>
3:
4: /*top*/
5: int main() { return 0; }
6: int t() { void ((*volatile p)()); p = (void ((*)()))deflateReset; return 0; }
/* end */

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -lz  -lws2_32  "
\mingw\lib\gcc\mingw32\3.4.5\..\..\..\..\mingw32\bin\ld.exe: cannot find -lz
collect2: ld returned 1 exit status
checked program was:
/* begin */
1: /*top*/
2: int main() { return 0; }
3: int t() { deflateReset(); return 0; }
/* end */

--------------------

have_library: checking for deflateReset() in -llibz... -------------------- no

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -llibz  -lws2_32  "
conftest.c: In function `t':
conftest.c:6: error: `deflateReset' undeclared (first use in this function)
conftest.c:6: error: (Each undeclared identifier is reported only once
conftest.c:6: error: for each function it appears in.)
checked program was:
/* begin */
1: #include <windows.h>
2: #include <winsock.h>
3:
4: /*top*/
5: int main() { return 0; }
6: int t() { void ((*volatile p)()); p = (void ((*)()))deflateReset; return 0; }
/* end */

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -llibz  -lws2_32  "
\mingw\lib\gcc\mingw32\3.4.5\..\..\..\..\mingw32\bin\ld.exe: cannot find -llibz
collect2: ld returned 1 exit status
checked program was:
/* begin */
1: /*top*/
2: int main() { return 0; }
3: int t() { deflateReset(); return 0; }
/* end */

--------------------

have_library: checking for deflateReset() in -lzlib... -------------------- no

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -lzlib  -lws2_32  "
conftest.c: In function `t':
conftest.c:6: error: `deflateReset' undeclared (first use in this function)
conftest.c:6: error: (Each undeclared identifier is reported only once
conftest.c:6: error: for each function it appears in.)
checked program was:
/* begin */
1: #include <windows.h>
2: #include <winsock.h>
3:
4: /*top*/
5: int main() { return 0; }
6: int t() { void ((*volatile p)()); p = (void ((*)()))deflateReset; return 0; }
/* end */

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -lzlib  -lws2_32  "
\mingw\lib\gcc\mingw32\3.4.5\..\..\..\..\mingw32\bin\ld.exe: cannot find -lzlib
collect2: ld returned 1 exit status
checked program was:
/* begin */
1: /*top*/
2: int main() { return 0; }
3: int t() { deflateReset(); return 0; }
/* end */

--------------------

have_library: checking for deflateReset() in -lzdll... -------------------- no

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -lzdll  -lws2_32  "
conftest.c: In function `t':
conftest.c:6: error: `deflateReset' undeclared (first use in this function)
conftest.c:6: error: (Each undeclared identifier is reported only once
conftest.c:6: error: for each function it appears in.)
checked program was:
/* begin */
1: #include <windows.h>
2: #include <winsock.h>
3:
4: /*top*/
5: int main() { return 0; }
6: int t() { void ((*volatile p)()); p = (void ((*)()))deflateReset; return 0; }
/* end */

"gcc -o conftest -I../.. -I../../../ruby_1_8
-I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
    -lmsvcrt-ruby18-static -lzdll  -lws2_32  "
\mingw\lib\gcc\mingw32\3.4.5\..\..\..\..\mingw32\bin\ld.exe: cannot find -lzdll
collect2: ld returned 1 exit status
checked program was:
/* begin */
1: /*top*/
2: int main() { return 0; }
3: int t() { deflateReset(); return 0; }
/* end */

--------------------

So it seems lzdll cannot be found, but it's there, inside
extract_utils and inside mingw/bin.Could this be another Vista related
bug in mingw?

Again, thanks for your work, Luis.

Cheers,

David

From gthiesfeld at gmail.com  Thu Apr 10 13:27:17 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Thu, 10 Apr 2008 12:27:17 -0500
Subject: [Rubyinstaller-devel] compiling OpenSSL
Message-ID: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>

So, I just stumbled across this article[1] that talks about compiling
OpenSSL with Mingw for Ruby and Python.  I thought I'd give it a try
to see if it would help with the openssl test failures.  I'm having a
few problems.

The first problem is in the extract step.  Bsdtar errors out when it
tries to extract the latest OpenSSL tar.gz[2].  So, I've sort of
skipped that step for the moment.  I run rake extract, then I manually
extract these files where I need them.

The second problem is getting the order of things right in rake.  I'm
assuming I would need to configure, compile, and install the OpenSSL
files (*.dll and *.a) before I try to build Ruby, and my rake skills
aren't good enough to figure out how to do this.

So, I think I'm going to try to configure msys and mingw properly, and
try this as a manual process to see if there's any worth to this
approach.  But, before I do that, I thought I'd ask here.  Does this
seem like a worthwhile approach, or am I on a wild goose chase? :-)

[1] http://bowmansolutions.com/mingw-openldap/
[2] http://www.openssl.org/source/openssl-0.9.8g.tar.gz

Thanks,

Gordon

From luislavena at gmail.com  Thu Apr 10 16:01:47 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Thu, 10 Apr 2008 17:01:47 -0300
Subject: [Rubyinstaller-devel] compiling OpenSSL
In-Reply-To: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>
References: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>
Message-ID: <71166b3b0804101301h4a1b1f96j51e0c1814bf5c922@mail.gmail.com>

On Thu, Apr 10, 2008 at 2:27 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> So, I just stumbled across this article[1] that talks about compiling
>  OpenSSL with Mingw for Ruby and Python.  I thought I'd give it a try
>  to see if it would help with the openssl test failures.  I'm having a
>  few problems.
>

Hello Gordon!

I wasn't aware there are OpenSSL failures in the MinGW build. AFAIK
the only blocker I still have is test_readline, which hangs and halt
the whole 'rake check' process.

If you see the recipes, I've been using a pre-build binary from
GNUwin32 project instead of build it myself.

Right now the way the Rake tasks are organized is not good enough, and
can get brittle over time.

>  The first problem is in the extract step.  Bsdtar errors out when it
>  tries to extract the latest OpenSSL tar.gz[2].  So, I've sort of
>  skipped that step for the moment.  I run rake extract, then I manually
>  extract these files where I need them.
>

Can you provide the output or the error message you're getting when it
tries to extract it? Also, this OpenSSL need ot be extracted in it's
own directory, instead of using the OpenSSL package target location,
instead of MinGW one.

>  The second problem is getting the order of things right in rake.  I'm
>  assuming I would need to configure, compile, and install the OpenSSL
>  files (*.dll and *.a) before I try to build Ruby, and my rake skills
>  aren't good enough to figure out how to do this.
>

Yes and no, configure will be chained with configure of ruby
interpreter, and for compile will happen the same.

That wouldn't work unless we provide --with-openssl-dir or something
to the ruby configure process (to avoid clutter mingw and msys
installation with leftovers of these packages).

I think we need a new abstraction layer on top of Rake, that can be
more descriptive and precise of the dependencies, but don't have
enough time right now to design something :-P

>  So, I think I'm going to try to configure msys and mingw properly, and
>  try this as a manual process to see if there's any worth to this
>  approach.  But, before I do that, I thought I'd ask here.  Does this
>  seem like a worthwhile approach, or am I on a wild goose chase? :-)
>

Please do, I'll suggest you revert your changes, use latest.zip and
let rake do it's stuff, then use the sandbox/mingw and sandbox/msys as
base MinGW and start playing with manual build of OpenSSL.

I think is real worthy, since one of the goals of these recipes is cut
the dependency of externals builds, for ruby or it's dependencies.

The bad is some GNU projects (like readline) mess a lot the packages
and make almost impossible replicate the build procedure from sources.

>  Thanks,

Thanks to you man!, please keep us posted on this!

I'll hang out at #ruby-lang the weekend, trying to get other gems
working on Windows and MinGW.

Regards,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From luislavena at gmail.com  Thu Apr 10 20:11:59 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Thu, 10 Apr 2008 21:11:59 -0300
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
Message-ID: <71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>

On Thu, Apr 10, 2008 at 12:35 PM, David Leal <dgleal at gmail.com> wrote:
>
>  Ok, I recompiled everything and got the same thing. There is a
>  mkmf.log, which I'm pasting below. I hope it helps.
>
>  have_library: checking for deflateReset() in -lz... -------------------- no
>
>  "gcc -o conftest -I../.. -I../../../ruby_1_8
>  -I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
>     -lmsvcrt-ruby18-static -lz  -lws2_32  "
>  conftest.c: In function `t':
>  conftest.c:6: error: `deflateReset' undeclared (first use in this function)
>  conftest.c:6: error: (Each undeclared identifier is reported only once
>  conftest.c:6: error: for each function it appears in.)
>  checked program was:
>  /* begin */
>  1: #include <windows.h>
>  2: #include <winsock.h>
>  3:
>  4: /*top*/
>  5: int main() { return 0; }
>  6: int t() { void ((*volatile p)()); p = (void ((*)()))deflateReset; return 0; }
>  /* end */
>

I only can point one difference between the build process in XP
compared to Vista and is the library lookup paths passed into ld:

-L"." -L"../.."

versus the one I got:

-L. -L../..

that is naive, but is the only difference I spot and is significant.

Can you share with me the PATH and drive where you extracted the recipes?

I'm just curious why it is quoting library (lib) lookup paths.

Please let me know about this or contact me via gtalk so we can
workout a solution.
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From dgleal at gmail.com  Fri Apr 11 05:49:08 2008
From: dgleal at gmail.com (David Leal)
Date: Fri, 11 Apr 2008 10:49:08 +0100
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
Message-ID: <1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>

Hi Luis,

here's my path:

C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Program
Files\ATI Technologies\ATI.ACE;C:\Program
Files\Git\cmd;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Program
Files\Bazaar;C:\Program
Files\Subversion\bin;c:\Utils\jruby-1.1\bin;C:\Program
Files\Java\jdk1.6.0_05\bin;C:\Program
Files\Mercurial;c:\utils\ruby\bin

I'm building this on c:\Users\David\Development\ruby-installer

I tried to do this using jruby (jruby -S rake ...) but the result is
the same. I also found out that the problem is not only with zlib.
Openssl doesn't build also, neither does readline, and I'm pretty sure
that every extension that depends on an external lib has this problem.
rbconfig is generated with the LIBPATHFLAGS option set to "-L\"%s\"",
but I can't figure out where that is coming from.

Cheers,

David

On Fri, Apr 11, 2008 at 1:11 AM, Luis Lavena <luislavena at gmail.com> wrote:
> On Thu, Apr 10, 2008 at 12:35 PM, David Leal <dgleal at gmail.com> wrote:
>  >
>  >  Ok, I recompiled everything and got the same thing. There is a
>  >  mkmf.log, which I'm pasting below. I hope it helps.
>  >
>  >  have_library: checking for deflateReset() in -lz... -------------------- no
>  >
>  >  "gcc -o conftest -I../.. -I../../../ruby_1_8
>  >  -I../../../ruby_1_8/ext/zlib  -g -O2  conftest.c  -L"." -L"../.." -L.
>  >     -lmsvcrt-ruby18-static -lz  -lws2_32  "
>  >  conftest.c: In function `t':
>  >  conftest.c:6: error: `deflateReset' undeclared (first use in this function)
>  >  conftest.c:6: error: (Each undeclared identifier is reported only once
>  >  conftest.c:6: error: for each function it appears in.)
>  >  checked program was:
>  >  /* begin */
>  >  1: #include <windows.h>
>  >  2: #include <winsock.h>
>  >  3:
>  >  4: /*top*/
>  >  5: int main() { return 0; }
>  >  6: int t() { void ((*volatile p)()); p = (void ((*)()))deflateReset; return 0; }
>  >  /* end */
>  >
>
>  I only can point one difference between the build process in XP
>  compared to Vista and is the library lookup paths passed into ld:
>
>  -L"." -L"../.."
>
>  versus the one I got:
>
>  -L. -L../..
>
>  that is naive, but is the only difference I spot and is significant.
>
>  Can you share with me the PATH and drive where you extracted the recipes?
>
>  I'm just curious why it is quoting library (lib) lookup paths.
>
>  Please let me know about this or contact me via gtalk so we can
>  workout a solution.
>
>
> --
>  Luis Lavena
>  Multimedia systems
>  -
>  Human beings, who are almost unique in having the ability to learn from
>  the experience of others, are also remarkable for their apparent
>  disinclination to do so.
>  Douglas Adams
>  _______________________________________________
>  Rubyinstaller-devel mailing list
>  Rubyinstaller-devel at rubyforge.org
>  http://rubyforge.org/mailman/listinfo/rubyinstaller-devel
>

From luislavena at gmail.com  Fri Apr 11 10:06:16 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Fri, 11 Apr 2008 11:06:16 -0300
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
Message-ID: <71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>

On Fri, Apr 11, 2008 at 6:49 AM, David Leal <dgleal at gmail.com> wrote:
> Hi Luis,
>
>  here's my path:
>
>  C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Program
>  Files\ATI Technologies\ATI.ACE;C:\Program
>  Files\Git\cmd;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Program
>  Files\Bazaar;C:\Program
>  Files\Subversion\bin;c:\Utils\jruby-1.1\bin;C:\Program
>  Files\Java\jdk1.6.0_05\bin;C:\Program
>  Files\Mercurial;c:\utils\ruby\bin
>
>  I'm building this on c:\Users\David\Development\ruby-installer
>

Wow, too weird... I can say is the first time I see something like this :-)

Just for the sake of testing, can you move c:\utils\ruby\bin on top of the path?

SET PATH=c:\utils\ruby\bin;%PATH%

I know that sounds stupid (not silly), but what to exhaust all these
issues before restore my notebook vista and give a whirl in there :-P

>  I tried to do this using jruby (jruby -S rake ...) but the result is
>  the same. I also found out that the problem is not only with zlib.
>  Openssl doesn't build also, neither does readline, and I'm pretty sure
>  that every extension that depends on an external lib has this problem.
>  rbconfig is generated with the LIBPATHFLAGS option set to "-L\"%s\"",
>  but I can't figure out where that is coming from.
>

That is taken from the generated config.h file in ruby_build

Just run configure step by its own and peek what ended in there (of
course, on a clean sandbox and after extract and prepare).

>  Cheers,
>

Thank you David, I'll see what can do to workaround this the weekend.
Sorry the inconvenience.

In the meantime you can grab a pre-build package from my dump server:

http://dump.mmediasys.com/installer3

Regards,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Fri Apr 11 16:57:34 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Fri, 11 Apr 2008 15:57:34 -0500
Subject: [Rubyinstaller-devel] compiling OpenSSL
In-Reply-To: <71166b3b0804101301h4a1b1f96j51e0c1814bf5c922@mail.gmail.com>
References: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>
	<71166b3b0804101301h4a1b1f96j51e0c1814bf5c922@mail.gmail.com>
Message-ID: <e9ece6840804111357y44d84fb0v510da35a7add552f@mail.gmail.com>

On Thu, Apr 10, 2008 at 3:01 PM, Luis Lavena <luislavena at gmail.com> wrote:
>  Hello Gordon!
>
>  I wasn't aware there are OpenSSL failures in the MinGW build. AFAIK
>  the only blocker I still have is test_readline, which hangs and halt
>  the whole 'rake check' process.

I'm running bzr revision 52.  I get this on 'rake check'.

./ruby_1_8/test/openssl/test_hmac.rb:14: [BUG] Segmentation fault
ruby 1.8.6 (2008-03-03) [i386-mingw32]


This application has requested the Runtime to terminate it in an unusual way.
Please contact the application's support team for more information.
make: *** [test-all] Error 3
rake aborted!
Command failed with status (2): ["C:/ruby/repo/installer3/sandbox/msys/bin/...]

(See full trace by running task with --trace)


>  Can you provide the output or the error message you're getting when it
>  tries to extract it?

