From tomasz2k at poczta.onet.pl  Tue Nov 10 10:32:02 2009
From: tomasz2k at poczta.onet.pl (tomasz2k at poczta.onet.pl)
Date: Tue, 10 Nov 2009 16:32:02 +0100
Subject: [Celerity-users] I cannot see HtmlUnit debug messages
Message-ID: <Q6557570-7be3f574d739de2c00162a6c5db39c76@pmq4.m5r2.onet.test.onet.pl>

Hello

I cannot get any debug messages from HtmlUnit's packages below. I use this code to set debug level:

    b = Celerity::Browser.new(:refresh_handler => :threaded,
      :resynchronize => true,
      :browser=>:internet_explorer)

    b.log_level= :all

I've also tried with :finest. Only messages I can see are INFO and above. I guess that there should be many DEBUG messages from HtmlUnit but I cannot see any DEBUG message at all. Am I missing something?

Thanks in advance.

Regards,
Tomasz Kalkosinski

From jari.bakken at gmail.com  Tue Nov 10 10:52:15 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Tue, 10 Nov 2009 16:52:15 +0100
Subject: [Celerity-users] I cannot see HtmlUnit debug messages
In-Reply-To: <Q6557570-7be3f574d739de2c00162a6c5db39c76@pmq4.m5r2.onet.test.onet.pl>
References: <Q6557570-7be3f574d739de2c00162a6c5db39c76@pmq4.m5r2.onet.test.onet.pl>
Message-ID: <eae5f6f90911100752p351765ffv251676fecf70333@mail.gmail.com>

On Tue, Nov 10, 2009 at 4:32 PM,  <tomasz2k at poczta.onet.pl> wrote:
>
> I've also tried with :finest. Only messages I can see are INFO and above. I guess that there should be many DEBUG messages from HtmlUnit but I cannot see any DEBUG message at all. Am I missing something?
>

Not sure - try the HtmlUnit list.

From dorikick at gmail.com  Wed Nov 11 06:15:54 2009
From: dorikick at gmail.com (doridori Jo)
Date: Wed, 11 Nov 2009 03:15:54 -0800
Subject: [Celerity-users] click_and_attach question
Message-ID: <3a8841390911110315u592dea76la5883ba95c3c0039@mail.gmail.com>

is it possible to use* click_and_attach* on a link that loads a page via
ajax ? assuming :resynchronize is on, do i add browser.wait immediately
after ?

*newtab = ajaxlink.click_and_attach *



can you stack your *click_and_attach* ? and then close everything by closing
the first browser instance ? I am doing this because* browser.back* is not
reliable in AJAX heavy sites where actually clicking back in your browser
will produce undesired results or simply not go back.

*firsttab = link.click_and_attach

secondtab = firsttab.link.click_and_attach

thirdtab = secondtab.link.click_and_attach

firstab.close* *#would this close all "tabs" opened? or would i need to
close it individually like secondtab.close, thirdtab.close.*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091111/40f1ca11/attachment.html>

From jasoninclass at googlemail.com  Wed Nov 11 15:17:48 2009
From: jasoninclass at googlemail.com (jason franklin-stokes)
Date: Wed, 11 Nov 2009 21:17:48 +0100
Subject: [Celerity-users] thread safe
Message-ID: <6F4F8DE9-47B3-4FB8-A531-13BB6A72EFF9@googlemail.com>

dear all,

i want to use celerity in a number of different threads, however i am  
getting the following error...


/Library/jruby-1.4.0/lib/ruby/site_ruby/shared/builtin/javasupport/ 
core_ext/module.rb:37 warning: already initialized constant  
NicelyResynchronizingAjaxController

/Library/jruby-1.4.0/lib/ruby/site_ruby/shared/builtin/javasupport/ 
core_ext/module.rb:37 warning: already initialized constant  
WebRequestSettings

any ideas???

apparently htmlunit is thread safe s i am wondering what this could be

thanks
jason.


From dorikick at gmail.com  Wed Nov 11 17:22:58 2009
From: dorikick at gmail.com (doridori Jo)
Date: Wed, 11 Nov 2009 14:22:58 -0800
Subject: [Celerity-users] thread safe
In-Reply-To: <6F4F8DE9-47B3-4FB8-A531-13BB6A72EFF9@googlemail.com>
References: <6F4F8DE9-47B3-4FB8-A531-13BB6A72EFF9@googlemail.com>
Message-ID: <3a8841390911111422ode17eabv8b695e42b9765d67@mail.gmail.com>

how did you create multiple threads with celerity ? i am just running
multiple instances.

