From peter at hexagile.com  Fri Oct 28 05:33:36 2011
From: peter at hexagile.com (Peter Szinek)
Date: Fri, 28 Oct 2011 11:33:36 +0200
Subject: [Celerity-users] wrong rendering of a page
Message-ID: <CACDiUNqHf9O4-0a3eVdBKsV+xgxJY1DcpN2Fohp08sxUQGkk4g@mail.gmail.com>

Hey guys,

Tried this without javascript, as well as :resynchronize => true (not
sure it should make any difference in this case - well it doesn't)

browser = Celerity::Browser.new(:user_agent =>'Mozilla/4.0
(compatible; MSIE 7.0b; Windows NT 6.0)')
browser.goto "http://careers.peopleclick.com/careerscp/client_motorola/external/search.do"
browser.element_by_xpath("//input[@id='searchButton']").exists?
 => false

That button should definitely be there (checked in a real browser, and
even with mechanize) - I know for a fact that until recently the above
code returned true, since we are having this scrape around since
months if not years, and it was working. The strange thing is the
above code is happening with 0.8.9, 0.9.0 as well as 0.9.1... maybe
jruby-1.6.3 is the culprit? Need to check this with jruby-1.6.4

Help!

Cheers,
Peter

From peter at hexagile.com  Fri Oct 28 06:37:52 2011
From: peter at hexagile.com (Peter Szinek)
Date: Fri, 28 Oct 2011 12:37:52 +0200
Subject: [Celerity-users] wrong rendering of a page
In-Reply-To: <CACDiUNqHf9O4-0a3eVdBKsV+xgxJY1DcpN2Fohp08sxUQGkk4g@mail.gmail.com>
References: <CACDiUNqHf9O4-0a3eVdBKsV+xgxJY1DcpN2Fohp08sxUQGkk4g@mail.gmail.com>
Message-ID: <CACDiUNrJitTKujOWS7BnRjFjv34uKB61SKUMmRi6Fr3O_Axjfw@mail.gmail.com>

Somewhat better with juby-1.6.5 - at least now when I examine
browser.xml (by outputting it to a html file and opening that in a
browser) the button is visible (with jruby-1.6.3 it's not even visible
in the browser). However,

browser.element_by_xpath("//input[@id='searchButton']").exists?

still returns false...

On Fri, Oct 28, 2011 at 11:33 AM, Peter Szinek <peter at hexagile.com> wrote:
> Hey guys,
>
> Tried this without javascript, as well as :resynchronize => true (not
> sure it should make any difference in this case - well it doesn't)
>
> browser = Celerity::Browser.new(:user_agent =>'Mozilla/4.0
> (compatible; MSIE 7.0b; Windows NT 6.0)')
> browser.goto "http://careers.peopleclick.com/careerscp/client_motorola/external/search.do"
> browser.element_by_xpath("//input[@id='searchButton']").exists?
> ?=> false
>
> That button should definitely be there (checked in a real browser, and
> even with mechanize) - I know for a fact that until recently the above
> code returned true, since we are having this scrape around since
> months if not years, and it was working. The strange thing is the
> above code is happening with 0.8.9, 0.9.0 as well as 0.9.1... maybe
> jruby-1.6.3 is the culprit? Need to check this with jruby-1.6.4
>
> Help!
>
> Cheers,
> Peter
>

From peter at hexagile.com  Fri Oct 28 08:10:11 2011
From: peter at hexagile.com (Peter Szinek)
Date: Fri, 28 Oct 2011 14:10:11 +0200
Subject: [Celerity-users] wrong rendering of a page
In-Reply-To: <CACDiUNrJitTKujOWS7BnRjFjv34uKB61SKUMmRi6Fr3O_Axjfw@mail.gmail.com>
References: <CACDiUNqHf9O4-0a3eVdBKsV+xgxJY1DcpN2Fohp08sxUQGkk4g@mail.gmail.com>
	<CACDiUNrJitTKujOWS7BnRjFjv34uKB61SKUMmRi6Fr3O_Axjfw@mail.gmail.com>
Message-ID: <CACDiUNot88FoRqAL65OrDNpcupBVFNVpfp0rEDqrRrF3Geu0GA@mail.gmail.com>

OK so this is getting really weird here.

browser.xml contains

<input "" type="submit" name="PARAMFILTER:functionName=resetCriteria|"
value="Reset" class="pc-rtg-submitButton" onclick="return
submitOnce();">

but I can't get it with an XPath.

also stuff like

like browser.elements_by_xpath("//input").map{|x|x.id}

doesn't reveal the id either.