"C:/ruby/repo/oci/sandbox/extract_utils/bsdtar.exe" xzf
"C:/ruby/repo/oci/downloads/openssl-0.9.8g.tar.gz"
openssl-0.9.8g/apps/md4.c: Can't create 'openssl-0.9.8g/apps/md4.c':
No such file or directory
openssl-0.9.8g/include/openssl/dtls1.h: Can't create
'openssl-0.9.8g/include/openssl/dtls1.h': No such file or directory
openssl-0.9.8g/include/openssl/kssl.h: Can't create
'openssl-0.9.8g/include/openssl/kssl.h': No such file or directory
openssl-0.9.8g/include/openssl/ssl.h: Can't create
'openssl-0.9.8g/include/openssl/ssl.h': No such file or directory
openssl-0.9.8g/include/openssl/ssl2.h: Can't create
'openssl-0.9.8g/include/openssl/ssl2.h': No such file or directory
openssl-0.9.8g/include/openssl/ssl23.h: Can't create
'openssl-0.9.8g/include/openssl/ssl23.h': No such file or directory
openssl-0.9.8g/include/openssl/ssl3.h: Can't create
'openssl-0.9.8g/include/openssl/ssl3.h': No such file or directory
openssl-0.9.8g/include/openssl/tls1.h: Can't create
'openssl-0.9.8g/include/openssl/tls1.h': No such file or directory
bsdtar.exe: Error exit delayed from previous errors.
rake aborted!
Command failed with status (1): ["C:/ruby/repo/oci/sandbox/extract_utils/bs...]

(See full trace by running task with --trace)

>  Also, this OpenSSL need ot be extracted in it's
>  own directory, instead of using the OpenSSL package target location,
>  instead of MinGW one.

Right, I had set it up with a similar folder structure to Ruby's
(openssl, openssl_build, openssl_mingw).

>  Please do, I'll suggest you revert your changes, use latest.zip and
>  let rake do it's stuff, then use the sandbox/mingw and sandbox/msys as
>  base MinGW and start playing with manual build of OpenSSL.

I did bzr revert, and I'm going to give this a try.

>  I'll hang out at #ruby-lang the weekend, trying to get other gems
>  working on Windows and MinGW.

I'll try to stop by.  I could use some help ;)

Gordon Thiesfeld

From luislavena at gmail.com  Fri Apr 11 17:57:35 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Fri, 11 Apr 2008 18:57:35 -0300
Subject: [Rubyinstaller-devel] compiling OpenSSL
In-Reply-To: <e9ece6840804111357y44d84fb0v510da35a7add552f@mail.gmail.com>
References: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>
	<71166b3b0804101301h4a1b1f96j51e0c1814bf5c922@mail.gmail.com>
	<e9ece6840804111357y44d84fb0v510da35a7add552f@mail.gmail.com>
Message-ID: <71166b3b0804111457p2f97b037r576ffdbc147a76a3@mail.gmail.com>

On Fri, Apr 11, 2008 at 5:57 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> On Thu, Apr 10, 2008 at 3:01 PM, Luis Lavena <luislavena at gmail.com> wrote:
>  >  Hello Gordon!
>  >
>  >  I wasn't aware there are OpenSSL failures in the MinGW build. AFAIK
>  >  the only blocker I still have is test_readline, which hangs and halt
>  >  the whole 'rake check' process.
>
>  I'm running bzr revision 52.  I get this on 'rake check'.
>
>  ./ruby_1_8/test/openssl/test_hmac.rb:14: [BUG] Segmentation fault
>  ruby 1.8.6 (2008-03-03) [i386-mingw32]
>

That segfault came from your OpenSSL change, or that is generated by
"out of the box" recipes?

Can you provide information about your Windows installation?
(XP/Vista, home, ultimate, service pack level, etc).

Also will be helpful know your PATH environment variable:

ECHO %PATH% (or SET PATH without params)

>
>  This application has requested the Runtime to terminate it in an unusual way.
>  Please contact the application's support team for more information.
>  make: *** [test-all] Error 3
>  rake aborted!
>  Command failed with status (2): ["C:/ruby/repo/installer3/sandbox/msys/bin/...]
>
>  (See full trace by running task with --trace)
>
>
>
>  >  Can you provide the output or the error message you're getting when it
>  >  tries to extract it?
>
>  "C:/ruby/repo/oci/sandbox/extract_utils/bsdtar.exe" xzf
>  "C:/ruby/repo/oci/downloads/openssl-0.9.8g.tar.gz"
>  openssl-0.9.8g/apps/md4.c: Can't create 'openssl-0.9.8g/apps/md4.c':
>  No such file or directory

Ok, no doubt bsdtar cannot handle latest openssl packages. I'm
checking right now what version can and cannot be unpacked:

0.9.7l, 0.9.7m, 0.9.8d, 0.9.8e, 0.9.8f and 0.9.8g all fail in the same files:

openssl-0.9.7l/apps/md4.c: Can't create 'openssl-0.9.7l/apps/md4.c':
No such file or directory
openssl-0.9.7l/include/openssl/kssl.h: Can't create
'openssl-0.9.7l/include/openssl/kssl.h': No such file or directory
openssl-0.9.7l/include/openssl/ssl23.h: Can't create
'openssl-0.9.7l/include/openssl/ssl23.h': No such file or directory
openssl-0.9.7l/include/openssl/ssl2.h: Can't create
'openssl-0.9.7l/include/openssl/ssl2.h': No such file or directory
openssl-0.9.7l/include/openssl/ssl3.h: Can't create
'openssl-0.9.7l/include/openssl/ssl3.h': No such file or directory
openssl-0.9.7l/include/openssl/ssl.h: Can't create
'openssl-0.9.7l/include/openssl/ssl.h': No such file or directory
openssl-0.9.7l/include/openssl/tls1.h: Can't create
'openssl-0.9.7l/include/openssl/tls1.h': No such file or directory
openssl-0.9.7l/test/fips_aes_data: Can't create
'openssl-0.9.7l/test/fips_aes_data': No such file or directory

If OpenSSL cannot be extracted using simple tar, then bsdtar is not
enough or openssl packages are corrupt somehow.

>
>  >  Also, this OpenSSL need ot be extracted in it's
>  >  own directory, instead of using the OpenSSL package target location,
>  >  instead of MinGW one.
>
>  Right, I had set it up with a similar folder structure to Ruby's
>  (openssl, openssl_build, openssl_mingw).
>
>
>  >  Please do, I'll suggest you revert your changes, use latest.zip and
>  >  let rake do it's stuff, then use the sandbox/mingw and sandbox/msys as
>  >  base MinGW and start playing with manual build of OpenSSL.
>
>  I did bzr revert, and I'm going to give this a try.
>
>
>  >  I'll hang out at #ruby-lang the weekend, trying to get other gems
>  >  working on Windows and MinGW.
>
>  I'll try to stop by.  I could use some help ;)
>

Good, me too ;-)

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From dgleal at gmail.com  Fri Apr 11 18:01:33 2008
From: dgleal at gmail.com (David Leal)
Date: Sat, 12 Apr 2008 00:01:33 +0200
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
	<71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
Message-ID: <1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>

On 4/11/08, Luis Lavena <luislavena at gmail.com> wrote:

>  Just for the sake of testing, can you move c:\utils\ruby\bin on top of the path?

Oh, I did that already. The ruby entry used to be the first in the
path, but I changed it to see if that's what was causing the troubles.

>  Just run configure step by its own and peek what ended in there (of
>  course, on a clean sandbox and after extract and prepare).

I sent you an email with all the files inside ruby_build. I hope it helps.

>  Thank you David, I'll see what can do to workaround this the weekend.
>  Sorry the inconvenience.

No need to apologize. You're doing a great favor to us all. We're the
ones in debt to you.

>  In the meantime you can grab a pre-build package from my dump server:
>
>  http://dump.mmediasys.com/installer3

Thanks, I'll do that.

Cheers,

David

From gthiesfeld at gmail.com  Fri Apr 11 20:12:06 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Fri, 11 Apr 2008 19:12:06 -0500
Subject: [Rubyinstaller-devel] compiling OpenSSL
In-Reply-To: <71166b3b0804111457p2f97b037r576ffdbc147a76a3@mail.gmail.com>
References: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>
	<71166b3b0804101301h4a1b1f96j51e0c1814bf5c922@mail.gmail.com>
	<e9ece6840804111357y44d84fb0v510da35a7add552f@mail.gmail.com>
	<71166b3b0804111457p2f97b037r576ffdbc147a76a3@mail.gmail.com>
Message-ID: <e9ece6840804111712p7919c141vd449b19dfd670463@mail.gmail.com>

On Fri, Apr 11, 2008 at 4:57 PM, Luis Lavena <luislavena at gmail.com> wrote:
>
>  That segfault came from your OpenSSL change, or that is generated by
>  "out of the box" recipes?
>

With the "out of the box" recipes.  I haven't actually gotten OpenSSL
to compile yet ;)

>  Can you provide information about your Windows installation?
>  (XP/Vista, home, ultimate, service pack level, etc).
>

Win XP Pro SP2

>  Also will be helpful know your PATH environment variable:
>
>  ECHO %PATH% (or SET PATH without params)


C:\ruby\repo\oci>path
PATH=c:\wix;c:\ruby\ruby186\bin;c:\windows\system32;c:\windows\system;C:\Program
Files\Subversion\bin;C:\Program Files\Git\cmd;C:\Program
Files\Bazaar;C:\bin;C:\WINDOWS\Microsoft.NET\Framework\v2.0.50727;C:\Program
Files\Putty;c:\bin\console2

I also tried putting ruby_mingw\bin folder to the front of the path,
and that didn't help.

>  Ok, no doubt bsdtar cannot handle latest openssl packages. I'm
>  checking right now what version can and cannot be unpacked:

>  If OpenSSL cannot be extracted using simple tar, then bsdtar is not
>  enough or openssl packages are corrupt somehow.

7zip can open them ok, and Winzip too.  I was thinking about writing a
new extract method using 7zip (7z.exe), but that would be a dependency
(required to be installed and required to be in path).  I guess before
I even think about that, I should worry about getting OpenSSL
compiled, and working ;)

Thanks,

Gordon

From luislavena at gmail.com  Fri Apr 11 20:18:12 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Fri, 11 Apr 2008 21:18:12 -0300
Subject: [Rubyinstaller-devel] compiling OpenSSL
In-Reply-To: <e9ece6840804111712p7919c141vd449b19dfd670463@mail.gmail.com>
References: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>
	<71166b3b0804101301h4a1b1f96j51e0c1814bf5c922@mail.gmail.com>
	<e9ece6840804111357y44d84fb0v510da35a7add552f@mail.gmail.com>
	<71166b3b0804111457p2f97b037r576ffdbc147a76a3@mail.gmail.com>
	<e9ece6840804111712p7919c141vd449b19dfd670463@mail.gmail.com>
Message-ID: <71166b3b0804111718y1cd855ecv7321e90e48db08a0@mail.gmail.com>

On Fri, Apr 11, 2008 at 9:12 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> On Fri, Apr 11, 2008 at 4:57 PM, Luis Lavena <luislavena at gmail.com> wrote:
>  >
>  >  That segfault came from your OpenSSL change, or that is generated by
>  >  "out of the box" recipes?
>  >
>
>  With the "out of the box" recipes.  I haven't actually gotten OpenSSL
>  to compile yet ;)
>

Hehehe, ok, then it seems something smells fishy down there...

Can you use Dependency Walker on top of the openssl.so file? it will
try to resolve to msvcrt-ruby18.dll, libeay32.dll and libssl32.dll.

Please take note the path where these files resolves to, it seems
openssl.so, even is linked correctly at build time, at runtime it's
resolving the location of one of these dependencies wrong.

>
> >  If OpenSSL cannot be extracted using simple tar, then bsdtar is not
>  >  enough or openssl packages are corrupt somehow.
>
>  7zip can open them ok, and Winzip too.  I was thinking about writing a
>  new extract method using 7zip (7z.exe), but that would be a dependency
>  (required to be installed and required to be in path).  I guess before
>  I even think about that, I should worry about getting OpenSSL
>  compiled, and working ;)
>

Maybe we can get a 7zip command line version package somewhere, as
long is .zip file (enough to bootstrap it with rubyzip).

Thank you for your time, I'll try to see what I can do for this later tonight.

Regards,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From luislavena at gmail.com  Fri Apr 11 21:32:12 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Fri, 11 Apr 2008 22:32:12 -0300
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
	<71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
	<1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>
Message-ID: <71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>

On Fri, Apr 11, 2008 at 7:01 PM, David Leal <dgleal at gmail.com> wrote:
> On 4/11/08, Luis Lavena <luislavena at gmail.com> wrote:
>
>  >  Just run configure step by its own and peek what ended in there (of
>  >  course, on a clean sandbox and after extract and prepare).
>
>  I sent you an email with all the files inside ruby_build. I hope it helps.
>

Yes, it helped and I think I spotted where the problem lies.

Based on the files, it appears you updated the autoconf package (from
2.59 to 2.61). That is correct?

"generated by GNU Autoconf 2.61."

that is part of the config.log file.

I sticked to 2.59 since the new set of files generated by 2.61 didn't
worked properly.

config.log (2.59):

LIBPATHFLAG=' -L%s'

your config.log (2.61):

LIBPATHFLAG=' -L"%s"'

Somehow, dunno from where, the newer version of autoconf is being
called. The only one I see from your path that could be doing that is
Git, even more precise if you have a .bashrc file in your HOME folder
that is altering the $PATH...

Try running msys.bat from sandbox/msys and 'which autoconf' and it
should point to /bin/autoconf (7,663 bytes) and check the version
information around line 130 of that file.

Please let me know how it ended.
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Fri Apr 11 22:26:12 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Fri, 11 Apr 2008 21:26:12 -0500
Subject: [Rubyinstaller-devel] compiling OpenSSL
In-Reply-To: <71166b3b0804111718y1cd855ecv7321e90e48db08a0@mail.gmail.com>
References: <e9ece6840804101027had502f6ub7be570a2f6a3744@mail.gmail.com>
	<71166b3b0804101301h4a1b1f96j51e0c1814bf5c922@mail.gmail.com>
	<e9ece6840804111357y44d84fb0v510da35a7add552f@mail.gmail.com>
	<71166b3b0804111457p2f97b037r576ffdbc147a76a3@mail.gmail.com>
	<e9ece6840804111712p7919c141vd449b19dfd670463@mail.gmail.com>
	<71166b3b0804111718y1cd855ecv7321e90e48db08a0@mail.gmail.com>
Message-ID: <e9ece6840804111926p401587a4jad6b40ad48141043@mail.gmail.com>

On Fri, Apr 11, 2008 at 7:18 PM, Luis Lavena <luislavena at gmail.com> wrote:
> On Fri, Apr 11, 2008 at 9:12 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
>  > On Fri, Apr 11, 2008 at 4:57 PM, Luis Lavena <luislavena at gmail.com> wrote:
>  >  >
>  >  >  That segfault came from your OpenSSL change, or that is generated by
>  >  >  "out of the box" recipes?
>  >  >
>  >
>  >  With the "out of the box" recipes.  I haven't actually gotten OpenSSL
>  >  to compile yet ;)
>  >
>
>  Hehehe, ok, then it seems something smells fishy down there...
>
>  Can you use Dependency Walker on top of the openssl.so file? it will
>  try to resolve to msvcrt-ruby18.dll, libeay32.dll and libssl32.dll.
>
>  Please take note the path where these files resolves to, it seems
>  openssl.so, even is linked correctly at build time, at runtime it's
>  resolving the location of one of these dependencies wrong.

Good call.  I have libeay32.dll and libssl32.dll in
C:\Windows\system32.  I'll clean that up and try again tomorrow
morning.

>  >
>  > >  If OpenSSL cannot be extracted using simple tar, then bsdtar is not
>  >  >  enough or openssl packages are corrupt somehow.
>  >
>  >  7zip can open them ok, and Winzip too.  I was thinking about writing a
>  >  new extract method using 7zip (7z.exe), but that would be a dependency
>  >  (required to be installed and required to be in path).  I guess before
>  >  I even think about that, I should worry about getting OpenSSL
>  >  compiled, and working ;)
>  >
>
>  Maybe we can get a 7zip command line version package somewhere, as
>  long is .zip file (enough to bootstrap it with rubyzip).
>

Yes, here it is:

http://downloads.sourceforge.net/sevenzip/7za457.zip

The only thing is that 7z.exe can't extract files in a single step,
you have to extract from gz or bz2, etc.,  then extract from tar.  I
don't think that would be too big of a problem, though.

Thanks again, Luis.


Gordon

From dgleal at gmail.com  Sat Apr 12 09:51:18 2008
From: dgleal at gmail.com (David Leal)
Date: Sat, 12 Apr 2008 14:51:18 +0100
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
	<71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
	<1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>
	<71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>
Message-ID: <1b86cb980804120651u6346a65en80dbfb4cbc8071da@mail.gmail.com>

On Sat, Apr 12, 2008 at 2:32 AM, Luis Lavena <luislavena at gmail.com> wrote:
>  Based on the files, it appears you updated the autoconf package (from
>  2.59 to 2.61). That is correct?
>
>  "generated by GNU Autoconf 2.61."