On Wed, Nov 11, 2009 at 12:17 PM, jason franklin-stokes <
jasoninclass at googlemail.com> wrote:

> dear all,
>
> i want to use celerity in a number of different threads, however i am
> getting the following error...
>
>
> /Library/jruby-1.4.0/lib/ruby/site_ruby/shared/builtin/javasupport/core_ext/module.rb:37
> warning: already initialized constant NicelyResynchronizingAjaxController
>
> /Library/jruby-1.4.0/lib/ruby/site_ruby/shared/builtin/javasupport/core_ext/module.rb:37
> warning: already initialized constant WebRequestSettings
>
> any ideas???
>
> apparently htmlunit is thread safe s i am wondering what this could be
>
> thanks
> jason.
>
> _______________________________________________
> Celerity-users mailing list
> Celerity-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/celerity-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091111/13e42b3e/attachment.html>

From ashley.moran at patchspace.co.uk  Fri Nov 13 14:33:29 2009
From: ashley.moran at patchspace.co.uk (Ashley Moran)
Date: Fri, 13 Nov 2009 19:33:29 +0000
Subject: [Celerity-users] Dojo/JsonRestStore problem with resynchronize and
	PUT requests
Message-ID: <D3462456-C91B-4C1C-AE13-6DE2CBF16BF7@patchspace.co.uk>

Hi

Anyone got any ideas what is going on here?  I'll copy the comment from the source file:

`:resynchronize => true` causes the an 'edit complete' event to fire and attempts to
send a PUT request with an empty body, but an exception is thrown before it leaves Celerity
`:resynchronize => false` apparently prevents the PUT from ever being attempted

It seems like browser resynchronising is affecting the way Dojo gets run.  Presumably an HtmlUnit issue, but I thought I'd ask here first in case anyone can help.

(To run the merb app, you need to do `merb -I json_server.rb`)

Thanks
Ashley

-------------- next part --------------
A non-text attachment was scrubbed...
Name: json_server.rb
Type: text/x-ruby-script
Size: 6495 bytes
Desc: not available
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091113/9cd0f183/attachment-0002.bin>
-------------- next part --------------

-------------- next part --------------
A non-text attachment was scrubbed...
Name: put_test.rb
Type: text/x-ruby-script
Size: 2200 bytes
Desc: not available
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091113/9cd0f183/attachment-0003.bin>
-------------- next part --------------


-- 
http://www.patchspace.co.uk/
http://www.linkedin.com/in/ashleymoran


From dorikick at gmail.com  Fri Nov 13 16:32:26 2009
From: dorikick at gmail.com (doridori Jo)
Date: Fri, 13 Nov 2009 13:32:26 -0800
Subject: [Celerity-users] Dojo/JsonRestStore problem with resynchronize
	and PUT requests
In-Reply-To: <D3462456-C91B-4C1C-AE13-6DE2CBF16BF7@patchspace.co.uk>
References: <D3462456-C91B-4C1C-AE13-6DE2CBF16BF7@patchspace.co.uk>
Message-ID: <3a8841390911131332u23f2a26el364efc27fa169f90@mail.gmail.com>

are you using the latest version of celerity ? 0.7.5

On Fri, Nov 13, 2009 at 11:33 AM, Ashley Moran <
ashley.moran at patchspace.co.uk> wrote:

> Hi
>
> Anyone got any ideas what is going on here?  I'll copy the comment from the
> source file:
>
> `:resynchronize => true` causes the an 'edit complete' event to fire and
> attempts to
> send a PUT request with an empty body, but an exception is thrown before it
> leaves Celerity
> `:resynchronize => false` apparently prevents the PUT from ever being
> attempted
>
> It seems like browser resynchronising is affecting the way Dojo gets run.
>  Presumably an HtmlUnit issue, but I thought I'd ask here first in case
> anyone can help.
>
> (To run the merb app, you need to do `merb -I json_server.rb`)
>
> Thanks
> Ashley
>
>
>
>
>
>
> --
> http://www.patchspace.co.uk/
> http://www.linkedin.com/in/ashleymoran
>
>
> _______________________________________________
> Celerity-users mailing list
> Celerity-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/celerity-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091113/725a0924/attachment.html>

