From jari.bakken at gmail.com  Thu Jun  4 21:51:16 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Fri, 5 Jun 2009 03:51:16 +0200
Subject: [Celerity-users] Show hide Divs
In-Reply-To: <F15BB23A-3F4E-4118-B00A-7C89991A5345@dinshawdesign.com>
References: <412BB2DA-C1D9-4755-B0CB-8DBD0FFC94F7@dinshawdesign.com>
	<D55537E8-3532-4174-A32C-B8E0FA59B368@patchspace.co.uk>
	<283B4C38-F263-4DA9-A01F-C3B112553A85@dinshawdesign.com>
	<74948418-B187-457A-9444-8F4F1E4F05E7@gmail.com>
	<F15BB23A-3F4E-4118-B00A-7C89991A5345@dinshawdesign.com>
Message-ID: <7B870F87-E4BD-46EC-8512-AD2579EB6D62@gmail.com>

On 4. juni. 2009, at 22.16, Dinshaw Gobhai wrote:

>
> I have tried:
> @browser.div(:id, 'foo').visible?

This should return false if the div is hidden. If not, can you tell me  
how your div is styled when in the hidden state?

> @browser.text.should_not =~ /Hidden form text/

Same as above.

> @browser.html.should_not =~ /Hidden form text/

@browser.html returns the HTML response from the server (before any JS  
changes). So if this fails, then that text _does_ exist before any  
scripts are executed.

> @browser.div(:id, 'foo').should_not =~ /Hidden form text/

This expectation should pass, since Celerity::Element does not  
override Object#=~ (which will always return false).
This should pass as well:

   @browser.div(:id, 'foo').should_not =~ /Hidden form text/

> And even Hipricot parsing the output of @browser.html
>
> Did I miss something?
>

Hard to tell what's going on without seeing an example that will  
reproduce this.
And as always, try installing the latest version of Celerity from  
GitHub to see if it resolves your issue.

jb

From jari.bakken at gmail.com  Fri Jun  5 17:04:48 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Fri, 5 Jun 2009 23:04:48 +0200
Subject: [Celerity-users] Celerity and LiveQuery
In-Reply-To: <6E35EC59-D8F9-40F8-8205-C7E79ACE3E68@dinshawdesign.com>
References: <412BB2DA-C1D9-4755-B0CB-8DBD0FFC94F7@dinshawdesign.com>
	<D55537E8-3532-4174-A32C-B8E0FA59B368@patchspace.co.uk>
	<283B4C38-F263-4DA9-A01F-C3B112553A85@dinshawdesign.com>
	<74948418-B187-457A-9444-8F4F1E4F05E7@gmail.com>
	<F15BB23A-3F4E-4118-B00A-7C89991A5345@dinshawdesign.com>
	<7B870F87-E4BD-46EC-8512-AD2579EB6D62@gmail.com>
	<6E35EC59-D8F9-40F8-8205-C7E79ACE3E68@dinshawdesign.com>
Message-ID: <eae5f6f90906051404q512334fcmc02432d8e09fa4b7@mail.gmail.com>

On Fri, Jun 5, 2009 at 10:44 PM, Dinshaw Gobhai <info at dinshawdesign.com>wrote:

> has anyone been testing any thing with LiveQuery (
> http://docs.jquery.com/Plugins/livequery) ?
> I have some forms that are replaced in the dom each time the form is
> submitted (successfully) and the "re-hooked" by LiveQuery to submit async.
> The LIveQuery hooks do not seem to be working and Culerity is posting the
> forms synchronously.
>
> I have tried setting :resynchronize => true even though I am not really
> sure what it does, but it didn't work.
>

Did you see my response to your previous email?

http://rubyforge.org/pipermail/celerity-users/2009-June/000083.html

(Not sure why your messages won't show up in the archive - anyone know
anything about that)?

I haven't used livequery, so can't help you unless you can show me some code
that will reproduce the issue.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20090605/763a934a/attachment.html>

From jari.bakken at gmail.com  Fri Jun  5 18:56:49 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Sat, 6 Jun 2009 00:56:49 +0200
Subject: [Celerity-users] Celerity and LiveQuery
In-Reply-To: <07787991-7D32-4B88-9B8B-E888FD483A43@dinshawdesign.com>
References: <412BB2DA-C1D9-4755-B0CB-8DBD0FFC94F7@dinshawdesign.com>
	<D55537E8-3532-4174-A32C-B8E0FA59B368@patchspace.co.uk>
	<283B4C38-F263-4DA9-A01F-C3B112553A85@dinshawdesign.com>
	<74948418-B187-457A-9444-8F4F1E4F05E7@gmail.com>
	<F15BB23A-3F4E-4118-B00A-7C89991A5345@dinshawdesign.com>
	<7B870F87-E4BD-46EC-8512-AD2579EB6D62@gmail.com>
	<6E35EC59-D8F9-40F8-8205-C7E79ACE3E68@dinshawdesign.com>
	<eae5f6f90906051404q512334fcmc02432d8e09fa4b7@mail.gmail.com>
	<07787991-7D32-4B88-9B8B-E888FD483A43@dinshawdesign.com>
Message-ID: <eae5f6f90906051556x263605a2ya88a175b0ef6d3f6@mail.gmail.com>

On Sat, Jun 6, 2009 at 12:09 AM, Dinshaw Gobhai <info at dinshawdesign.com>wrote:

>
> The really strange part is that even if i wait, refresh the page
> (@browser.refresh), hit the page again (@browser.goto('
> http://localhost:3001/tasks'), or navigate away and then come back to the
> page, it still fails to post asynchronously.
>

I guess by "fails to post asynchronously", you mean it does a full browser
HTTP request?