If you look at the ruby-1.8.6-p114.tar.gz, the configure file inside
already says generated by autoconf 2.61, so I think the problem lies
there.

Anyway, I cleared my PATH, only leaving c:\Windows,
c:\Windows\System32 and c:\Windows\System32\Wbem. It still says that
the files were generated by autoconf 2.61. The msys autoconf is
version 2.59.

Deleting configure, running autoconf and then running rake configure
doesn't help. I'm thinking that there's something else I should do,
but I don't know what.

What do you think, Luis?

Thanks,

David

From luislavena at gmail.com  Sat Apr 12 11:21:12 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sat, 12 Apr 2008 12:21:12 -0300
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <1b86cb980804120651u6346a65en80dbfb4cbc8071da@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
	<71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
	<1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>
	<71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>
	<1b86cb980804120651u6346a65en80dbfb4cbc8071da@mail.gmail.com>
Message-ID: <71166b3b0804120821l121a4de3i63115325fbf103ad@mail.gmail.com>

On Sat, Apr 12, 2008 at 10:51 AM, David Leal <dgleal at gmail.com> wrote:
> On Sat, Apr 12, 2008 at 2:32 AM, Luis Lavena <luislavena at gmail.com> wrote:
>  >  Based on the files, it appears you updated the autoconf package (from
>  >  2.59 to 2.61). That is correct?
>  >
>  >  "generated by GNU Autoconf 2.61."
>
>  If you look at the ruby-1.8.6-p114.tar.gz, the configure file inside
>  already says generated by autoconf 2.61, so I think the problem lies
>  there.
>

Let me try running from the source file, since I was using checkout
mode (rake CHECKOUT=1) to workaround a series of bugs related to
Readline and pooling (eat your cpu) and rubygems too.

Updating the recipes for RubyGems 1.1.1 and trying again right now :-)

>  Anyway, I cleared my PATH, only leaving c:\Windows,
>  c:\Windows\System32 and c:\Windows\System32\Wbem. It still says that
>  the files were generated by autoconf 2.61. The msys autoconf is
>  version 2.59.
>

It seems the problems lies in the values generated by autoconf 2.61
compared to the ones form 2.59. Since I was running form source, been
always generating the configure script with autoconf, so didn't found
that until you exposed it.

(which is good, give me enough arguments to scream at ruby-core list) :-)

>  Deleting configure, running autoconf and then running rake configure
>  doesn't help. I'm thinking that there's something else I should do,
>  but I don't know what.
>
>  What do you think, Luis?
>

Working on it, will post a followup in a few minutes.
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From luislavena at gmail.com  Sat Apr 12 13:25:45 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sat, 12 Apr 2008 14:25:45 -0300
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <71166b3b0804120821l121a4de3i63115325fbf103ad@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
	<71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
	<1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>
	<71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>
	<1b86cb980804120651u6346a65en80dbfb4cbc8071da@mail.gmail.com>
	<71166b3b0804120821l121a4de3i63115325fbf103ad@mail.gmail.com>
Message-ID: <71166b3b0804121025l2859524ai144225a138d83797@mail.gmail.com>

On Sat, Apr 12, 2008 at 12:21 PM, Luis Lavena <luislavena at gmail.com> wrote:
>
>  Let me try running from the source file, since I was using checkout
>  mode (rake CHECKOUT=1) to workaround a series of bugs related to
>  Readline and pooling (eat your cpu) and rubygems too.
>
>  Updating the recipes for RubyGems 1.1.1 and trying again right now :-)
>

Ok, it worked in this end. even with the included configure script
generated in 2.61, it works. (openssl get build, zlib too).

So I guess is related to Vista now.

Can you try the "rake CHECKOUT=1" build instead (you need to have
subversion on the path) and send me your results?

Based on that I'll know how to deal with this issue.

Please apologize all the inconveniences and my request of testing :-)

Regards,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From builder at mmediasys.com  Tue Apr 15 15:15:44 2008
From: builder at mmediasys.com (Integration Builder)
Date: Tue, 15 Apr 2008 16:15:44 -0300
Subject: [Rubyinstaller-devel] [CruiseControl] rubygems-mswin32 build 1704.1
	fixed
Message-ID: <4804fee035390_fb813de3a61ae@edicion01.tmail>

The build has been fixed.

  Note: if you set Configuration.dashboard_url in config/site_config.rb, you'd see a link to the build page here.

CHANGES
-------
New revision 1704 detected
Revision 1704 committed by drbrain on 2008-04-15 00:40:42
Add Gem::RemoteFetcher::fetcher= with lib/rubygems/test_utilities.rb
  A /trunk/lib/rubygems/test_utilities.rb
  M /trunk/test/test_gem_dependency_installer.rb
  M /trunk/test/test_gem_remote_fetcher.rb
  M /trunk/test/gemutilities.rb
  M /trunk/test/test_gem_commands_sources_command.rb
Revision 1703 committed by drbrain on 2008-04-14 23:59:58
Expose FakeFetcher, TempIO via rubygems/test_utilities.rb

  M /trunk/test/test_gem_dependency_installer.rb
  M /trunk/test/gemutilities.rb
  M /trunk/ChangeLog



BUILD LOG
---------
D:/Users/Luis/.cruise/projects/rubygems-mswin32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"  
(in D:/Users/Luis/.cruise/projects/rubygems-mswin32/work)
[CruiseControl] Invoking Rake task "test"
Loaded suite D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader
Started
...............................................................D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/openssl/x509.rb:24: warning: instance variable @config not initialized
D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/openssl/x509.rb:24: warning: instance variable @config not initialized
D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/openssl/x509.rb:24: warning: instance variable @config not initialized
................................D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/rdoc/parsers/parse_c.rb:204: warning: method redefined; discarding old progress
...WARNING:  Unable to use symlinks on Windows, installing wrapper
.WARNING:  Unable to use symlinks on Windows, installing wrapper
.WARNING:  Unable to use symlinks on Windows, installing wrapper
..............................................................WARNING:  Unable to use symlinks on Windows, installing wrapper
.............WARNING:  Unable to use symlinks on Windows, installing wrapper
..WARNING:  Unable to use symlinks on Windows, installing wrapper
................................................................................................................................................................................................................................................................................................................................................................................................................................
Finished in 59.188 seconds.

593 tests, 1771 assertions, 0 failures, 0 errors


PROJECT SETTINGS
----------------
# Project-specific configuration for CruiseControl.rb

Project.configure do |project|
  
  project.email_notifier.emails = ['rubygems-developers at rubyforge.org', 'rubyinstaller-devel at rubyforge.org']
  
  project.scheduler.polling_interval = 5.minutes

end


From luislavena at gmail.com  Tue Apr 15 19:37:57 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Tue, 15 Apr 2008 20:37:57 -0300
Subject: [Rubyinstaller-devel] Ruby 1.8.7-preview1 binaries uploaded
Message-ID: <71166b3b0804151637h2fe51cadv4130013321c6f0b1@mail.gmail.com>

Hello Guys,

The readline issues are still forcing me do it this manually, even
when I have Integration builder in place [1]

Anyway, I've updated the 7z files in the dump location [2], still
waiting to have more spare time to work on the MSI scripts.

I think this coming weekend will be a great time to work on those,
even I still have a series of [BUG] and segfaults to track down
(wxRuby).

Also, when worked getting DataMapper run on Windows, found that the
current version of OpenSSL we ship with Installer3 is older and made
PostgreSQL fail.

So, after all, I think we will require manually build the external
libraries that Ruby depends on.

Thoughts?

[1] http://builder.mmediasys.com:8333/
[2] http://dump.mmediasys.com/installer3/

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Wed Apr 16 00:07:55 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Tue, 15 Apr 2008 23:07:55 -0500
Subject: [Rubyinstaller-devel] Ruby 1.8.7-preview1 binaries uploaded
In-Reply-To: <71166b3b0804151637h2fe51cadv4130013321c6f0b1@mail.gmail.com>
References: <71166b3b0804151637h2fe51cadv4130013321c6f0b1@mail.gmail.com>
Message-ID: <e9ece6840804152107l63d4b2b5hb07c32c31ff0dba0@mail.gmail.com>

>  So, after all, I think we will require manually build the external
>  libraries that Ruby depends on.
>
>  Thoughts?
>
>  [1] http://builder.mmediasys.com:8333/
>  [2] http://dump.mmediasys.com/installer3/
>

I'll try to work on compiling OpenSSL some more this weekend.

Also, I get a "Connection timed out" error when I go to
http://builder.mmediasys.com:8333/

Thanks,

Gordon

From luislavena at gmail.com  Wed Apr 16 00:13:07 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Wed, 16 Apr 2008 01:13:07 -0300
Subject: [Rubyinstaller-devel] Ruby 1.8.7-preview1 binaries uploaded
In-Reply-To: <e9ece6840804152107l63d4b2b5hb07c32c31ff0dba0@mail.gmail.com>
References: <71166b3b0804151637h2fe51cadv4130013321c6f0b1@mail.gmail.com>
	<e9ece6840804152107l63d4b2b5hb07c32c31ff0dba0@mail.gmail.com>
Message-ID: <71166b3b0804152113y18b3dfa0sbd0dfb152eafb682@mail.gmail.com>

On Wed, Apr 16, 2008 at 1:07 AM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> >  So, after all, I think we will require manually build the external
>  >  libraries that Ruby depends on.
>  >
>  >  Thoughts?
>  >
>  >  [1] http://builder.mmediasys.com:8333/
>  >  [2] http://dump.mmediasys.com/installer3/
>  >
>
>  I'll try to work on compiling OpenSSL some more this weekend.
>
>  Also, I get a "Connection timed out" error when I go to
>
> http://builder.mmediasys.com:8333/
>

Yeah, forgot to put that system in the UPS, and turned down the power
when leaving the office.

Sorry for that :-(

I think I need to re-work the whole rake recipes and compilation
procedures, to allow building theses dependencies with the rake files.

Maybe abuse of Ruby power and do my own DSL that generates the require
rake task for it? After all, done that several times for projects like
mongrel_service and it's dependencies.

The only ones that can turn into a nightmare are gettext and libiconv,
they need build two times since they depend each other... (what kind
of sadistic programmer designed that?)

I'm installing Vista at the office to see what is happening with it :-P

Regards,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From hramrach at centrum.cz  Fri Apr 18 11:16:13 2008
From: hramrach at centrum.cz (Michal Suchanek)
Date: Fri, 18 Apr 2008 17:16:13 +0200
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
	<71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
	<1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>
	<71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>
Message-ID: <a5d587fb0804180816g6b9040e1i633150d040f199d9@mail.gmail.com>

On 12/04/2008, Luis Lavena <luislavena at gmail.com> wrote:
> On Fri, Apr 11, 2008 at 7:01 PM, David Leal <dgleal at gmail.com> wrote:
>  > On 4/11/08, Luis Lavena <luislavena at gmail.com> wrote:
>  >
>
> >  >  Just run configure step by its own and peek what ended in there (of
>  >  >  course, on a clean sandbox and after extract and prepare).
>  >
>  >  I sent you an email with all the files inside ruby_build. I hope it helps.
>  >
>
>
> Yes, it helped and I think I spotted where the problem lies.
>
>  Based on the files, it appears you updated the autoconf package (from
>  2.59 to 2.61). That is correct?
>
>  "generated by GNU Autoconf 2.61."
>
>  that is part of the config.log file.
>
>  I sticked to 2.59 since the new set of files generated by 2.61 didn't
>  worked properly.
>
>  config.log (2.59):
>
>  LIBPATHFLAG=' -L%s'
>
>  your config.log (2.61):
>
>  LIBPATHFLAG=' -L"%s"'
>
>  Somehow, dunno from where, the newer version of autoconf is being
>  called. The only one I see from your path that could be doing that is
>  Git, even more precise if you have a .bashrc file in your HOME folder
>  that is altering the $PATH...
>
>  Try running msys.bat from sandbox/msys and 'which autoconf' and it
>  should point to /bin/autoconf (7,663 bytes) and check the version
>  information around line 130 of that file.
>

Did you find a solution?

I patched configure.in on OS X to remove the quotes. There are similar
linking problems as on Vista so I hope I would get rid of them this
way.

Looks like the shipped configure has LIBPATHFLAG with the quotes, though.

Thanks

Michal
-------------- next part --------------
A non-text attachment was scrubbed...
Name: configure.diff
Type: application/octet-stream
Size: 272 bytes
Desc: not available
Url : http://rubyforge.org/pipermail/rubyinstaller-devel/attachments/20080418/f09ceb83/attachment.obj 

From gthiesfeld at gmail.com  Fri Apr 18 11:57:38 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Fri, 18 Apr 2008 10:57:38 -0500
Subject: [Rubyinstaller-devel] make.cmd?
Message-ID: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>

So, I was looking at the way msysGit wraps the git command in a batch
file, and I thought we could do the same thing for compiling
extensions in gems.
Something like this in \path_to_ruby\bin:

  REM make.cmd
  @echo off
  setlocal
  set path=%path%;c:\ruby\repo\oci\sandbox\mingw
  C:\ruby\repo\oci\sandbox\msys\bin\bash.exe  --login -i -c "make %*"
  exit /B %ErrorLevel%

This way we wouldn't need to worry about adding mingw to the path,
except when it's needed (ie. when make is called). We may need to add
msys to the path as well, I'm not sure.  I can use this method to `gem
install mongrel`.  It compiles native extensions and all that.  Of
course when I try to run mongrel it fails ;), but I think that's a
separate issue.

What do you think?  Am I oversimplifying the problem again?

My setup:

  Windows XP, SP2
  ruby 1.8.6 (2008-03-03 patchlevel 114) [i386-mingw32]
  rubygems 1.1.1

Thanks,

Gordon

From luislavena at gmail.com  Fri Apr 18 19:31:05 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Fri, 18 Apr 2008 20:31:05 -0300
Subject: [Rubyinstaller-devel] Ruby MinGW on Vista
In-Reply-To: <a5d587fb0804180816g6b9040e1i633150d040f199d9@mail.gmail.com>
References: <1b86cb980804100322t2ae23245s54a4315303fb0139@mail.gmail.com>
	<1b86cb980804100333u5408567cn553c3b53d915c1b1@mail.gmail.com>
	<71166b3b0804100513o6f60d52aicd89dfc174f23fc6@mail.gmail.com>
	<1b86cb980804100835i375ecc7fq898caabe06012d07@mail.gmail.com>
	<71166b3b0804101711m7d986144ga68cccc0ee30b0b5@mail.gmail.com>
	<1b86cb980804110249o3dfd9880vcf06a87634a83fec@mail.gmail.com>
	<71166b3b0804110706r7fe9600ardf898d3e27b21c2f@mail.gmail.com>
	<1b86cb980804111501h4c5286d0r42e845111795bafd@mail.gmail.com>
	<71166b3b0804111832i18732567ya4e768126481e7ec@mail.gmail.com>
	<a5d587fb0804180816g6b9040e1i633150d040f199d9@mail.gmail.com>
Message-ID: <71166b3b0804181631j3d4306c7j78d726e639ee8f41@mail.gmail.com>

On Fri, Apr 18, 2008 at 12:16 PM, Michal Suchanek <hramrach at centrum.cz> wrote:
> On 12/04/2008, Luis Lavena <luislavena at gmail.com> wrote:
>  > On Fri, Apr 11, 2008 at 7:01 PM, David Leal <dgleal at gmail.com> wrote:
>  >  > On 4/11/08, Luis Lavena <luislavena at gmail.com> wrote:
>  >  >
>  >
>  > >  >  Just run configure step by its own and peek what ended in there (of
>  >  >  >  course, on a clean sandbox and after extract and prepare).
>  >  >
>  >  >  I sent you an email with all the files inside ruby_build. I hope it helps.
>  >  >
>  >
>  >
>  > Yes, it helped and I think I spotted where the problem lies.
>  >
>  >  Based on the files, it appears you updated the autoconf package (from
>  >  2.59 to 2.61). That is correct?
>  >
>  >  "generated by GNU Autoconf 2.61."
>  >
>  >  that is part of the config.log file.
>  >
>  >  I sticked to 2.59 since the new set of files generated by 2.61 didn't
>  >  worked properly.
>  >
>  >  config.log (2.59):
>  >
>  >  LIBPATHFLAG=' -L%s'
>  >
>  >  your config.log (2.61):
>  >
>  >  LIBPATHFLAG=' -L"%s"'
>  >
>  >  Somehow, dunno from where, the newer version of autoconf is being
>  >  called. The only one I see from your path that could be doing that is
>  >  Git, even more precise if you have a .bashrc file in your HOME folder
>  >  that is altering the $PATH...
>  >
>  >  Try running msys.bat from sandbox/msys and 'which autoconf' and it
>  >  should point to /bin/autoconf (7,663 bytes) and check the version
>  >  information around line 130 of that file.
>  >
>
>  Did you find a solution?
>

After all it seems is not related to quoted paths, but a GCC problem on Vista:

http://sourceforge.net/mailarchive/message.php?msg_name=4808B254.9000403%40tdragon.net

http://sourceforge.net/mailarchive/message.php?msg_name=4808B667.584B9FC9%40dessent.net

I'm trying to report these problems back to MinGW developers since we
(one-click project) don't have enough resources to run GCC compilation
besides ruby one :-D

>  I patched configure.in on OS X to remove the quotes. There are similar
>  linking problems as on Vista so I hope I would get rid of them this
>  way.
>

Done that, didn't work, it seems that even small things like a gcc
hello world is broken on Vista.