From ashley.moran at patchspace.co.uk  Fri Nov 13 18:05:52 2009
From: ashley.moran at patchspace.co.uk (Ashley Moran)
Date: Fri, 13 Nov 2009 23:05:52 +0000
Subject: [Celerity-users] Dojo/JsonRestStore problem with resynchronize
	and PUT requests
In-Reply-To: <3a8841390911131332u23f2a26el364efc27fa169f90@mail.gmail.com>
References: <D3462456-C91B-4C1C-AE13-6DE2CBF16BF7@patchspace.co.uk>
	<3a8841390911131332u23f2a26el364efc27fa169f90@mail.gmail.com>
Message-ID: <59C91964-09A2-4203-8DD3-5C50A1440CD0@patchspace.co.uk>


On 13 Nov 2009, at 21:32, doridori Jo wrote:

> are you using the latest version of celerity ? 0.7.5

Hi doridori

Yes I am.  Well, client was - I just realised I was on 0.7.4, but I've  
upgraded and still get the same behaviour.

Ashley

-- 
http://www.patchspace.co.uk/
http://www.linkedin.com/in/ashleymoran




From AEtzold at gmx.de  Mon Nov 16 18:57:22 2009
From: AEtzold at gmx.de (Axel Etzold)
Date: Tue, 17 Nov 2009 00:57:22 +0100
Subject: [Celerity-users] A simple naming question - comboboxes ?
Message-ID: <20091116235722.322110@gmx.net>

Dear all,

how are elements like the ones with the text "All Fields" on them
on the website

http://patft.uspto.gov/netahtml/PTO/search-bool.html

named and how could I access them ?

I thought of something like

browser = Celerity::Browser.new
browser.goto("http://patft.uspto.gov/netahtml/PTO/search-bool.html")
browser.comboboxes, 

but there is no such method ...

Thank you very much!

Best regards,

Axel 
-- 
DSL-Preisknaller: DSL Komplettpakete von GMX schon f?r 
16,99 Euro mtl.!* Hier klicken: http://portal.gmx.net/de/go/dsl02

From jari.bakken at gmail.com  Mon Nov 16 19:11:23 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Tue, 17 Nov 2009 01:11:23 +0100
Subject: [Celerity-users] A simple naming question - comboboxes ?
In-Reply-To: <20091116235722.322110@gmx.net>
References: <20091116235722.322110@gmx.net>
Message-ID: <eae5f6f90911161611y7d7c8114xcdeddc4b8e19e5d9@mail.gmail.com>

Try

  browser.select_list(:name, 'FIELD1')

In the future, reading the docs at
http://rdoc.info/projects/jarib/celerity should be helpful for
questions like this, especially the Celerity::Container module and the
various Element subclasses.

On Tue, Nov 17, 2009 at 12:57 AM, Axel Etzold <AEtzold at gmx.de> wrote:
> Dear all,
>
> how are elements like the ones with the text "All Fields" on them
> on the website
>
> http://patft.uspto.gov/netahtml/PTO/search-bool.html
>
> named and how could I access them ?
>
> I thought of something like
>
> browser = Celerity::Browser.new
> browser.goto("http://patft.uspto.gov/netahtml/PTO/search-bool.html")
> browser.comboboxes,
>
> but there is no such method ...
>
> Thank you very much!
>
> Best regards,
>
> Axel
> --
> DSL-Preisknaller: DSL Komplettpakete von GMX schon f?r
> 16,99 Euro mtl.!* Hier klicken: http://portal.gmx.net/de/go/dsl02
> _______________________________________________
> Celerity-users mailing list
> Celerity-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/celerity-users
>

From geoffrey at geoffreyarnold.com  Tue Nov 17 23:36:11 2009
From: geoffrey at geoffreyarnold.com (Geoffrey Arnold)
Date: Tue, 17 Nov 2009 23:36:11 -0500
Subject: [Celerity-users] Downloading dynamic Javascript images
Message-ID: <C0C61205-FEB5-494E-B6D6-F975211F8426@geoffreyarnold.com>

Hello,

We have a page that dynamically downloads images in the background using Javascript similar to the following:

    var image = new Image();
    image.src = '...';

... and are attempting to test it using Cucumber/Celerity, however it appears that the images are not recognized.  Here's a snippet from our Cucumber step:

    @browser.goto_quietly("...")
    @browser.wait

    @browser.wait_until do |browser|
        browser.images.each do |image|
            return false unless image.loaded?
        end
    end

If I print the browser.images property it is empty.  Any help would be greatly appreciated.