> This is the first time i am trying to post a form twice in the
> same feature . Any thoughts would be appreciated.
>

If you could create a small example that will reproduce the problem, it
would be a lot easier to help.
You can try enabling JavaScript execptions and turn on full logging:

  Celerity::Browser.new(:javascript_exceptions => true, :log_level => :all)

The :resynchronize => true setting makes sure that everything happens
synchronously - so you don't have to explicitly wait for Ajax calls to
finish. It's recommended whenever you have Ajax in your app.

Is there some sort of cache anywhere in the Celerity @browser?
>

Yes, HtmlUnit will cache "compiled JavaScript files and parsed CSS
snippets". You can clear this cache by calling

  @browser.clear_cache

on Celerity versions >= 0.0.6.4
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20090606/6cd238d4/attachment.html>

From jari.bakken at gmail.com  Tue Jun  9 17:28:10 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Tue, 9 Jun 2009 23:28:10 +0200
Subject: [Celerity-users] Logger / js output for Rails and Culerity
In-Reply-To: <c948d2470906091127q12f846d2rd439035d1d967114@mail.gmail.com>
References: <c948d2470906091127q12f846d2rd439035d1d967114@mail.gmail.com>
Message-ID: <517D9075-347A-40E3-B78D-170054145327@gmail.com>


On 9. juni. 2009, at 20.27, Dinshaw Gobhai wrote:

> Is there a way to do do some js debugging in a celerity session?
> I found the HtmlUnit log docs but don't really see how to use it  
> through Rails and Culerity.

My best suggestion is using alert() and in your Ruby code do

   browser.add_listener(:alert) { |page, message| ... }

That will probably get quite annoying when browsing the app manually  
though. It would be great to be able to do console.debug(...) in  
JavaScript and have HtmlUnit expose that log somewhere. However, http://htmlunit.sourceforge.net/javascript.html 
  says:

> Testing pages with JavaScript on them can produce a number of  
> errors. These errors occur for several reasons. Some browsers accept  
> shortcuts that are not part of the DOM model and currently not  
> duplicated by HtmlUnit. Also, HtmlUnit still does not implement all  
> of the extensions that have been added by the different browsers.

If you can convince someone to add the console object to HtmlUnit's JS  
implementation, I'll be happy to add support for it in Celerity. Their  
mailing lists are at http://htmlunit.sourceforge.net/mail-lists.html :)



From info at dinshawdesign.com  Fri Jun 12 18:12:10 2009
From: info at dinshawdesign.com (Dinshaw Gobhai)
Date: Fri, 12 Jun 2009 18:12:10 -0400
Subject: [Celerity-users] Sessions and subdomains
Message-ID: <c948d2470906121512o405c7747kf22120799716e25b@mail.gmail.com>

Seems that after setting a session for example.com, the @browser
cannot set a session for sub.example.com.

Is this 'works-as-designed'? or maybe a problem with my implementation?

I am doing features for a registration wizard and ended up having to
stop one scenario after my wizard redirected to a subdomain, and start
a new one.

We are using standard Rails session setters (eg session[:user_id] = 1)
and cookie based storage.

Thanks,
-Dinshaw

From jari.bakken at gmail.com  Fri Jun 12 18:44:39 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Sat, 13 Jun 2009 00:44:39 +0200
Subject: [Celerity-users] Sessions and subdomains
In-Reply-To: <c948d2470906121512o405c7747kf22120799716e25b@mail.gmail.com>
References: <c948d2470906121512o405c7747kf22120799716e25b@mail.gmail.com>
Message-ID: <674AC4B8-770B-448F-905D-EDF01DC96669@gmail.com>


On 13. juni. 2009, at 00.12, Dinshaw Gobhai wrote:

> Seems that after setting a session for example.com, the @browser
> cannot set a session for sub.example.com.
>
> Is this 'works-as-designed'? or maybe a problem with my  
> implementation?
>

Celerity/HtmlUnit should work like normal browsers.

> I am doing features for a registration wizard and ended up having to
> stop one scenario after my wizard redirected to a subdomain, and start
> a new one.
>
> We are using standard Rails session setters (eg session[:user_id] = 1)
> and cookie based storage.
>

If you can provide an example that will reproduce the issue, I'll be  
happy to take a look.


From info at dinshawdesign.com  Mon Jun 15 23:21:55 2009
From: info at dinshawdesign.com (Dinshaw Gobhai)
Date: Mon, 15 Jun 2009 23:21:55 -0400
Subject: [Celerity-users] ActionMailer
Message-ID: <c948d2470906152021l54e89053y734241b3d831fa23@mail.gmail.com>

I am running into some strange results trying to test emails.
Is there anything diferent about how one should treat mailer tests
when using Culerity?

I am using the email_spec plugin and the method described for Cucumber
but, inside the Celerity session i always get
ActionMailer::Base.deliveries.size == 0.
If I dubug and actually send an email
(UserMailer.deliver_invite(user)), then I see it in the
ActionMailer::Base.deliveries.

The log shows the emails being sent normally.

P.S. Thanks for you help on the other emails I have sent, Jari. It
seems that the app I am working on has some non-standard
implementations that I am just getting to know so please seemingly
crazy questions.

-Dinshaw