Will keep working on getting things properly for it, since without the
One-Click Developer Kit will be useless.

>  Looks like the shipped configure has LIBPATHFLAG with the quotes, though.
>

Is because they used autconf 2.61, older versions (2.58 AFAIK) didn't do that.

>  Thanks
>

Thanks to you for your interest,

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From luislavena at gmail.com  Fri Apr 18 21:15:00 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Fri, 18 Apr 2008 22:15:00 -0300
Subject: [Rubyinstaller-devel] make.cmd?
In-Reply-To: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>
References: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>
Message-ID: <71166b3b0804181815k39ba53eevc09ff651d1a3470e@mail.gmail.com>

On Fri, Apr 18, 2008 at 12:57 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> So, I was looking at the way msysGit wraps the git command in a batch
>  file, and I thought we could do the same thing for compiling
>  extensions in gems.
>  Something like this in \path_to_ruby\bin:
>
>   REM make.cmd
>   @echo off
>   setlocal
>   set path=%path%;c:\ruby\repo\oci\sandbox\mingw
>   C:\ruby\repo\oci\sandbox\msys\bin\bash.exe  --login -i -c "make %*"
>   exit /B %ErrorLevel%
>
>  This way we wouldn't need to worry about adding mingw to the path,
>  except when it's needed (ie. when make is called). We may need to add
>  msys to the path as well, I'm not sure.  I can use this method to `gem
>  install mongrel`.  It compiles native extensions and all that.  Of
>  course when I try to run mongrel it fails ;), but I think that's a
>  separate issue.
>

There is no need to add mingw or msys to the path, since when bash
start it parses profile and setup $PATH to include both, mingw and
msys tools.

Anyway, we will need wrap 'gcc' too, since is something extconf uses
to try definitions that create the makefile.

The only problem is that ruby doesn't like it, and I had problems in
the past due that, and was ruby exec code.

Let me explain:

When you issue 'make' in the command line, the cmd interpreter
(cmd.exe) will try to find an executable of make, based on 'make' +
one of the extensions in PATHEXT environment variable, in the order
they are listed:

make.com, make.exe, make.bat, make.cmd, etc.

The problem is ruby exec doesn't work like that. if you supply a
command without the extension, it will look for executables (com and
exe), but you must explicitly indicate .bat or .cmd when executing
batch files.

try creating a make.bat that just echo something and do:

ruby -e "system('make --version')"

Without mingw/msys in the path, that will fail until you supply
.bat/.cmd to the system command.

So, that will require us patch rubygems and replace conditions for
RUBY_PLATFORM and issue 'make.bat/cmd' instead of plain 'make'. Take
in consideration that right now there are conditions for mswin (nmake)
and others platforms, plain make.

>  What do you think?  Am I oversimplifying the problem again?
>

I like the idea and you aren't oversimplifying it, I think Ruby is
broken in that aspect.

Maybe we can fix that damn thing in ruby C code and workaround that?

win32/win32.c:989 (CreateChild function), call dln_find_exe
dlc.c:1671 (dln_find_exe function), call dln_find_1
dlc.c:1796 (dln_find_1 function), call eaccess
file.c:897 (eaccess function), call access API with 'path' and mode =
X_OK (io.h)

Or maybe I ended tracing it to the wrong place :-P

>  Thanks,

Thanks to you for your interest.

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Sat Apr 19 10:51:10 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Sat, 19 Apr 2008 09:51:10 -0500
Subject: [Rubyinstaller-devel] make.cmd?
In-Reply-To: <71166b3b0804181815k39ba53eevc09ff651d1a3470e@mail.gmail.com>
References: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>
	<71166b3b0804181815k39ba53eevc09ff651d1a3470e@mail.gmail.com>
Message-ID: <e9ece6840804190751n1957715bpac6274812955fc43@mail.gmail.com>

On Fri, Apr 18, 2008 at 8:15 PM, Luis Lavena <luislavena at gmail.com> wrote:
>
>  There is no need to add mingw or msys to the path, since when bash
>  start it parses profile and setup $PATH to include both, mingw and
>  msys tools.
>
>  Anyway, we will need wrap 'gcc' too, since is something extconf uses
>  to try definitions that create the makefile.
>
>  The only problem is that ruby doesn't like it, and I had problems in
>  the past due that, and was ruby exec code.
>
>  Let me explain:
>
>  When you issue 'make' in the command line, the cmd interpreter
>  (cmd.exe) will try to find an executable of make, based on 'make' +
>  one of the extensions in PATHEXT environment variable, in the order
>  they are listed:
>
>  make.com, make.exe, make.bat, make.cmd, etc.
>
>  The problem is ruby exec doesn't work like that. if you supply a
>  command without the extension, it will look for executables (com and
>  exe), but you must explicitly indicate .bat or .cmd when executing
>  batch files.
>
>  try creating a make.bat that just echo something and do:
>
>  ruby -e "system('make --version')"

:-P

C:\ruby\ruby_mingw>which make.exe
which: no make.exe in
(c:\wix;c:\windows\system32;c:\windows\system;C:\Program
Files\Subversion\bin;C:\Program Files\Bazaar;C:\Program
Files\Git\cmd;C:\ruby\ruby_mingw\bin;C:\bin;C:\WINDOWS\Microsoft.NET\Framework\v2.0.50727;C:\Program
Files\Putty;c:\bin\console2)

C:\ruby\ruby_mingw>which make.cmd
C:\ruby\ruby_mingw\bin\make.cmd

C:\ruby\ruby_mingw>irb
>> `make --version`
=> "GNU Make version 3.79.1, by Richard Stallman and Roland McGrath.\nBuilt for
i686-pc-msys\nCopyright (C) 1988, 89, 90, 91, 92, 93, 94, 95, 96, 97,
98, 99, 2000\n\tFree Software Foundation, Inc.\nThis is free software;
see the source for
copying conditions.\nThere is NO warranty; not even for
MERCHANTABILITY or FITNESS FOR A\nPARTICULAR PURPOSE.\n\nReport bugs
to <bug-make at gnu.org>.\n\n"
>>

OK, I've seen this problem before too, with Rakefiles that shell out
to use 'gem' and things like that, and I've always made the same
assumption that Ruby didn't process bat or cmd file. But what was
bugging me after I read your email was that my make.cmd worked, and
there is no other make in the path.  First of all, it does seem to run
bat and cmd files, but they have to be in the path not in the same
folder you're in.  This may be a 'feature' ;-), similar to the way
*nix won't let you run shell commands in your current folder without
adding ./ in front of them. Make a simple batch file, like this:

REM test.cmd
@echo off
date /t

Stick it in ruby\bin, or somewhere else in the path and then try to
run it.  It works for me.

The problem with the gem command and commands created by RubyGems, is
that Ruby will find the ruby file at the command first and try to
execute it, and the shell doesn't know how to run it (Windows doesn't
do shebang lines). So, if you move the ruby file with no extension out
of the bin folder, and adjust the batch file to point to it, it will
work.  At least it works on my machine ;-).

>  I like the idea and you aren't oversimplifying it, I think Ruby is
>  broken in that aspect.
>
>  Maybe we can fix that damn thing in ruby C code and workaround that?
>
>  win32/win32.c:989 (CreateChild function), call dln_find_exe
>  dlc.c:1671 (dln_find_exe function), call dln_find_1
>  dlc.c:1796 (dln_find_1 function), call eaccess
>  file.c:897 (eaccess function), call access API with 'path' and mode =
>  X_OK (io.h)
>

Maybe Ruby's not totally broken here, just undocumented?  Take a look
and let me know what you think.

--Gordon

From luislavena at gmail.com  Sat Apr 19 19:23:04 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sat, 19 Apr 2008 20:23:04 -0300
Subject: [Rubyinstaller-devel] make.cmd?
In-Reply-To: <e9ece6840804190751n1957715bpac6274812955fc43@mail.gmail.com>
References: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>
	<71166b3b0804181815k39ba53eevc09ff651d1a3470e@mail.gmail.com>
	<e9ece6840804190751n1957715bpac6274812955fc43@mail.gmail.com>
Message-ID: <71166b3b0804191623t654d4c1pa7011e8ef594d3c4@mail.gmail.com>

On Sat, Apr 19, 2008 at 11:51 AM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> On Fri, Apr 18, 2008 at 8:15 PM, Luis Lavena <luislavena at gmail.com> wrote:
>  >
>  >  There is no need to add mingw or msys to the path, since when bash
>  >  start it parses profile and setup $PATH to include both, mingw and
>  >  msys tools.
>  >
>  >  Anyway, we will need wrap 'gcc' too, since is something extconf uses
>  >  to try definitions that create the makefile.
>  >
>  >  The only problem is that ruby doesn't like it, and I had problems in
>  >  the past due that, and was ruby exec code.
>  >
>  >  Let me explain:
>  >
>  >  When you issue 'make' in the command line, the cmd interpreter
>  >  (cmd.exe) will try to find an executable of make, based on 'make' +
>  >  one of the extensions in PATHEXT environment variable, in the order
>  >  they are listed:
>  >
>  >  make.com, make.exe, make.bat, make.cmd, etc.
>  >
>  >  The problem is ruby exec doesn't work like that. if you supply a
>  >  command without the extension, it will look for executables (com and
>  >  exe), but you must explicitly indicate .bat or .cmd when executing
>  >  batch files.
>  >
>  >  try creating a make.bat that just echo something and do:
>  >
>  >  ruby -e "system('make --version')"
>
>  :-P
>
>  C:\ruby\ruby_mingw>which make.exe
>  which: no make.exe in
>  (c:\wix;c:\windows\system32;c:\windows\system;C:\Program
>  Files\Subversion\bin;C:\Program Files\Bazaar;C:\Program
>  Files\Git\cmd;C:\ruby\ruby_mingw\bin;C:\bin;C:\WINDOWS\Microsoft.NET\Framework\v2.0.50727;C:\Program
>  Files\Putty;c:\bin\console2)
>
>  C:\ruby\ruby_mingw>which make.cmd
>  C:\ruby\ruby_mingw\bin\make.cmd
>
>  C:\ruby\ruby_mingw>irb
>  >> `make --version`
>  => "GNU Make version 3.79.1, by Richard Stallman and Roland McGrath.\nBuilt for
>  i686-pc-msys\nCopyright (C) 1988, 89, 90, 91, 92, 93, 94, 95, 96, 97,
>  98, 99, 2000\n\tFree Software Foundation, Inc.\nThis is free software;
>  see the source for
>  copying conditions.\nThere is NO warranty; not even for
>  MERCHANTABILITY or FITNESS FOR A\nPARTICULAR PURPOSE.\n\nReport bugs
>  to <bug-make at gnu.org>.\n\n"
>  >>
>

What The Hell???

I tried that before send the mail to the list, just to confirm what I
was getting for the past years... and it failed...

Now that I try again, and again, and again, it works???

(make.bat) put into %HOME%/bin (which is in PATH).

@ECHO OFF
ECHO.Simulate Make.
ECHO.parameters to be passed: %*

Now, using CMD.exe, in HOME or any other drive letter:

C:\>make --version
Simulate Make.
parameters to be passed: --version

(make.bat is in D:\Users\Luis\bin\make.bat), don't have msys tools in
the path for this test.

Ok, now Ruby:

C:\>ruby -e "puts Dir.pwd; system('make --version')"
C:/
Simulate Make.
parameters to be passed: --version

The damn thing worked!

I always got mixed results from system/backticks/exec ...

But!, now I know how to make it fail...

1) Create a empty 'make' file and put in a path that is listed
*before* the one that holds your make.bat (system32 is a good place).

D:\Users\Luis>ruby -e "puts Dir.pwd; system('make --version')"
D:/Users/Luis

No output from system call... because it found the first occurrence of
make (extension less).

2) Now move the empty make file into a directory of PATH that happens
*after* the position make.bat is located (example, I moved it to
c;\program files\putty)

C:\>ruby -e "puts Dir.pwd; system('make --version')"
C:/
Simulate Make.
parameters to be passed: --version

Bingo!

so it seems is not just looking for executable files, but any file
that matches the search criteria (make.*), the first found, the first
served, no matter if is an executable or not.

>  OK, I've seen this problem before too, with Rakefiles that shell out
>  to use 'gem' and things like that, and I've always made the same
>  assumption that Ruby didn't process bat or cmd file. But what was
>  bugging me after I read your email was that my make.cmd worked, and
>  there is no other make in the path.  First of all, it does seem to run
>  bat and cmd files, but they have to be in the path not in the same
>  folder you're in.  This may be a 'feature' ;-), similar to the way
>  *nix won't let you run shell commands in your current folder without
>  adding ./ in front of them. Make a simple batch file, like this:
>

Yes, a true feature :-P

>  REM test.cmd
>  @echo off
>  date /t
>
>  Stick it in ruby\bin, or somewhere else in the path and then try to
>  run it.  It works for me.
>
>  The problem with the gem command and commands created by RubyGems, is
>  that Ruby will find the ruby file at the command first and try to
>  execute it, and the shell doesn't know how to run it (Windows doesn't
>  do shebang lines). So, if you move the ruby file with no extension out
>  of the bin folder, and adjust the batch file to point to it, it will
>  work.  At least it works on my machine ;-).

The issue with 'gem' is that rubygems generates to files: gem and
gem.bat which is a stub to execute the true gem ruby script.

If we move from script+stub to scripts similar of what ruby ships
(take a look at irb.bat) can workaround the issues with
"cross-compatibility" raised of being using
system/exec/backticks/whatever

>  >  I like the idea and you aren't oversimplifying it, I think Ruby is
>  >  broken in that aspect.
>  >
>  >  Maybe we can fix that damn thing in ruby C code and workaround that?
>  >
>  >  win32/win32.c:989 (CreateChild function), call dln_find_exe
>  >  dlc.c:1671 (dln_find_exe function), call dln_find_1
>  >  dlc.c:1796 (dln_find_1 function), call eaccess
>  >  file.c:897 (eaccess function), call access API with 'path' and mode =
>  >  X_OK (io.h)
>  >
>
>  Maybe Ruby's not totally broken here, just undocumented?  Take a look
>  and let me know what you think.
>

Please let me congrats you Gordon, you just hit one of the big issues
that blocks true cross-platform implementation.

Also, these injected stubs/scripts can workaround the pipes issues for
scripts using .rb and being tried to executed standalone (this is an
old issue of ruby and how cmd creates the process and pass the
information to the child process).

With the fix for Vista the MinGW guys are working, maybe then we could
add a 'gcc' and 'make' stubs that transparently load the mingw/msys
environment for you :-D

So the Developer Kit only will be the packages we already have (a bit
of cleanup) and let it drop these stubs into ruby/bin :-)

I love the idea! Thank you for taking the time exposing your point, It
was really enlightening

Regards,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Sun Apr 20 11:15:17 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Sun, 20 Apr 2008 10:15:17 -0500
Subject: [Rubyinstaller-devel] Installer3 rake compile problem
Message-ID: <e9ece6840804200815x1c9f04b4rba615badefd69bd3@mail.gmail.com>

I've been having a problem with the compile step in installer3

compiling openssl
C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:197:in `initialize':
Permission denied - mkmftmp1.log (Errno::EACCES)
        from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:197:in `open'
        from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:197:in `open'
        from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:228:in `postpone'
        from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:201:in `open'
        from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:224:in `postpone'
        from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:591:in
`checking_for'
        from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:695:in
`have_func'
        from ../../../ruby_1_8/ext/openssl/extconf.rb:73
        from ../ruby_1_8/ext/extmk.rb:168:in `load'
        from ../ruby_1_8/ext/extmk.rb:168:in `extmake'
        from ../ruby_1_8/ext/extmk.rb:441
        from ../ruby_1_8/ext/extmk.rb:440:in `each'
        from ../ruby_1_8/ext/extmk.rb:440
make: *** [all] Error 1
rake aborted!
Command failed with status (2): ["C:/ruby/repo/oci_187/sandbox/msys/bin/bas...]