Best,
Geoff.

From dorikick at gmail.com  Wed Nov 18 16:56:40 2009
From: dorikick at gmail.com (doridori Jo)
Date: Wed, 18 Nov 2009 13:56:40 -0800
Subject: [Celerity-users] Error: Your application used more memory than the
	safety cap of 500m.Specify -J-Xmx####m to increase it (#### =
	cap size in MB).Specify -w for full OutOfMemoryError stack trace
Message-ID: <3a8841390911181356o3b604549v40e40f30193b8ab6@mail.gmail.com>

Error: Your application used more memory than the safety cap of 500m.
Specify -J-Xmx####m to increase it (#### = cap size in MB).
Specify -w for full OutOfMemoryError stack trace
I get this message. where do i specify the -w option ? jruby -w myscript.rb
?

also, I am using lot of .click_and_attach() which creates many new browser
instances. could this be the problem ?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091118/9c9b55cc/attachment.html>

From dorikick at gmail.com  Sun Nov 22 03:58:44 2009
From: dorikick at gmail.com (doridori Jo)
Date: Sun, 22 Nov 2009 00:58:44 -0800
Subject: [Celerity-users] browser.click_and_attach works but not
	browser.click
Message-ID: <3a8841390911220058y1161e61ndeb0217123193197@mail.gmail.com>

i found a very weird situation today where

puts browser.url # => something.com
> puts browser.element_by_xpath(somelink).click_and_attach.url  #=>
> something.com/new/
>

works as it should.

i deleted above, and decided to just use .click as below.

puts browser.url # => something.com
> browser.element_by_xpath(somelink).click
> browser.wait
> puts browser.url # => something.com
>

no exception is raised, meaning it clicked it right? then, why is it that
something.com/new/ is not loading ?

what does this mean ?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091122/9fe6248e/attachment.html>

From jari.bakken at gmail.com  Sun Nov 22 10:13:18 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Sun, 22 Nov 2009 16:13:18 +0100
Subject: [Celerity-users] browser.click_and_attach works but not
	browser.click
In-Reply-To: <3a8841390911220058y1161e61ndeb0217123193197@mail.gmail.com>
References: <3a8841390911220058y1161e61ndeb0217123193197@mail.gmail.com>
Message-ID: <eae5f6f90911220713h95aacf3id185ba7e7fabb2c6@mail.gmail.com>

On Sun, Nov 22, 2009 at 9:58 AM, doridori Jo <dorikick at gmail.com> wrote:
>
> what does this mean ?
>

It's impossible to say without knowing more about the web app. If you
can provide a way for us to reproduce the problem locally, we might be
able to help further.

Following the instructions on
http://wiki.github.com/jarib/celerity/ajax, especially turning on
javascript exceptions and full logging, might give you further
insights.

From ben at bridesmere.com  Wed Nov 25 10:37:34 2009
From: ben at bridesmere.com (Ben Butler-Cole)
Date: Wed, 25 Nov 2009 15:37:34 +0000
Subject: [Celerity-users] Magic transmogrifying row object
Message-ID: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>

Hello

I need to get at the underlying HtmlUnit object for a table row. When
I first tried it I discovered that it didn't have the methods I was
expecting. By trial and error I have discovered that the first time I
access the row's HtmlUnit object (via the .object method) it returns a
List. The second time it returns an HtmlTableRow. Ummm????

Here is the log of a real, live irb session where you can see this
happing. As a bonus we learn that Google's front page has tables in
it.

irb(main):019:0> browser = Celerity::Browser.new
irb(main):020:0> browser.goto('http://www.google.com')
irb(main):021:0> e = browser.row(:index, 1)

irb(main):022:0> puts e.object.getClass
class java.util.Collections$UnmodifiableRandomAccessList

irb(main):023:0> puts e.object.getClass
class com.gargoylesoftware.htmlunit.html.HtmlTableRow

I can live with this because a random e.object in my code fixes the
problem. But any idea what's going on? Shall I raise a bug?