I'm wondering whether the "" after the input tag is screwing things up....

On Fri, Oct 28, 2011 at 12:37 PM, Peter Szinek <peter at hexagile.com> wrote:
> Somewhat better with juby-1.6.5 - at least now when I examine
> browser.xml (by outputting it to a html file and opening that in a
> browser) the button is visible (with jruby-1.6.3 it's not even visible
> in the browser). However,
>
> browser.element_by_xpath("//input[@id='searchButton']").exists?
>
> still returns false...
>
> On Fri, Oct 28, 2011 at 11:33 AM, Peter Szinek <peter at hexagile.com> wrote:
>> Hey guys,
>>
>> Tried this without javascript, as well as :resynchronize => true (not
>> sure it should make any difference in this case - well it doesn't)
>>
>> browser = Celerity::Browser.new(:user_agent =>'Mozilla/4.0
>> (compatible; MSIE 7.0b; Windows NT 6.0)')
>> browser.goto "http://careers.peopleclick.com/careerscp/client_motorola/external/search.do"
>> browser.element_by_xpath("//input[@id='searchButton']").exists?
>> ?=> false
>>
>> That button should definitely be there (checked in a real browser, and
>> even with mechanize) - I know for a fact that until recently the above
>> code returned true, since we are having this scrape around since
>> months if not years, and it was working. The strange thing is the
>> above code is happening with 0.8.9, 0.9.0 as well as 0.9.1... maybe
>> jruby-1.6.3 is the culprit? Need to check this with jruby-1.6.4
>>
>> Help!
>>
>> Cheers,
>> Peter
>>
>

From peter at hexagile.com  Fri Oct 28 08:24:07 2011
From: peter at hexagile.com (Peter Szinek)
Date: Fri, 28 Oct 2011 14:24:07 +0200
Subject: [Celerity-users] wrong rendering of a page
In-Reply-To: <CACDiUNot88FoRqAL65OrDNpcupBVFNVpfp0rEDqrRrF3Geu0GA@mail.gmail.com>
References: <CACDiUNqHf9O4-0a3eVdBKsV+xgxJY1DcpN2Fohp08sxUQGkk4g@mail.gmail.com>
	<CACDiUNrJitTKujOWS7BnRjFjv34uKB61SKUMmRi6Fr3O_Axjfw@mail.gmail.com>
	<CACDiUNot88FoRqAL65OrDNpcupBVFNVpfp0rEDqrRrF3Geu0GA@mail.gmail.com>
Message-ID: <CACDiUNp1mj7Xjg8FH2XnoCG+jZJXYe-T1Fy8BQQeC19x+qrrzw@mail.gmail.com>

OK so this is as far as I got:

jruby-1.6.5 :042 >
browser.element_by_xpath("//div[@id='pc-rtg-lastCriteria']").html
 => "<div id=\"pc-rtg-lastCriteria\">\n  <table width=\"100%\"
border=\"0\" cellspacing=\"0\" cellpadding=\"0\">\n    <tbody
align=\"left\">\n      <tr valign=\"top\">\n        <td
colspan=\"3\"/>\n      </tr>\n      <tr valign=\"bottom\">\n
<td width=\"400px\">\n          \n                        Results per
page:\302\240\n\n    \n          <select class=\"\" size=\"1\"
name=\"com.peopleclick.cp.formdata.hitsPerPage\"
id=\"com.peopleclick.cp.formdata.hitsPerPage\">\n            <option
value=\"10\" selected=\"selected\">\n              10\n
</option>\n            <option value=\"20\">\n              20\n
     </option>\n            <option value=\"30\">\n              30\n
          </option>\n            <option value=\"50\">\n
50\n            </option>\n          </select>\n        </td>\n
<td align=\"right\" width=\"75px\" class=\"pc-rtg-searchButton\">\n
      <input type=\"text\"/>\n        </td>\n        <td
align=\"right\" width=\"75px\" class=\"pc-rtg-searchButton\">\n
  <input type=\"text\"/>\n        </td>\n      </tr>\n    </tbody>\n
</table>\n</div>\n"

As you can see the two inputs are there, but they are detected as
"text" and also missing all the other attributes. I think this might
be exactly because the  "" after the input tag.