(See full trace by running task with --trace)

If I run it again, it might get past openssl, but fail on another
extension.  So I just keep running it until it finishes.  If I comment
out line 234 of ruby_1_8\lib\mkmf.rb it runs all the way through, but
this seems suboptimal.  Is anyone else seeing this?

Thanks,

Gordon

From luislavena at gmail.com  Sun Apr 20 13:21:50 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sun, 20 Apr 2008 14:21:50 -0300
Subject: [Rubyinstaller-devel] Installer3 rake compile problem
In-Reply-To: <e9ece6840804200815x1c9f04b4rba615badefd69bd3@mail.gmail.com>
References: <e9ece6840804200815x1c9f04b4rba615badefd69bd3@mail.gmail.com>
Message-ID: <71166b3b0804201021w59c56ca3nbcb2364cfa843bae@mail.gmail.com>

On Sun, Apr 20, 2008 at 12:15 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> I've been having a problem with the compile step in installer3
>
>  compiling openssl
>  C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:197:in `initialize':
>  Permission denied - mkmftmp1.log (Errno::EACCES)
>         from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:197:in `open'
>         from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:197:in `open'
>         from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:228:in `postpone'
>         from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:201:in `open'
>         from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:224:in `postpone'
>         from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:591:in
>  `checking_for'
>         from C:/ruby/repo/oci_187/sandbox/ruby_1_8/lib/mkmf.rb:695:in
>  `have_func'
>         from ../../../ruby_1_8/ext/openssl/extconf.rb:73
>         from ../ruby_1_8/ext/extmk.rb:168:in `load'
>         from ../ruby_1_8/ext/extmk.rb:168:in `extmake'
>         from ../ruby_1_8/ext/extmk.rb:441
>         from ../ruby_1_8/ext/extmk.rb:440:in `each'
>         from ../ruby_1_8/ext/extmk.rb:440
>  make: *** [all] Error 1
>  rake aborted!
>  Command failed with status (2): ["C:/ruby/repo/oci_187/sandbox/msys/bin/bas...]
>
>  (See full trace by running task with --trace)
>
>  If I run it again, it might get past openssl, but fail on another
>  extension.  So I just keep running it until it finishes.  If I comment
>  out line 234 of ruby_1_8\lib\mkmf.rb it runs all the way through, but
>  this seems suboptimal.  Is anyone else seeing this?
>

Mmm, looks like the OS is blocking the .log file when compiling the extension.

Can you cleanup ruby_build/ext/openssl (remove folder) and also the
.ext/i386-mingw32/openssl.so ?

Also, check for a mkmf*.log in your %TEMP%.

Sometimes some antivirus software block these files.

HTH,
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From luislavena at gmail.com  Sun Apr 20 18:01:06 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sun, 20 Apr 2008 19:01:06 -0300
Subject: [Rubyinstaller-devel] Reverting from 1.8.7
Message-ID: <71166b3b0804201501o1180176cn688a079cb2156621@mail.gmail.com>

Hello Folks,

I was so dumb perform a release and also some announcement of the
preview of 1.8.7 based on MinGW. I neglected the importance of
backward compatibility and followed the joy of new release, which
introduced several bugs and incompatibilities, as reported in
ruby-core [1]

I'm aware that svn version after PATCHLEVEL 114 introduced some
changes that made things like wxRuby crash. So I'm releasing right now
a test version based on packaged ruby-1.8.6-p114 tarball.

Also, I'm working right now on make rubyspecs (part of rubinius
innitiative) work on Windows and ensure proper releases at least for
the core of ruby.

In the mean time, I also have added a text file alongside the 7z
package which contains the output of 'make check' for each release.

Hopefully someone will find this useful :-)

The place were these files can be downloaded:

http://dump.mmediasys.com/installer3

The only known problem with this release is the readline/irb CPU
eating bug, but that cannot be fixed right now without monkey
patching.

I'm looking forward ruby maintainers workaround these issues, but for
the time being, I only have hope :-)

Regards and good weekend everybody.

[1] http://www.ruby-forum.com/topic/150251
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Sun Apr 20 18:14:38 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Sun, 20 Apr 2008 17:14:38 -0500
Subject: [Rubyinstaller-devel] Installer3 rake compile problem
In-Reply-To: <71166b3b0804201021w59c56ca3nbcb2364cfa843bae@mail.gmail.com>
References: <e9ece6840804200815x1c9f04b4rba615badefd69bd3@mail.gmail.com>
	<71166b3b0804201021w59c56ca3nbcb2364cfa843bae@mail.gmail.com>
Message-ID: <e9ece6840804201514x13024951ga637e3daaede0a9a@mail.gmail.com>

On Sun, Apr 20, 2008 at 12:21 PM, Luis Lavena <luislavena at gmail.com> wrote:
>
>
>  Mmm, looks like the OS is blocking the .log file when compiling the extension.
>
>  Can you cleanup ruby_build/ext/openssl (remove folder) and also the
>  .ext/i386-mingw32/openssl.so ?
>
>  Also, check for a mkmf*.log in your %TEMP%.
>
>  Sometimes some antivirus software block these files.

It was the antivirus software.  Thanks, Luis.

Gordon

From gthiesfeld at gmail.com  Mon Apr 21 22:03:45 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Mon, 21 Apr 2008 21:03:45 -0500
Subject: [Rubyinstaller-devel] make.cmd?
In-Reply-To: <71166b3b0804191623t654d4c1pa7011e8ef594d3c4@mail.gmail.com>
References: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>
	<71166b3b0804181815k39ba53eevc09ff651d1a3470e@mail.gmail.com>
	<e9ece6840804190751n1957715bpac6274812955fc43@mail.gmail.com>
	<71166b3b0804191623t654d4c1pa7011e8ef594d3c4@mail.gmail.com>
Message-ID: <e9ece6840804211903l44bed8dbxb23c488e0d4b4a56@mail.gmail.com>

>  The issue with 'gem' is that rubygems generates to files: gem and
>  gem.bat which is a stub to execute the true gem ruby script.
>
>  If we move from script+stub to scripts similar of what ruby ships
>  (take a look at irb.bat) can workaround the issues with
>  "cross-compatibility" raised of being using
>  system/exec/backticks/whatever
>

Ok, I've stubbed make, gcc, and sh, and rubygems tests pass.  I'm
working on a patch for rubygems bin scripts as well, but I can't find
a simple way to do it without breaking a lot of tests.  I'll keep at
it.  I did discover that the exit /b %errorlevel% doesn't do the right
thing when called in Ruby, so I'm just making the cmd files so that
the last line is the one we care about..

This is my rake.cmd.  It seems to be working fine.  Let me know if you
see any problems with it.

@ECHO OFF
goto endofruby
#!c:/ruby/ruby_mingw/bin/ruby.exe
#
# This file was generated by RubyGems.
#
# The application 'rake' is installed as part of a gem, and
# this file is here to facilitate running it.
#

require 'rubygems'

version = ">= 0"

if ARGV.first =~ /^_(.*)_$/ and Gem::Version.correct? $1 then
  version = $1
  ARGV.shift
end

gem 'rake', version
load 'rake'
__END__
:endofruby
"%~d0%~p0ruby" -x "%~f0" %*


>  Please let me congrats you Gordon, you just hit one of the big issues
>  that blocks true cross-platform implementation.
>
>  Also, these injected stubs/scripts can workaround the pipes issues for
>  scripts using .rb and being tried to executed standalone (this is an
>  old issue of ruby and how cmd creates the process and pass the
>  information to the child process).

Yeah, I've been bitten by this one too :-).  The only problem with
batch files is that it makes ctrl-c cause the extremely annoying
"Terminate batch job (Y/N)?".  I did find a way to modify cmd.exe so
that it doesn't do this, but it would be nice if there was a cleaner
solution.

From luislavena at gmail.com  Mon Apr 21 22:55:09 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Mon, 21 Apr 2008 23:55:09 -0300
Subject: [Rubyinstaller-devel] make.cmd?
In-Reply-To: <e9ece6840804211903l44bed8dbxb23c488e0d4b4a56@mail.gmail.com>
References: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>
	<71166b3b0804181815k39ba53eevc09ff651d1a3470e@mail.gmail.com>
	<e9ece6840804190751n1957715bpac6274812955fc43@mail.gmail.com>
	<71166b3b0804191623t654d4c1pa7011e8ef594d3c4@mail.gmail.com>
	<e9ece6840804211903l44bed8dbxb23c488e0d4b4a56@mail.gmail.com>
Message-ID: <71166b3b0804211955k2e75c9ebya808e8d51ae8874e@mail.gmail.com>

On Mon, Apr 21, 2008 at 11:03 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> >  The issue with 'gem' is that rubygems generates to files: gem and
>  >  gem.bat which is a stub to execute the true gem ruby script.
>  >
>  >  If we move from script+stub to scripts similar of what ruby ships
>  >  (take a look at irb.bat) can workaround the issues with
>  >  "cross-compatibility" raised of being using
>  >  system/exec/backticks/whatever
>  >
>
>  Ok, I've stubbed make, gcc, and sh, and rubygems tests pass.  I'm
>  working on a patch for rubygems bin scripts as well, but I can't find
>  a simple way to do it without breaking a lot of tests.  I'll keep at
>  it.  I did discover that the exit /b %errorlevel% doesn't do the right
>  thing when called in Ruby, so I'm just making the cmd files so that
>  the last line is the one we care about..
>
>  This is my rake.cmd.  It seems to be working fine.  Let me know if you
>  see any problems with it.
>
>  @ECHO OFF
>  goto endofruby
>  #!c:/ruby/ruby_mingw/bin/ruby.exe
>  #
>  # This file was generated by RubyGems.
>  #
>  # The application 'rake' is installed as part of a gem, and
>  # this file is here to facilitate running it.
>  #
>
>  require 'rubygems'
>
>  version = ">= 0"
>
>  if ARGV.first =~ /^_(.*)_$/ and Gem::Version.correct? $1 then
>   version = $1
>   ARGV.shift
>  end
>
>  gem 'rake', version
>  load 'rake'
>  __END__
>  :endofruby
>  "%~d0%~p0ruby" -x "%~f0" %*
>

I was one of the responsible of change the .cmd files to .bat in
latest rubygems releases.

Why? because .bat take precedence instead of .cmd, also because, even
we don't *support* win9x, at least it will work out of the box.

I'll change to uppercase all the references to windows/batch commands (GOTO).

I'll take a look at rubygems (again) to see if we can get rid of the
separate batch/script files for the next release, or at least, our
implementation, since I'm thinking seriously not ship 1.8.7...
(there is a heated discussion about that).

An alternative will be have a series of patches that can be
rolled/applied to current 1.8.6-p114 and fix particular things in ruby
source or rubygems until an official release ship them.

As a RubyGems contributor I can slip these changes in, but we will
require wait for the next release to make them available.

>  >  Please let me congrats you Gordon, you just hit one of the big issues
>  >  that blocks true cross-platform implementation.
>  >
>  >  Also, these injected stubs/scripts can workaround the pipes issues for
>  >  scripts using .rb and being tried to executed standalone (this is an
>  >  old issue of ruby and how cmd creates the process and pass the
>  >  information to the child process).
>
>  Yeah, I've been bitten by this one too :-).  The only problem with
>  batch files is that it makes ctrl-c cause the extremely annoying
>  "Terminate batch job (Y/N)?".  I did find a way to modify cmd.exe so
>  that it doesn't do this, but it would be nice if there was a cleaner
>  solution.
>

There is no easy solution without binary patching cmd.exe, since
in-memory patches will require us alter ruby.exe and we can get caught
by DEP protection :-P

We need to talk about the the future on this, do you feel comfortable
we set a wiki page and start drawing a picture about the goals for the
release?

Thank you for your time Gordon.

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Tue Apr 22 09:03:50 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Tue, 22 Apr 2008 08:03:50 -0500
Subject: [Rubyinstaller-devel] make.cmd?
In-Reply-To: <71166b3b0804211955k2e75c9ebya808e8d51ae8874e@mail.gmail.com>
References: <e9ece6840804180857oeac2e0r9484fb5cab1940b2@mail.gmail.com>
	<71166b3b0804181815k39ba53eevc09ff651d1a3470e@mail.gmail.com>
	<e9ece6840804190751n1957715bpac6274812955fc43@mail.gmail.com>
	<71166b3b0804191623t654d4c1pa7011e8ef594d3c4@mail.gmail.com>
	<e9ece6840804211903l44bed8dbxb23c488e0d4b4a56@mail.gmail.com>
	<71166b3b0804211955k2e75c9ebya808e8d51ae8874e@mail.gmail.com>
Message-ID: <e9ece6840804220603y41d15799x717b41dc196cff29@mail.gmail.com>

On Mon, Apr 21, 2008 at 9:55 PM, Luis Lavena <luislavena at gmail.com> wrote:

>  We need to talk about the the future on this, do you feel comfortable
>  we set a wiki page and start drawing a picture about the goals for the
>  release?
>

This sounds like a great idea.  I've been meaning to do some work on
the wiki for a few weeks now.  I should be on gtalk most of the day,
too.

Thanks,

Gordon

From builder at mmediasys.com  Tue Apr 22 17:04:15 2008
From: builder at mmediasys.com (Integration Builder)
Date: Tue, 22 Apr 2008 18:04:15 -0300
Subject: [Rubyinstaller-devel] [CruiseControl] mongrel-stable_1-2-mingw32
	build 1012 failed
Message-ID: <480e52cf2da78_fdc1462a8819@edicion01.tmail>

The build failed.

  Note: if you set Configuration.dashboard_url in config/site_config.rb, you'd see a link to the build page here.

CHANGES
-------
New revision 1012 detected
Revision 1012 committed by luislavena on 2008-04-18 07:09:44
Wait longer for child process terminate properly (max 20 seconds). 
Imported tests from RubyServices project. (Closes #18).

  M /branches/stable_1-2/projects/mongrel_service/CHANGELOG
  M /branches/stable_1-2/projects/mongrel_service/tests/test_helpers.bas
  M /branches/stable_1-2/projects/mongrel_service/native/console_process.bas
  M /branches/stable_1-2
  M /branches/stable_1-2/projects/mongrel_service/tests/test_console_process.bas
Revision 1011 committed by luislavena on 2008-04-18 07:09:41
Added emulation of slow termination processes.

  M /branches/stable_1-2/projects/mongrel_service/tests/fixtures/mock_process.bas
  M /branches/stable_1-2
Revision 1010 committed by luislavena on 2008-04-18 07:09:38
Include tests projects and child process fixtures.

  M /branches/stable_1-2/projects/mongrel_service/Rakefile
  M /branches/stable_1-2
Revision 1009 committed by luislavena on 2008-04-18 07:09:36
Ported ConsoleProcess tests from RubyServices.

  A /branches/stable_1-2/projects/mongrel_service/tests/fixtures/mock_process.bas
  A /branches/stable_1-2/projects/mongrel_service/tests/test_helpers.bas
  A /branches/stable_1-2/projects/mongrel_service/tests/fixtures
  A /branches/stable_1-2/projects/mongrel_service/tests/test_helpers.bi
  A /branches/stable_1-2/projects/mongrel_service/tests/all_tests.bas
  A /branches/stable_1-2/projects/mongrel_service/tests
  M /branches/stable_1-2
  A /branches/stable_1-2/projects/mongrel_service/tests/test_console_process.bas
Revision 1008 committed by luislavena on 2008-04-18 07:09:33
Updated FreeBASIC compilation tools.

  M /branches/stable_1-2
  M /branches/stable_1-2/projects/mongrel_service/tools/freebasic.rb
Revision 1007 committed by luislavena on 2008-04-17 23:20:50
mongrel_service: Merged changes from trunk.
  M /branches/stable_1-2/projects/mongrel_service/CHANGELOG
  M /branches/stable_1-2/projects/mongrel_service/lib/mongrel_service/init.rb
  M /branches/stable_1-2/projects/mongrel_service/Rakefile
Revision 1006 committed by luislavena on 2008-04-17 18:59:55
mongrel_service: updated ServiceFB library (pistoned).
  M /branches/stable_1-2/projects/mongrel_service/lib/ServiceFB/ServiceFB_Utils.bas
  M /branches/stable_1-2/projects/mongrel_service/lib/ServiceFB
  M /branches/stable_1-2/projects/mongrel_service/lib/ServiceFB/_utils_internals.bi
Revision 1005 committed by evanweaver on 2008-04-17 04:32:11
Document issue with soft restart pid file setting (closes #29).