Thanks
Ben

From jari.bakken at gmail.com  Wed Nov 25 11:24:51 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Wed, 25 Nov 2009 17:24:51 +0100
Subject: [Celerity-users] Magic transmogrifying row object
In-Reply-To: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>
References: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>
Message-ID: <DA8A5D2D-213C-4900-924A-47486D403854@gmail.com>


Den 25. nov. 2009 kl. 16.37 skrev Ben Butler-Cole <ben at bridesmere.com>:

> Hello
>
> I need to get at the underlying HtmlUnit object for a table row. When
> I first tried it I discovered that it didn't have the methods I was
> expecting. By trial and error I have discovered that the first time I
> access the row's HtmlUnit object (via the .object method) it returns a
> List. The second time it returns an HtmlTableRow. Ummm????


Without having the code in front of me, I would guess this is caused  
by the lazy locating of elements. If you try row(...).locate you  
should get the correct HU object directly.

Out of interest, why do you need the underlying object?

>
> Here is the log of a real, live irb session where you can see this
> happing. As a bonus we learn that Google's front page has tables in
> it.
>
> irb(main):019:0> browser = Celerity::Browser.new
> irb(main):020:0> browser.goto('http://www.google.com')
> irb(main):021:0> e = browser.row(:index, 1)
>
> irb(main):022:0> puts e.object.getClass
> class java.util.Collections$UnmodifiableRandomAccessList
>
> irb(main):023:0> puts e.object.getClass
> class com.gargoylesoftware.htmlunit.html.HtmlTableRow
>
> I can live with this because a random e.object in my code fixes the
> problem. But any idea what's going on? Shall I raise a bug?
>
> Thanks
> Ben
> _______________________________________________
> Celerity-users mailing list
> Celerity-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/celerity-users

From ben at bridesmere.com  Wed Nov 25 11:59:56 2009
From: ben at bridesmere.com (Ben Butler-Cole)
Date: Wed, 25 Nov 2009 16:59:56 +0000
Subject: [Celerity-users] Magic transmogrifying row object
In-Reply-To: <DA8A5D2D-213C-4900-924A-47486D403854@gmail.com>
References: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>
	<DA8A5D2D-213C-4900-924A-47486D403854@gmail.com>
Message-ID: <fecee110911250859n4fcc2218h81380e86487f365@mail.gmail.com>

2009/11/25 Jari Bakken <jari.bakken at gmail.com>:
>
>> I need to get at the underlying HtmlUnit object for a table row. When
>> I first tried it I discovered that it didn't have the methods I was
>> expecting. By trial and error I have discovered that the first time I
>> access the row's HtmlUnit object (via the .object method) it returns a
>> List. The second time it returns an HtmlTableRow. Ummm????
>
> Without having the code in front of me, I would guess this is caused by the
> lazy locating of elements. If you try row(...).locate you should get the
> correct HU object directly.

Calling .locate *always* gives the wrong type (the list instead of the
HtmlTableRow).

I have put the code in front of myself and had a look. I think there
is a bug in Celerity::TableRow.locate. The relevant bits look like
this:

module Celerity
  class Element
    def locate
      @object = ElementLocator.new(@container,
self.class).find_by_conditions(@conditions)
    end

    def object
      @object || locate
    end
  end

  class TableRow < Element
    def locate
      super
      @cells = @object.getCells if @object
    end
  end
end

So the .object method in Element expects .locate to return the object
as well as assign it. But TableRow's overridden .locate returns the
cells instead. The easiest fix is:

module Celerity
  class TableRow
    def locate
      super
      @cells = @object.getCells if @object
      @object
    end
  end
end

Which I have verified locally.

> Out of interest, why do you need the underlying object?

I want to call the mouse* methods. I've added this:

class Celerity::Element
  def mouse_down
    object.mouseDown
  end
  def mouse_up
    object.mouseUp
  end
  def mouse_move
    object.mouseMove
  end
end

Ben

From jari.bakken at gmail.com  Wed Nov 25 12:47:44 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Wed, 25 Nov 2009 18:47:44 +0100
Subject: [Celerity-users] Magic transmogrifying row object
In-Reply-To: <fecee110911250859n4fcc2218h81380e86487f365@mail.gmail.com>
References: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>
	<DA8A5D2D-213C-4900-924A-47486D403854@gmail.com>
	<fecee110911250859n4fcc2218h81380e86487f365@mail.gmail.com>
Message-ID: <F45D020E-BA59-4D81-99D5-9801C4E4A1DF@gmail.com>