On Fri, Oct 28, 2011 at 2:10 PM, Peter Szinek <peter at hexagile.com> wrote:
> OK so this is getting really weird here.
>
> browser.xml contains
>
> <input "" type="submit" name="PARAMFILTER:functionName=resetCriteria|"
> value="Reset" class="pc-rtg-submitButton" onclick="return
> submitOnce();">
>
> but I can't get it with an XPath.
>
> also stuff like
>
> like browser.elements_by_xpath("//input").map{|x|x.id}
>
> doesn't reveal the id either.
>
> I'm wondering whether the "" after the input tag is screwing things up....
>
> On Fri, Oct 28, 2011 at 12:37 PM, Peter Szinek <peter at hexagile.com> wrote:
>> Somewhat better with juby-1.6.5 - at least now when I examine
>> browser.xml (by outputting it to a html file and opening that in a
>> browser) the button is visible (with jruby-1.6.3 it's not even visible
>> in the browser). However,
>>
>> browser.element_by_xpath("//input[@id='searchButton']").exists?
>>
>> still returns false...
>>
>> On Fri, Oct 28, 2011 at 11:33 AM, Peter Szinek <peter at hexagile.com> wrote:
>>> Hey guys,
>>>
>>> Tried this without javascript, as well as :resynchronize => true (not
>>> sure it should make any difference in this case - well it doesn't)
>>>
>>> browser = Celerity::Browser.new(:user_agent =>'Mozilla/4.0
>>> (compatible; MSIE 7.0b; Windows NT 6.0)')
>>> browser.goto "http://careers.peopleclick.com/careerscp/client_motorola/external/search.do"
>>> browser.element_by_xpath("//input[@id='searchButton']").exists?
>>> ?=> false
>>>
>>> That button should definitely be there (checked in a real browser, and
>>> even with mechanize) - I know for a fact that until recently the above
>>> code returned true, since we are having this scrape around since
>>> months if not years, and it was working. The strange thing is the
>>> above code is happening with 0.8.9, 0.9.0 as well as 0.9.1... maybe
>>> jruby-1.6.3 is the culprit? Need to check this with jruby-1.6.4
>>>
>>> Help!
>>>
>>> Cheers,
>>> Peter
>>>
>>
>

From jari.bakken at gmail.com  Fri Oct 28 22:21:25 2011
From: jari.bakken at gmail.com (Jari Bakken)
Date: Fri, 28 Oct 2011 19:21:25 -0700
Subject: [Celerity-users] wrong rendering of a page
In-Reply-To: <CACDiUNrJitTKujOWS7BnRjFjv34uKB61SKUMmRi6Fr3O_Axjfw@mail.gmail.com>
References: <CACDiUNqHf9O4-0a3eVdBKsV+xgxJY1DcpN2Fohp08sxUQGkk4g@mail.gmail.com>
	<CACDiUNrJitTKujOWS7BnRjFjv34uKB61SKUMmRi6Fr3O_Axjfw@mail.gmail.com>
Message-ID: <CAEMjNnLHCtecnZ8Gu0+fZKvg5a=j4xxu-qd1Dejt-f8AszOCyA@mail.gmail.com>

You'll probably have more luck bringing this up on the HtmlUnit list.

On Fri, Oct 28, 2011 at 3:37 AM, Peter Szinek <peter at hexagile.com> wrote:

> Somewhat better with juby-1.6.5 - at least now when I examine
> browser.xml (by outputting it to a html file and opening that in a
> browser) the button is visible (with jruby-1.6.3 it's not even visible
> in the browser). However,
>
> browser.element_by_xpath("//input[@id='searchButton']").exists?
>
> still returns false...
>
> On Fri, Oct 28, 2011 at 11:33 AM, Peter Szinek <peter at hexagile.com> wrote:
> > Hey guys,
> >
> > Tried this without javascript, as well as :resynchronize => true (not
> > sure it should make any difference in this case - well it doesn't)
> >
> > browser = Celerity::Browser.new(:user_agent =>'Mozilla/4.0
> > (compatible; MSIE 7.0b; Windows NT 6.0)')
> > browser.goto "
> http://careers.peopleclick.com/careerscp/client_motorola/external/search.do
> "
> > browser.element_by_xpath("//input[@id='searchButton']").exists?
> >  => false
> >
> > That button should definitely be there (checked in a real browser, and
> > even with mechanize) - I know for a fact that until recently the above
> > code returned true, since we are having this scrape around since
> > months if not years, and it was working. The strange thing is the
> > above code is happening with 0.8.9, 0.9.0 as well as 0.9.1... maybe
> > jruby-1.6.3 is the culprit? Need to check this with jruby-1.6.4
> >
> > Help!
> >
> > Cheers,
> > Peter
> >
> _______________________________________________
> 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/20111028/1bb53d30/attachment.html>