  M /branches/stable_1-2/bin/mongrel_rails
Revision 1004 committed by evanweaver on 2008-04-17 04:29:44
Apply r1003 to stable.

  M /branches/stable_1-2/bin/mongrel_rails



BUILD LOG
---------
D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"  
D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/bin/ruby extconf.rb
(in D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work)
[CruiseControl] Invoking Rake task "clobber"
Cleaning
- ext/http11/http11.o
- ext/http11/http11_parser.o
- ext/http11/Makefile
- lib/http11.so
- log
- test_stderr.log
- test_stdout.log
[CruiseControl] Invoking Rake task "compile"
checking for main() in -lc... no
creating Makefile
gcc -I. -I. -ID:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/1.8/i386-mingw32 -I.  -g -O2   -c http11.c
gcc -I. -I. -ID:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/1.8/i386-mingw32 -I.  -g -O2   -c http11_parser.c
gcc -shared -s -o http11.so http11.o http11_parser.o -L. -LD:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib -L.  -Wl,--enable-auto-image-base,--enable-auto-import,--export-all   -lmsvcrt-ruby18  -lshell32 -lws2_32  
D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/bin/ruby -Ilib;ext;bin;test "D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader.rb" "test/test_helper.rb" "test/unit/test_cgi_wrapper.rb" "test/unit/test_command.rb" "test/unit/test_conditional.rb" "test/unit/test_configurator.rb" "test/unit/test_debug.rb" "test/unit/test_handlers.rb" "test/unit/test_http_parser.rb" "test/unit/test_redirect_handler.rb" "test/unit/test_request_progress.rb" "test/unit/test_response.rb" "test/unit/test_stats.rb" "test/unit/test_uriclassifier.rb" "test/unit/test_ws.rb" 
[CruiseControl] Invoking Rake task "test"
./test/test_helper.rb:8: warning: already initialized constant HERE
Loaded suite D:/Users/Luis/ruby/ruby-186-svn-MINGW-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader
Started
.......................F..E................................
Finished in 34.157 seconds.

  1) Failure:
test_horrible_queries(HttpParserTest)
    [./test/unit/test_http_parser.rb:105:in `test_horrible_queries'
     ./test/unit/test_http_parser.rb:103:in `times'
     ./test/unit/test_http_parser.rb:103:in `test_horrible_queries']:
<Mongrel::HttpParserError> exception expected but none was thrown.

  2) Error:
test_parse_strange_headers(HttpParserTest):
Mongrel::HttpParserError: Invalid HTTP format, parsing fails.
    ./test/unit/test_http_parser.rb:49:in `execute'
    ./test/unit/test_http_parser.rb:49:in `test_parse_strange_headers'

59 tests, 458 assertions, 1 failures, 1 errors
rake aborted!
Command failed with status (1): [D:/Users/Luis/ruby/ruby-186-svn-MINGW-ruby...]

(See full trace by running task with --trace)

dir : D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work
command : ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"
executed command : echo D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mingw32/build-1012/build.log && ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:\Users\Luis\.cruise\projects\mongrel-stable_1-2-mingw32\build-1012\build.log 2>&1
exitstatus: 1
STDERR TAIL START

STDERR TAIL END


PROJECT SETTINGS
----------------
# Project-specific configuration for CruiseControl.rb

Project.configure do |project|

  project.email_notifier.emails = ['mongrel-development at rubyforge.org', 'rubyinstaller-devel at rubyforge.org']

  project.rake_task = 'clobber compile test package'
  
  project.scheduler.polling_interval = 5.minutes

end


From builder at mmediasys.com  Tue Apr 22 17:05:08 2008
From: builder at mmediasys.com (Integration Builder)
Date: Tue, 22 Apr 2008 18:05:08 -0300
Subject: [Rubyinstaller-devel] [CruiseControl] mongrel-stable_1-2-mswin32
	build 1012 failed
Message-ID: <480e5304631f0_e3813de3a6187@edicion01.tmail>

The build failed.

  Note: if you set Configuration.dashboard_url in config/site_config.rb, you'd see a link to the build page here.

CHANGES
-------
New revision 1012 detected
Revision 1012 committed by luislavena on 2008-04-18 07:09:44
Wait longer for child process terminate properly (max 20 seconds). 
Imported tests from RubyServices project. (Closes #18).

  M /branches/stable_1-2/projects/mongrel_service/CHANGELOG
  M /branches/stable_1-2/projects/mongrel_service/tests/test_helpers.bas
  M /branches/stable_1-2/projects/mongrel_service/native/console_process.bas
  M /branches/stable_1-2
  M /branches/stable_1-2/projects/mongrel_service/tests/test_console_process.bas
Revision 1011 committed by luislavena on 2008-04-18 07:09:41
Added emulation of slow termination processes.

  M /branches/stable_1-2/projects/mongrel_service/tests/fixtures/mock_process.bas
  M /branches/stable_1-2
Revision 1010 committed by luislavena on 2008-04-18 07:09:38
Include tests projects and child process fixtures.

  M /branches/stable_1-2/projects/mongrel_service/Rakefile
  M /branches/stable_1-2
Revision 1009 committed by luislavena on 2008-04-18 07:09:36
Ported ConsoleProcess tests from RubyServices.

  A /branches/stable_1-2/projects/mongrel_service/tests/fixtures/mock_process.bas
  A /branches/stable_1-2/projects/mongrel_service/tests/test_helpers.bas
  A /branches/stable_1-2/projects/mongrel_service/tests/fixtures
  A /branches/stable_1-2/projects/mongrel_service/tests/test_helpers.bi
  A /branches/stable_1-2/projects/mongrel_service/tests/all_tests.bas
  A /branches/stable_1-2/projects/mongrel_service/tests
  M /branches/stable_1-2
  A /branches/stable_1-2/projects/mongrel_service/tests/test_console_process.bas
Revision 1008 committed by luislavena on 2008-04-18 07:09:33
Updated FreeBASIC compilation tools.

  M /branches/stable_1-2
  M /branches/stable_1-2/projects/mongrel_service/tools/freebasic.rb
Revision 1007 committed by luislavena on 2008-04-17 23:20:50
mongrel_service: Merged changes from trunk.
  M /branches/stable_1-2/projects/mongrel_service/CHANGELOG
  M /branches/stable_1-2/projects/mongrel_service/lib/mongrel_service/init.rb
  M /branches/stable_1-2/projects/mongrel_service/Rakefile
Revision 1006 committed by luislavena on 2008-04-17 18:59:55
mongrel_service: updated ServiceFB library (pistoned).
  M /branches/stable_1-2/projects/mongrel_service/lib/ServiceFB/ServiceFB_Utils.bas
  M /branches/stable_1-2/projects/mongrel_service/lib/ServiceFB
  M /branches/stable_1-2/projects/mongrel_service/lib/ServiceFB/_utils_internals.bi
Revision 1005 committed by evanweaver on 2008-04-17 04:32:11
Document issue with soft restart pid file setting (closes #29).

  M /branches/stable_1-2/bin/mongrel_rails
Revision 1004 committed by evanweaver on 2008-04-17 04:29:44
Apply r1003 to stable.

  M /branches/stable_1-2/bin/mongrel_rails



BUILD LOG
---------
D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"  
D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/bin/ruby extconf.rb
(in D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work)
[CruiseControl] Invoking Rake task "clobber"
Cleaning
- ext/http11/http11-i386-mswin32.def
- ext/http11/http11-i386-mswin32.exp
- ext/http11/http11-i386-mswin32.lib
- ext/http11/http11-i386-mswin32.pdb
- ext/http11/http11.obj
- ext/http11/http11_parser.obj
- ext/http11/Makefile
- ext/http11/vc60.pdb
- lib/http11.so
- log
- test_stderr.log
- test_stdout.log
[CruiseControl] Invoking Rake task "compile"
checking for main() in c.lib... no
creating Makefile

Microsoft (R) Program Maintenance Utility   Version 6.00.9782.0
Copyright (C) Microsoft Corp 1988-1998. All rights reserved.

	D:\Users\Luis\ruby\ruby-186-p114-VC6-rubygems-110\bin\ruby -e "puts 'EXPORTS', 'Init_http11'"  > http11-i386-mswin32.def
	cl -nologo -I. -I. -ID:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/i386-mswin32 -I. -MD -Zi -O2b2xg- -G6  -c -Tchttp11.c
http11.c
	cl -nologo -I. -I. -ID:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/1.8/i386-mswin32 -I. -MD -Zi -O2b2xg- -G6  -c -Tchttp11_parser.c
http11_parser.c
	cl -nologo -LD -Fehttp11.so http11.obj http11_parser.obj msvcrt-ruby18.lib  oldnames.lib user32.lib advapi32.lib ws2_32.lib  -link -incremental:no -debug -opt:ref -opt:icf -dll -libpath:"." -libpath:"D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib" -def:http11-i386-mswin32.def -implib:http11-i386-mswin32.lib -pdb:http11-i386-mswin32.pdb
   Creating library http11-i386-mswin32.lib and object http11-i386-mswin32.exp
D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/bin/ruby -Ilib;ext;bin;test "D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader.rb" "test/test_helper.rb" "test/unit/test_cgi_wrapper.rb" "test/unit/test_command.rb" "test/unit/test_conditional.rb" "test/unit/test_configurator.rb" "test/unit/test_debug.rb" "test/unit/test_handlers.rb" "test/unit/test_http_parser.rb" "test/unit/test_redirect_handler.rb" "test/unit/test_request_progress.rb" "test/unit/test_response.rb" "test/unit/test_stats.rb" "test/unit/test_uriclassifier.rb" "test/unit/test_ws.rb" 
[CruiseControl] Invoking Rake task "test"
./test/test_helper.rb:8: warning: already initialized constant HERE
Loaded suite D:/Users/Luis/ruby/ruby-186-p114-VC6-rubygems-110/lib/ruby/gems/1.8/gems/rake-0.8.1/lib/rake/rake_test_loader
Started
.......................F..E................................
Finished in 32.687 seconds.

  1) Failure:
test_horrible_queries(HttpParserTest)
    [./test/unit/test_http_parser.rb:105:in `test_horrible_queries'
     ./test/unit/test_http_parser.rb:103:in `times'
     ./test/unit/test_http_parser.rb:103:in `test_horrible_queries']:
<Mongrel::HttpParserError> exception expected but none was thrown.

  2) Error:
test_parse_strange_headers(HttpParserTest):
Mongrel::HttpParserError: Invalid HTTP format, parsing fails.
    ./test/unit/test_http_parser.rb:49:in `execute'
    ./test/unit/test_http_parser.rb:49:in `test_parse_strange_headers'

59 tests, 458 assertions, 1 failures, 1 errors
rake aborted!
Command failed with status (1): [D:/Users/Luis/ruby/ruby-186-p114-VC6-rubyg...]

(See full trace by running task with --trace)

dir : D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work
command : ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear"
executed command : echo D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/work Luis$ ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:/Users/Luis/.cruise/projects/mongrel-stable_1-2-mswin32/build-1012/build.log && ruby -e "require 'rubygems' rescue nil; require 'rake'; load 'D:/Users/Luis/tools/ccrb/tasks/cc_build.rake'; ARGV << '--nosearch' << 'cc:build'; Rake.application.run; ARGV.clear" >> D:\Users\Luis\.cruise\projects\mongrel-stable_1-2-mswin32\build-1012\build.log 2>&1
exitstatus: 1
STDERR TAIL START

STDERR TAIL END


PROJECT SETTINGS
----------------
# Project-specific configuration for CruiseControl.rb

Project.configure do |project|

  project.email_notifier.emails = ['mongrel-development at rubyforge.org', 'rubyinstaller-devel at rubyforge.org']

  project.rake_task = 'clobber compile test package'
  
  project.scheduler.polling_interval = 5.minutes

end


From luislavena at gmail.com  Wed Apr 23 01:36:39 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Wed, 23 Apr 2008 02:36:39 -0300
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
Message-ID: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>

Hello fellow developers,

It seems I neglected to push or decide what to do regarding the code
being developed for what I call it 'installer3' MinGW build.

So far, the idea was push these changes back into svn, which will
lower the entry point for others developers wanting to contribute back
to it.

Since 'git' is the new cool toy, I wondered a lot of Installer3 should
also move to it, even when Git support is not 100% what I will love to
have.

Or, instead, I was thinking also move the repo from my host to
launchpad, which will raise some critics about using closed-source
software to host an open source project (github is closed source
anyway)...

What do you guys think will be best? I'm open to suggestions.

What I love of the dvcs workflow is that I can keep local branches to
foster experimentation and try new things before I merge them back
cherry pick stuff from them.

Thank you for your time guys, feel free to comment.
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From will at hotgazpacho.org  Wed Apr 23 08:01:59 2008
From: will at hotgazpacho.org (Will Green)
Date: Wed, 23 Apr 2008 08:01:59 -0400
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
In-Reply-To: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
References: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
Message-ID: <480F2537.7030907@hotgazpacho.org>

SVN has the lowest barrier to entry for Win32 developers (Collabnet's 
SVN installer for Windows, TortoiseSVN).

To the best of knowledge, all the other tools (bzr, git, etc) do not 
have a stable, native win32 client. That is, they all require an 
existing Cygwin or MinGW install.

The other benefit of SVN is that if folks want to use git, git-svn can 
push their changes back to the SVN repo.

Given that this is a Win32 project for Win32 developers, my vote is for SVN.

==
Will Green

Luis Lavena wrote:
> Hello fellow developers,
>
> It seems I neglected to push or decide what to do regarding the code
> being developed for what I call it 'installer3' MinGW build.
>
> So far, the idea was push these changes back into svn, which will
> lower the entry point for others developers wanting to contribute back
> to it.
>
> Since 'git' is the new cool toy, I wondered a lot of Installer3 should
> also move to it, even when Git support is not 100% what I will love to
> have.
>
> Or, instead, I was thinking also move the repo from my host to
> launchpad, which will raise some critics about using closed-source
> software to host an open source project (github is closed source
> anyway)...
>
> What do you guys think will be best? I'm open to suggestions.
>
> What I love of the dvcs workflow is that I can keep local branches to
> foster experimentation and try new things before I merge them back
> cherry pick stuff from them.
>
> Thank you for your time guys, feel free to comment.
>   


From luislavena at gmail.com  Wed Apr 23 10:12:06 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Wed, 23 Apr 2008 11:12:06 -0300
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
In-Reply-To: <480F2537.7030907@hotgazpacho.org>
References: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
	<480F2537.7030907@hotgazpacho.org>
Message-ID: <71166b3b0804230712j5d4ee585nb905b16bc19469cd@mail.gmail.com>

On Wed, Apr 23, 2008 at 9:01 AM, Will Green <will at hotgazpacho.org> wrote:
> SVN has the lowest barrier to entry for Win32 developers (Collabnet's
>  SVN installer for Windows, TortoiseSVN).
>

Good points. Even I have installed TortoiseSVN, no longer uses it due
some memory consumption that imposes on explorer.exe....

Anyway...

>  To the best of knowledge, all the other tools (bzr, git, etc) do not
>  have a stable, native win32 client. That is, they all require an
>  existing Cygwin or MinGW install.
>

On that topic, based in my personal experience. Bzr was one, if not
the only one that worked out of the box, using the standalone version
or the python packages. but as you state, there are no "stable"
releases or sometimes requires fulfill dependencies that can make
things a bit harder for developers.

>  The other benefit of SVN is that if folks want to use git, git-svn can
>  push their changes back to the SVN repo.
>

Right know I'm using bzr-svn which perform similar operations that
git-svn, so I'm good with that.

>  Given that this is a Win32 project for Win32 developers, my vote is for SVN.
>

Then I'll push the changes back into existing repo (but it's own
branch), and keep git or bzr for local workflow only.