On 25. nov. 2009, at 17.59, Ben Butler-Cole wrote:

> 2009/11/25 Jari Bakken <jari.bakken at gmail.com>:
>> 
>>> I need to get at the underlying HtmlUnit object for a table row. When
>>> I first tried it I discovered that it didn't have the methods I was
>>> expecting. By trial and error I have discovered that the first time I
>>> access the row's HtmlUnit object (via the .object method) it returns a
>>> List. The second time it returns an HtmlTableRow. Ummm????
>> 
>> Without having the code in front of me, I would guess this is caused by the
>> lazy locating of elements. If you try row(...).locate you should get the
>> correct HU object directly.
> 
> Calling .locate *always* gives the wrong type (the list instead of the
> HtmlTableRow).
> 
> I have put the code in front of myself and had a look. I think there
> is a bug in Celerity::TableRow.locate. The relevant bits look like
> this:
> 
> module Celerity
>  class Element
>    def locate
>      @object = ElementLocator.new(@container,
> self.class).find_by_conditions(@conditions)
>    end
> 
>    def object
>      @object || locate
>    end
>  end
> 
>  class TableRow < Element
>    def locate
>      super
>      @cells = @object.getCells if @object
>    end
>  end
> end
> 
> So the .object method in Element expects .locate to return the object
> as well as assign it. But TableRow's overridden .locate returns the
> cells instead. The easiest fix is:
> 
> module Celerity
>  class TableRow
>    def locate
>      super
>      @cells = @object.getCells if @object
>      @object
>    end
>  end
> end
> 
> Which I have verified locally.

Nice work. Could you fork Celerity on GitHub and commit this fix?




From jari.bakken at gmail.com  Wed Nov 25 13:24:15 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Wed, 25 Nov 2009 19:24:15 +0100
Subject: [Celerity-users] Magic transmogrifying row object
In-Reply-To: <fecee110911250859n4fcc2218h81380e86487f365@mail.gmail.com>
References: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>
	<DA8A5D2D-213C-4900-924A-47486D403854@gmail.com>
	<fecee110911250859n4fcc2218h81380e86487f365@mail.gmail.com>
Message-ID: <FC1FEA7A-87B5-44EE-BE47-CC163AF462B9@gmail.com>


On 25. nov. 2009, at 17.59, Ben Butler-Cole wrote:

> 2009/11/25 Jari Bakken <jari.bakken at gmail.com>:
>> 
>>> I need to get at the underlying HtmlUnit object for a table row. When
>>> I first tried it I discovered that it didn't have the methods I was
>>> expecting. By trial and error I have discovered that the first time I
>>> access the row's HtmlUnit object (via the .object method) it returns a
>>> List. The second time it returns an HtmlTableRow. Ummm????
>> 
>> Without having the code in front of me, I would guess this is caused by the
>> lazy locating of elements. If you try row(...).locate you should get the
>> correct HU object directly.
> 
> Calling .locate *always* gives the wrong type (the list instead of the
> HtmlTableRow).
> 
> I have put the code in front of myself and had a look. I think there
> is a bug in Celerity::TableRow.locate. The relevant bits look like
> this:
> 
> module Celerity
>  class Element
>    def locate
>      @object = ElementLocator.new(@container,
> self.class).find_by_conditions(@conditions)
>    end
> 
>    def object
>      @object || locate
>    end
>  end
> 
>  class TableRow < Element
>    def locate
>      super
>      @cells = @object.getCells if @object
>    end
>  end
> end
> 
> So the .object method in Element expects .locate to return the object
> as well as assign it. But TableRow's overridden .locate returns the
> cells instead. The easiest fix is:
> 
> module Celerity
>  class TableRow
>    def locate
>      super
>      @cells = @object.getCells if @object
>      @object
>    end
>  end
> end
> 
> Which I have verified locally.

Could you fork Celerity on GitHub and commit this fix?