Thank you for your comments Will,
.
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gerardo.santana at gmail.com  Wed Apr 23 11:59:29 2008
From: gerardo.santana at gmail.com (=?ISO-8859-1?Q?Gerardo_Santana_G=F3mez_Garrido?=)
Date: Wed, 23 Apr 2008 10:59:29 -0500
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
In-Reply-To: <71166b3b0804230712j5d4ee585nb905b16bc19469cd@mail.gmail.com>
References: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
	<480F2537.7030907@hotgazpacho.org>
	<71166b3b0804230712j5d4ee585nb905b16bc19469cd@mail.gmail.com>
Message-ID: <b06e22130804230859x4a3e7d1fv98254bb82b8b8111@mail.gmail.com>

I'm a recent converted from CVS to Git, and find it very nice.

One of its advantages is precisely what Luis mentions, to have the
ability to create local experimental branches without impacting the
main repository, and cherry-picking what you want.

github.com helps open source projects to get even more exposure which
is a good thing, getting attention and potentially getting more
contributions back. Nothing that I can scientifically prove though.

AFAIK, there's a Git version for MSYS. According to my logic book,
since installer3 will run on MSYS, it doesn't feel wrong to use
msysgit, which is already used in production environments (according
to its Wikipedia entry).

And more importantly, I don't like SVN and its long list of dependencies :-)

That's my biased opinion, for whatever it's worth.

-- 
Gerardo Santana

From luislavena at gmail.com  Wed Apr 23 12:10:02 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Wed, 23 Apr 2008 13:10:02 -0300
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
In-Reply-To: <b06e22130804230859x4a3e7d1fv98254bb82b8b8111@mail.gmail.com>
References: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
	<480F2537.7030907@hotgazpacho.org>
	<71166b3b0804230712j5d4ee585nb905b16bc19469cd@mail.gmail.com>
	<b06e22130804230859x4a3e7d1fv98254bb82b8b8111@mail.gmail.com>
Message-ID: <71166b3b0804230910u27a04773x9e8ba08cef0e4565@mail.gmail.com>

On Wed, Apr 23, 2008 at 12:59 PM, Gerardo Santana G?mez Garrido
<gerardo.santana at gmail.com> wrote:
> I'm a recent converted from CVS to Git, and find it very nice.
>

Still get confused due the high quantity of commands available and the
huge amount of combos to get simple things going... That's what I call
a bad UI versus a good UI...

>  One of its advantages is precisely what Luis mentions, to have the
>  ability to create local experimental branches without impacting the
>  main repository, and cherry-picking what you want.
>
>  github.com helps open source projects to get even more exposure which
>  is a good thing, getting attention and potentially getting more
>  contributions back. Nothing that I can scientifically prove though.
>

I'm not 100% positive abut github. Like Zed wasn't with the move of
the svn repository outside rubyforge. We could better use gitorius [1]
instead of a closed source implementation, like if we go for LaunchPad
instead.

Even we have a full repo that can be relocated/pushed to another
hosting/site. I found quite annoying the relocation of repositories of
OSS projects.

>  AFAIK, there's a Git version for MSYS. According to my logic book,
>  since installer3 will run on MSYS, it doesn't feel wrong to use
>  msysgit, which is already used in production environments (according
>  to its Wikipedia entry).
>

Well, I mention msysGit in the original mail, basically it works,
still it misses a few bits but works. Still, we should keep using the
SVN repo at least for users already using Subversion at their
companies or working with other open-source projects that want to
contribute back, at least with patches.

What we do with MSYS is quite different of what Git does. we only use
it for make and sh (bash) to get compatible support for configure
scripts and generated makefiles.

On the other hand, msysGit have it's own copy that I wouldn't touch
since could break something :-P

>  And more importantly, I don't like SVN and its long list of dependencies :-)
>

We wouldn't be building svn from scratch, trust me, I'll not go that
path... tried several times (more than 15 times) and failed all of
them.

>  That's my biased opinion, for whatever it's worth.
>

Not biased, but valid. Is good to know there are users in Windows
using Git, and I'll love to heard back regarding it's usage.

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Wed Apr 23 13:49:58 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Wed, 23 Apr 2008 12:49:58 -0500
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
In-Reply-To: <71166b3b0804230910u27a04773x9e8ba08cef0e4565@mail.gmail.com>
References: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
	<480F2537.7030907@hotgazpacho.org>
	<71166b3b0804230712j5d4ee585nb905b16bc19469cd@mail.gmail.com>
	<b06e22130804230859x4a3e7d1fv98254bb82b8b8111@mail.gmail.com>
	<71166b3b0804230910u27a04773x9e8ba08cef0e4565@mail.gmail.com>
Message-ID: <e9ece6840804231049m54d43a03o2d7a6e2e42940e4e@mail.gmail.com>

On Wed, Apr 23, 2008 at 11:10 AM, Luis Lavena <luislavena at gmail.com> wrote:
> On Wed, Apr 23, 2008 at 12:59 PM, Gerardo Santana G?mez Garrido
>  <gerardo.santana at gmail.com> wrote:
>  > I'm a recent converted from CVS to Git, and find it very nice.
>  >
>
>  Still get confused due the high quantity of commands available and the
>  huge amount of combos to get simple things going... That's what I call
>  a bad UI versus a good UI...
>
>
>  >  One of its advantages is precisely what Luis mentions, to have the
>  >  ability to create local experimental branches without impacting the
>  >  main repository, and cherry-picking what you want.
>  >
>  >  github.com helps open source projects to get even more exposure which
>  >  is a good thing, getting attention and potentially getting more
>  >  contributions back. Nothing that I can scientifically prove though.
>  >
>
>  I'm not 100% positive abut github. Like Zed wasn't with the move of
>  the svn repository outside rubyforge. We could better use gitorius [1]
>  instead of a closed source implementation, like if we go for LaunchPad
>  instead.
>

A lot of projects do primary souce control on Git, but merge to SVN on
rubyforge on a regular basis.  We could do the same, or the reverse.
Maybe we could keep SVN as the primary, but sync it up to Git.
Rubyforge also allows for a Git repo, although I don't know if it
would be worth changing to it at this point.  Also, I don't think we
can fault GitHub for being closed source.  After all, we're all here
because we want Ruby to work as well as possible on a closed source
OS.  GitHub is friendly _and free_ to open source projects, and I
agree with Gerardo that it could help us get more attention.

>  Even we have a full repo that can be relocated/pushed to another
>  hosting/site. I found quite annoying the relocation of repositories of
>  OSS projects.
>
>
>  >  AFAIK, there's a Git version for MSYS. According to my logic book,
>  >  since installer3 will run on MSYS, it doesn't feel wrong to use
>  >  msysgit, which is already used in production environments (according
>  >  to its Wikipedia entry).
>  >
>
>  Well, I mention msysGit in the original mail, basically it works,
>  still it misses a few bits but works. Still, we should keep using the
>  SVN repo at least for users already using Subversion at their
>  companies or working with other open-source projects that want to
>  contribute back, at least with patches.

I like this idea, there is no reason we can't do both.

That's my two cents.

Thanks,

Gordon

From luislavena at gmail.com  Wed Apr 23 14:07:53 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Wed, 23 Apr 2008 15:07:53 -0300
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
In-Reply-To: <e9ece6840804231049m54d43a03o2d7a6e2e42940e4e@mail.gmail.com>
References: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
	<480F2537.7030907@hotgazpacho.org>
	<71166b3b0804230712j5d4ee585nb905b16bc19469cd@mail.gmail.com>
	<b06e22130804230859x4a3e7d1fv98254bb82b8b8111@mail.gmail.com>
	<71166b3b0804230910u27a04773x9e8ba08cef0e4565@mail.gmail.com>
	<e9ece6840804231049m54d43a03o2d7a6e2e42940e4e@mail.gmail.com>
Message-ID: <71166b3b0804231107u6ddfd9d3p16e392ca9d1b0ab7@mail.gmail.com>

On Wed, Apr 23, 2008 at 2:49 PM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> On Wed, Apr 23, 2008 at 11:10 AM, Luis Lavena <luislavena at gmail.com> wrote:
>  >
>  >  I'm not 100% positive abut github. Like Zed wasn't with the move of
>  >  the svn repository outside rubyforge. We could better use gitorius [1]
>  >  instead of a closed source implementation, like if we go for LaunchPad
>  >  instead.
>  >
>
>  A lot of projects do primary souce control on Git, but merge to SVN on
>  rubyforge on a regular basis.  We could do the same, or the reverse.
>  Maybe we could keep SVN as the primary, but sync it up to Git.
>  Rubyforge also allows for a Git repo, although I don't know if it
>  would be worth changing to it at this point.  Also, I don't think we
>  can fault GitHub for being closed source.  After all, we're all here
>  because we want Ruby to work as well as possible on a closed source
>  OS.  GitHub is friendly _and free_ to open source projects, and I
>  agree with Gerardo that it could help us get more attention.
>
>

I think I missed the point about Zed's comments. The issue with the
svn repo is the "ownership" of it, the viability of the system you're
depending on.

DVCS break that model into pieces, all the users contributing to a
project have copies _almost_ up-to date, so I don't see issues with
that, Git or Bzr models.

>  >
>  >  Well, I mention msysGit in the original mail, basically it works,
>  >  still it misses a few bits but works. Still, we should keep using the
>  >  SVN repo at least for users already using Subversion at their
>  >  companies or working with other open-source projects that want to
>  >  contribute back, at least with patches.
>
>  I like this idea, there is no reason we can't do both.
>

I'll check if latest msysGit support dcommit to push changes back into
SVN, after all, we will only maintain a mirror branch over there.

>  That's my two cents.

Why only two cents? we all demand more! ;-)

>  Thanks,
>

So basically:

- GitHub rubyinstaller project that holds the new code we have been creating.
- Push via dcommit the changes back into SVN repo [1] as installer3 branch.
- Provide simple instructions in the docs and rake helpers into the
recipes for managing the git repo.

We all agree?

[1] http://rubyinstaller.rubyforge.org/svn

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From hramrach at centrum.cz  Thu Apr 24 06:28:41 2008
From: hramrach at centrum.cz (Michal Suchanek)
Date: Thu, 24 Apr 2008 12:28:41 +0200
Subject: [Rubyinstaller-devel] RFC: Version Control for new Installer?
In-Reply-To: <480F2537.7030907@hotgazpacho.org>
References: <71166b3b0804222236xe73e744qfd92a92085605b5c@mail.gmail.com>
	<480F2537.7030907@hotgazpacho.org>
Message-ID: <a5d587fb0804240328r1b5753c0r5b6e26e98059ffdc@mail.gmail.com>

On 23/04/2008, Will Green <will at hotgazpacho.org> wrote:
> SVN has the lowest barrier to entry for Win32 developers (Collabnet's
>  SVN installer for Windows, TortoiseSVN).
>
>  To the best of knowledge, all the other tools (bzr, git, etc) do not
>  have a stable, native win32 client. That is, they all require an
>  existing Cygwin or MinGW install.
>
>  The other benefit of SVN is that if folks want to use git, git-svn can
>  push their changes back to the SVN repo.
>
>  Given that this is a Win32 project for Win32 developers, my vote is for SVN.
>

I also find SVN or CVS a tool more accessible than git, at least on Windows.

Still I can live with snapshot zips (or tarballs) exported from anything.
Zip archives are probably the most acceptable option for Windows
users, the OS has built-in support for them.

Thanks

Michal

From gthiesfeld at gmail.com  Fri Apr 25 00:01:21 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Thu, 24 Apr 2008 23:01:21 -0500
Subject: [Rubyinstaller-devel] Gem compatibility and platform specific
	dependencies
Message-ID: <e9ece6840804242101u13bf89fh1647b6d22f2382fa@mail.gmail.com>

This may be better asked on the RubyGems list, but I thought I'd try
here first.  I'm working on a patch[1] for the github gem[2].  The
github command requires open3, which doesn't work on Windows because
of fork, so I have it check for platform, and require WIn32-Open3 if
on Windows.  So, I have two questions about how to go about this.

First, is Gem.win_platform?  the proper way to go about this.  I think
it's safe to say this package will always be distributed as a gem, so
Gem will always exist when I need it.

Second, is there a way to have Gem depedencies that are platform
specific?  Open3 is part of the stdlib, so no dependency there, but on
Windows we need the Win32 gem.  I've borrowed a pattern from RSpec for
this:

  begin
    require 'win32 package'
  rescue LoadError
    warn "You must 'gem install the win32 package to do this on Windows"
    exit 1
  end

But in the case of RSpec, it's only used for colored console output.
If you don't use that feature, you'll never know you need to have the
gem installed.  In this case, the whole thing just won't work.
That's what dependencies are all about, I'd think.  If there isn't
such a thing as platform specific dependencies, then what I'm fine
doing it this way.  I'm just wondering if there's a better way.

Thanks,

Gordon

[1] http://github.com/vertiginous/github-gem/commit/a4f54d3b8d51549ff9922c55de160b713a761d8b
[2] http://github.com/defunkt/github-gem/tree/master

From luislavena at gmail.com  Fri Apr 25 00:21:02 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Fri, 25 Apr 2008 01:21:02 -0300
Subject: [Rubyinstaller-devel] Gem compatibility and platform specific
	dependencies
In-Reply-To: <e9ece6840804242101u13bf89fh1647b6d22f2382fa@mail.gmail.com>
References: <e9ece6840804242101u13bf89fh1647b6d22f2382fa@mail.gmail.com>
Message-ID: <71166b3b0804242121q5e1b60f6of55465a51f657e16@mail.gmail.com>

On Fri, Apr 25, 2008 at 1:01 AM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
> This may be better asked on the RubyGems list, but I thought I'd try
>  here first.  I'm working on a patch[1] for the github gem[2].  The
>  github command requires open3, which doesn't work on Windows because
>  of fork, so I have it check for platform, and require WIn32-Open3 if
>  on Windows.  So, I have two questions about how to go about this.
>
>  First, is Gem.win_platform?  the proper way to go about this.  I think
>  it's safe to say this package will always be distributed as a gem, so
>  Gem will always exist when I need it.
>

I think is safe to assume Gem.win_platform? is available since you're
implementing that into a project that get's distributed/installed as
gem.

The only problem is that win_platform? will be 'true' even for cygwin
implementation, which will render win32/open3 useless and I can smell
that will crash :-P

>  Second, is there a way to have Gem depedencies that are platform
>  specific?  Open3 is part of the stdlib, so no dependency there, but on
>  Windows we need the Win32 gem.  I've borrowed a pattern from RSpec for
>  this:
>

To our bad, there isn't a way to do that, you can create a
"windows-specific" version and in that gem include the dependencies,
but without a real need to do that, I think workaround in the code
should be used.

>   begin
>     require 'win32 package'
>   rescue LoadError
>     warn "You must 'gem install the win32 package to do this on Windows"
>     exit 1
>   end
>
>  But in the case of RSpec, it's only used for colored console output.
>  If you don't use that feature, you'll never know you need to have the
>  gem installed.  In this case, the whole thing just won't work.
>  That's what dependencies are all about, I'd think.  If there isn't
>  such a thing as platform specific dependencies, then what I'm fine
>  doing it this way.  I'm just wondering if there's a better way.
>

A good workaround will be poke the ruby-core developers to merge
win32/open3 package in and reduce our code cluttery, but I don't see
that will happen anytime soon :-P

>  Thanks,

Thanks to you Gordon!

BTW: just pushed installer3 into the repository:
http://rubyinstaller.rubyforge.org/svn/trunk/installer3/

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From gthiesfeld at gmail.com  Fri Apr 25 02:02:11 2008
From: gthiesfeld at gmail.com (Gordon Thiesfeld)
Date: Fri, 25 Apr 2008 01:02:11 -0500
Subject: [Rubyinstaller-devel] Gem compatibility and platform specific
	dependencies
In-Reply-To: <71166b3b0804242121q5e1b60f6of55465a51f657e16@mail.gmail.com>
References: <e9ece6840804242101u13bf89fh1647b6d22f2382fa@mail.gmail.com>
	<71166b3b0804242121q5e1b60f6of55465a51f657e16@mail.gmail.com>
Message-ID: <e9ece6840804242302i3a97c1eehaa2dbc8e7cb3aa68@mail.gmail.com>

On Thu, Apr 24, 2008 at 11:21 PM, Luis Lavena <luislavena at gmail.com> wrote:
> On Fri, Apr 25, 2008 at 1:01 AM, Gordon Thiesfeld <gthiesfeld at gmail.com> wrote:
>  > This may be better asked on the RubyGems list, but I thought I'd try
>  >  here first.  I'm working on a patch[1] for the github gem[2].  The
>  >  github command requires open3, which doesn't work on Windows because
>  >  of fork, so I have it check for platform, and require WIn32-Open3 if
>  >  on Windows.  So, I have two questions about how to go about this.
>  >
>  >  First, is Gem.win_platform?  the proper way to go about this.  I think
>  >  it's safe to say this package will always be distributed as a gem, so
>  >  Gem will always exist when I need it.
>  >
>
>  I think is safe to assume Gem.win_platform? is available since you're
>  implementing that into a project that get's distributed/installed as
>  gem.
>
>  The only problem is that win_platform? will be 'true' even for cygwin
>  implementation, which will render win32/open3 useless and I can smell
>  that will crash :-P
>

Oh yeah, I forgot about cygwin.  I guess open3 works on cygwin?  Is
this the proper regex, then?