From ben at bridesmere.com  Wed Nov 25 14:53:51 2009
From: ben at bridesmere.com (Ben Butler-Cole)
Date: Wed, 25 Nov 2009 19:53:51 +0000
Subject: [Celerity-users] Magic transmogrifying row object
In-Reply-To: <F45D020E-BA59-4D81-99D5-9801C4E4A1DF@gmail.com>
References: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>
	<DA8A5D2D-213C-4900-924A-47486D403854@gmail.com>
	<fecee110911250859n4fcc2218h81380e86487f365@mail.gmail.com>
	<F45D020E-BA59-4D81-99D5-9801C4E4A1DF@gmail.com>
Message-ID: <fecee110911251153g4970b482kfc3ce3a8df870a28@mail.gmail.com>

>> I think there is a bug in Celerity::TableRow.locate.
>
> Nice work. Could you fork Celerity on GitHub and commit this fix?

Will do. Where is the right place to put a test for this? I can't see
anywhere obvious.

Ben

From jari.bakken at gmail.com  Wed Nov 25 16:05:04 2009
From: jari.bakken at gmail.com (Jari Bakken)
Date: Wed, 25 Nov 2009 22:05:04 +0100
Subject: [Celerity-users] Magic transmogrifying row object
In-Reply-To: <fecee110911251153g4970b482kfc3ce3a8df870a28@mail.gmail.com>
References: <fecee110911250737m385bf7d9v9eca803acde29c86@mail.gmail.com>
	<DA8A5D2D-213C-4900-924A-47486D403854@gmail.com>
	<fecee110911250859n4fcc2218h81380e86487f365@mail.gmail.com>
	<F45D020E-BA59-4D81-99D5-9801C4E4A1DF@gmail.com>
	<fecee110911251153g4970b482kfc3ce3a8df870a28@mail.gmail.com>
Message-ID: <DA7CC2F1-8D0D-45A2-9E5D-F2607A379CF6@gmail.com>

The specs are in a separate project (watirspec), included as a git  
submodule. Check the rake tasks.

Den 25. nov. 2009 kl. 20.53 skrev Ben Butler-Cole <ben at bridesmere.com>:

>>> I think there is a bug in Celerity::TableRow.locate.
>>
>> Nice work. Could you fork Celerity on GitHub and commit this fix?
>
> Will do. Where is the right place to put a test for this? I can't see
> anywhere obvious.
>
> Ben
> _______________________________________________
> Celerity-users mailing list
> Celerity-users at rubyforge.org
> http://rubyforge.org/mailman/listinfo/celerity-users

From dorikick at gmail.com  Thu Nov 26 00:39:19 2009
From: dorikick at gmail.com (doridori Jo)
Date: Wed, 25 Nov 2009 21:39:19 -0800
Subject: [Celerity-users] browser.xml.to_s seems to consume lot of resource ?
Message-ID: <3a8841390911252139g3adf8381ide6440969da22a9@mail.gmail.com>

i need to display the HTML source, and have it written to disk.

currently i am doing

browser.xml.to_s and writing it to a text file.

is this efficient or is there a better approach ? I am trying to just
download what celerity::browser sees.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091125/3b8dec49/attachment.html>

From dorikick at gmail.com  Sat Nov 28 21:36:28 2009
From: dorikick at gmail.com (doridori Jo)
Date: Sat, 28 Nov 2009 18:36:28 -0800
Subject: [Celerity-users] Celerity renders site differently than firefox
Message-ID: <3a8841390911281836i5288e1f4me251e96bcc48426@mail.gmail.com>

I discovered today that in my firefox browser, a link is

/html/body/center/p[2]/table/tbody/tr/td[2]/small/a

but when celerity loads the site, to my surprise, it couldn't locate it via
XPATH. curious i decided to

puts browser.link(:text => "Someuniquelink").xpath

/html/body/p[3]/table/tbody/tr/td[2]/small/a


What gives ???!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091128/3e821a5f/attachment-0001.html>

From dorikick at gmail.com  Mon Nov 30 15:45:23 2009
From: dorikick at gmail.com (doridori Jo)
Date: Mon, 30 Nov 2009 12:45:23 -0800
Subject: [Celerity-users] Celerity opening pop up ads !
Message-ID: <3a8841390911301245ka4438ebn37bcabe60004c274@mail.gmail.com>

i discover that when i submit a form through celerity, and open the
resulting page in new browser session via #click_and_attach,

the new browser session contains an advertisement and not the expected
result page !

what to do ? I can't ignore all popunder because the app depends on it as
well (for setting certain form options).
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rubyforge.org/pipermail/celerity-users/attachments/20091130/e0cd00ff/attachment.html>