RUBY_PLATFORM =~ /mswin|mingw/

>
>  >  Second, is there a way to have Gem depedencies that are platform
>  >  specific?  Open3 is part of the stdlib, so no dependency there, but on
>  >  Windows we need the Win32 gem.  I've borrowed a pattern from RSpec for
>  >  this:
>  >
>
>  To our bad, there isn't a way to do that, you can create a
>  "windows-specific" version and in that gem include the dependencies,
>  but without a real need to do that, I think workaround in the code
>  should be used.
>

I was hoping I'd missed something simple, like:

add_dependency() if Gem.win_platform?

But if it's not there I'm not going to ask for it.  At least not in this case.

>
>  >   begin
>  >     require 'win32 package'
>  >   rescue LoadError
>  >     warn "You must 'gem install the win32 package to do this on Windows"
>  >     exit 1
>  >   end
>  >
>  >  But in the case of RSpec, it's only used for colored console output.
>  >  If you don't use that feature, you'll never know you need to have the
>  >  gem installed.  In this case, the whole thing just won't work.
>  >  That's what dependencies are all about, I'd think.  If there isn't
>  >  such a thing as platform specific dependencies, then what I'm fine
>  >  doing it this way.  I'm just wondering if there's a better way.
>  >
>
>  A good workaround will be poke the ruby-core developers to merge
>  win32/open3 package in and reduce our code cluttery, but I don't see
>  that will happen anytime soon :-P
>

I don't follow ruby-core that much, has this ever been suggested?  I'm
sure it must have at some point.

>  >  Thanks,
>
>  Thanks to you Gordon!
>
>  BTW: just pushed installer3 into the repository:
>  http://rubyinstaller.rubyforge.org/svn/trunk/installer3/

Should I switch from bzr to svn, or hold out for the git repo announcement ;-)

-- Gordon

From luislavena at gmail.com  Sun Apr 27 19:18:38 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sun, 27 Apr 2008 20:18:38 -0300
Subject: [Rubyinstaller-devel] [Fwd: GCC 4.x on Windows (now with more
	statistics!)]
In-Reply-To: <4814F564.10003@gmail.com>
References: <4814F564.10003@gmail.com>
Message-ID: <71166b3b0804271618h147b6ba1jf75998acdd822f50@mail.gmail.com>

On Sun, Apr 27, 2008 at 6:51 PM, Phillip Gawlowski
<cmdjackryan at googlemail.com> wrote:
>
>  Since I have no clue if this arrived on the ruby installer list:
>

Nop, it didn't got into the list, you need to be subscribed to allow mails :-P

Anyway, see my comments inline.

>  - -------- Original Message --------
>  Subject: GCC 4.x on Windows (now with more statistics!)
>  Date: Fri, 25 Apr 2008 19:16:23 +0200
>  From: Phillip Gawlowski <cmdjackryan at gmail.com>
>  To: For RubyInstaller Developers <rubyinstaller-devel at rubyforge.org>
>
>  I just read Luis' blog post on the state of the project[0], and saw that
>  MinGW with GCC4 is missing.
>
>  While it is not officially supported, there are binaries available:
>
>  http://www.develer.com/oss/GccWinBinaries
>

I prefer TDM release, since a downloadable file that could be extracted:

http://www.tdragon.net/recentgcc/

The one that Giovanni maintains contains all bundled into a nice
setup, which is good if you plan to install it on end-user computer,
but prove no advantage for us.

Also, it contains mingw32-make, which proven to be problematic for
Ruby assumptions (the makefile extconf generates is more posix like
than windows like).

>  GCC4 is a major update to GCC, especially in terms of speed (of the
>  compiler, and the compiled source), and the Ruby-Installer might want to
>  use it (for testing purposes, since MinGW themselves don't consider it
>  production ready yet, but GCC4 is out for years now...).
>

Yes, have played with it several times, you can see it in my older
pastie, using noven.net release instead of TDM at that time (July
2007).

http://pastie.caboo.se/95210

>  Okay, for comparison (using Luis' 'benchmark'):
>
>  mswin32: 30.334 sec | speedup: 1
>  gcc4: 18.557 sec | speedup: 1.63
>  gcc3: 19.528 sec | speedup: 1.55
>

Nice numbers, the benchmark just use an average test instead of more
of the specialized benchmarks out there, The nice thing is all the
implementations support the script out of the box.

Now that Installer3 is mirrored in the svn repository[1] you can grab
them and provide patches for use of GCC4 instead of GCC3, you're
welcome! :-D

(The good thing of TDM releases is they keep using .tar.gz and .zip
files for the packages, so there is no need to do some tricks for
getting the files).

Thank you for your time and effort on testing all this!
PS: CC'ed this mail for the list to know about this!

[1] http://rubyinstaller.rubyforge.org/svn/trunk/installer3/
-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From luislavena at gmail.com  Sun Apr 27 19:44:14 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sun, 27 Apr 2008 20:44:14 -0300
Subject: [Rubyinstaller-devel] [Fwd: GCC 4.x on Windows (now with more
	statistics!)]
In-Reply-To: <48150E1F.5040208@gmail.com>
References: <4814F564.10003@gmail.com>
	<71166b3b0804271618h147b6ba1jf75998acdd822f50@mail.gmail.com>
	<48150E1F.5040208@gmail.com>
Message-ID: <71166b3b0804271644y2ba92882uddb7e2a070a2f6d2@mail.gmail.com>

On Sun, Apr 27, 2008 at 8:37 PM, Phillip Gawlowski
<cmdjackryan at googlemail.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
>  Hash: SHA1
>
>  Luis Lavena wrote:
>  | On Sun, Apr 27, 2008 at 6:51 PM, Phillip Gawlowski
>  | <cmdjackryan at googlemail.com> wrote:
>  |>  Since I have no clue if this arrived on the ruby installer list:
>  |>
>  |
>  | Nop, it didn't got into the list, you need to be subscribed to allow
>  mails :-P
>
>  Actually, I am. I didn't get a bounce, no nothing. Strange. I'll
>  investigate (and add all the possible GMail addresses I have, if need be
>
>  :P).
>

Yeah, mailman sometimes does weird things :-P

>  | Anyway, see my comments inline.
>  |
>  | I prefer TDM release, since a downloadable file that could be extracted:
>  |
>  | http://www.tdragon.net/recentgcc/
>  |
>  | The one that Giovanni maintains contains all bundled into a nice
>  | setup, which is good if you plan to install it on end-user computer,
>  | but prove no advantage for us.
>
>  What ever floats your boat, really. I just picked that one, since it was
>  the least hassle to install for me, when I needed to grab a GCC.
>

Yes, as end user is more easy to install it, but if you take a look at
what we do with the recipes, you will see we are actually downloading
GCC and MSYS to actually create a development environment, and not
forcing the developer already have one installed ;-)

>
>  |
>  | Yes, have played with it several times, you can see it in my older
>  | pastie, using noven.net release instead of TDM at that time (July
>  | 2007).
>  |
>  | http://pastie.caboo.se/95210
>
>  Hm, interesting..
>
>  The speed increase GCC3 vs GCC4 seems most notable in lower end
>  machines, like mine.
>

Agree there too, running on the Core 2 Duo no ruby implemenation took
real advantage of the CPU power it have available: it never used more
than 45% :-P

>
>  |>  Okay, for comparison (using Luis' 'benchmark'):
>  |>
>  |>  mswin32: 30.334 sec | speedup: 1
>  |>  gcc4: 18.557 sec | speedup: 1.63
>  |>  gcc3: 19.528 sec | speedup: 1.55
>  |>
>  |
>  | Nice numbers, the benchmark just use an average test instead of more
>  | of the specialized benchmarks out there, The nice thing is all the
>  | implementations support the script out of the box.
>
>  Yep, it is a good baseline, but not a 'real' benchmark, unfortunately.
>  But good enough to see trends, anyway.
>

Well, actually is more than a real benchmark. As example, the YARV
bench only prove and excersice especific parts of the Ruby VM that has
been optimized, but is not a real life test like PetStore is done for
testing Rails performance.

On average, I complete testing of a Rails application from 80 seconds
(580 sepcs) down to 48 seconds running with MinGW build of Ruby.

>  | Now that Installer3 is mirrored in the svn repository[1] you can grab
>  | them and provide patches for use of GCC4 instead of GCC3, you're
>  | welcome! :-D
>
>  Well, if I can do that, I will. But no guarantees (My GCC Fu is very,
>  very weak). ;)
>

You don't need a lot of GCC fu, just replace the locations for the GCC
packages, check they download, extract and can successfuly 'configure'
and compile Ruby :-)

>  However, when I built Ruby with GCC4, I had no problems to build it,
>  actually. I think that hinges more on the makefiles than the C source.
>

Did you use configure to generate the makefile? if so that means you
need MSYS installed too, don't remember if Giovanni installed provides
it.

>
>  | Thank you for your time and effort on testing all this!
>
>  No problem, really.
>
>
>  | PS: CC'ed this mail for the list to know about this!
>
>  Thanks.
>
>
>  | [1] http://rubyinstaller.rubyforge.org/svn/trunk/installer3/
>
>
>  - --
>  Phillip Gawlowski
>  Twitter: twitter.com/cynicalryan
>
>  Use recursive procedures for recursively-defined data structures.
>  ~            - The Elements of Programming Style (Kernighan & Plaugher)
>
>  -----BEGIN PGP SIGNATURE-----
>  Version: GnuPG v1.4.8 (MingW32)
>  Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
>
>  iEYEARECAAYFAkgVDh0ACgkQbtAgaoJTgL928ACfcBlaRrX/edneJpttroysYHMh
>  ttsAnikkUq2jFwX9YUvh/Z2gh4eiVZMP
>  =IYIS
>  -----END PGP SIGNATURE-----
>



-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams

From cmdjackryan at googlemail.com  Sun Apr 27 19:56:47 2008
From: cmdjackryan at googlemail.com (Phillip Gawlowski)
Date: Mon, 28 Apr 2008 01:56:47 +0200
Subject: [Rubyinstaller-devel] [Fwd: GCC 4.x on Windows (now with more
	statistics!)]
In-Reply-To: <71166b3b0804271644y2ba92882uddb7e2a070a2f6d2@mail.gmail.com>
References: <4814F564.10003@gmail.com>	
	<71166b3b0804271618h147b6ba1jf75998acdd822f50@mail.gmail.com>	
	<48150E1F.5040208@gmail.com>
	<71166b3b0804271644y2ba92882uddb7e2a070a2f6d2@mail.gmail.com>
Message-ID: <481512BF.5010008@gmail.com>

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

Luis Lavena wrote:
|
| Yeah, mailman sometimes does weird things :-P

Don't get me started. :P

|
| Yes, as end user is more easy to install it, but if you take a look at
| what we do with the recipes, you will see we are actually downloading
| GCC and MSYS to actually create a development environment, and not
| forcing the developer already have one installed ;-)

Well, yeah. But it'd be nice if you'd have checked if I had MSYS/MinGW
installed, like I do. :P

| Agree there too, running on the Core 2 Duo no ruby implemenation took
| real advantage of the CPU power it have available: it never used more
| than 45% :-P

No surprise there. MRI isn't multi-threaded, so of course it only uses
one CPU :P

| Well, actually is more than a real benchmark. As example, the YARV
| bench only prove and excersice especific parts of the Ruby VM that has
| been optimized, but is not a real life test like PetStore is done for
| testing Rails performance.

Yes and no. It it is a bit of number crunching, but doesn't test things
like IO, compression, Arrays, etc. I'm not talking about benchmarks
showing off features. ;)

| You don't need a lot of GCC fu, just replace the locations for the GCC
| packages, check they download, extract and can successfuly 'configure'
| and compile Ruby :-)

Well, that is easy (Search and replace FTW :P). But if something stumps
because GCC decides to roll over and die, I'm of no help. ;)

| Did you use configure to generate the makefile? if so that means you
| need MSYS installed too, don't remember if Giovanni installed provides
| it.

I have a full MSYS environment with GCC4, so of course I used configure. :P

P.S.: Sent to Rubyinstaller ML, too, in hope of a working subscription
now...

- --
Phillip Gawlowski
Twitter: twitter.com/cynicalryan

~ - You know you've been hacking too long when...
...a fly lands on the screen and you try to pick it up with the mouse
and put it in the onscreen trashcan!
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkgVEr4ACgkQbtAgaoJTgL/ABQCfWxP4elzegkVvmhnnClzVtH7h
zrYAoIC8W7qpkcVDMJU98aADCEwJLe/Y
=OYUE
-----END PGP SIGNATURE-----

From luislavena at gmail.com  Sun Apr 27 20:38:06 2008
From: luislavena at gmail.com (Luis Lavena)
Date: Sun, 27 Apr 2008 21:38:06 -0300
Subject: [Rubyinstaller-devel] [Fwd: GCC 4.x on Windows (now with more
	statistics!)]
In-Reply-To: <481512BF.5010008@gmail.com>
References: <4814F564.10003@gmail.com>
	<71166b3b0804271618h147b6ba1jf75998acdd822f50@mail.gmail.com>
	<48150E1F.5040208@gmail.com>
	<71166b3b0804271644y2ba92882uddb7e2a070a2f6d2@mail.gmail.com>
	<481512BF.5010008@gmail.com>
Message-ID: <71166b3b0804271738v6ec75f54t407989edf9147c55@mail.gmail.com>

On Sun, Apr 27, 2008 at 8:56 PM, Phillip Gawlowski
<cmdjackryan at googlemail.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
>  Hash: SHA1
>
>  Luis Lavena wrote:
>  |
>  | Yeah, mailman sometimes does weird things :-P
>
>  Don't get me started. :P
>
>
>  |
>  | Yes, as end user is more easy to install it, but if you take a look at
>  | what we do with the recipes, you will see we are actually downloading
>  | GCC and MSYS to actually create a development environment, and not
>  | forcing the developer already have one installed ;-)
>
>  Well, yeah. But it'd be nice if you'd have checked if I had MSYS/MinGW
>  installed, like I do. :P
>

Oh sorry, you have GCC installed? ;-)

Well, the idea was bootstrap ruby with the less tools installed possible.

That's why the recipes even have a pure ruby-unzip that download bsdtar :-D

>
>  | Well, actually is more than a real benchmark. As example, the YARV
>  | bench only prove and excersice especific parts of the Ruby VM that has
>  | been optimized, but is not a real life test like PetStore is done for
>  | testing Rails performance.
>
>  Yes and no. It it is a bit of number crunching, but doesn't test things
>  like IO, compression, Arrays, etc. I'm not talking about benchmarks
>  showing off features. ;)
>

Is hard to compare IO processing against different platform and filesystems.
I found that jRuby couldn't get this script "compiled" due the multi
return point... funny.

>
>  | You don't need a lot of GCC fu, just replace the locations for the GCC
>  | packages, check they download, extract and can successfuly 'configure'
>  | and compile Ruby :-)
>
>  Well, that is easy (Search and replace FTW :P). But if something stumps
>  because GCC decides to roll over and die, I'm of no help. ;)
>

Don't worry, we are two :-D

>  | Did you use configure to generate the makefile? if so that means you
>  | need MSYS installed too, don't remember if Giovanni installed provides
>  | it.
>
>  I have a full MSYS environment with GCC4, so of course I used configure. :P
>

Just asked, since that mean is another pre-requisite to get you
started with MinGW.

The original idea was have a exerb "rake.exe" that let you get Ruby
build from scratch. but anyway, the sandbox and the recipes are for
developers wanting to contribute back to One-Click project, and not
creating the nightmare like Gentoo did for packages :-P

>  P.S.: Sent to Rubyinstaller ML, too, in hope of a working subscription
>  now...
>

Yep, it seems is now in the list, great! :-)

Regards and good nights!

-- 
Luis Lavena
Multimedia systems
-
Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so.
Douglas Adams