[nekohtml-dev] [ nekohtml-Bugs-3527659 ] table tag does not close surrounding inline tags in browsers

[SourceForge.net <no...@so...>]

Bugs item #3527659, was opened at 2012-05-17 11:00
Message generated for change (Comment added) made by mguillem
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=952178&aid=3527659&group_id=195122

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: tag-balancer
Group: None
>Status: Closed
>Resolution: Fixed
Priority: 5
Private: No
Submitted By: RBRi (rbri)
>Assigned to: Marc Guillemot (mguillem)
Summary: table tag does not close surrounding inline tags in browsers

Initial Comment:
<p><table></table> is balanced to <p></p><table></table> but browsers are building <p><table></table></p>

attached a patch including one new test
The patch includes some more minor fixes (not needed null checks and a fix for the element definition)

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

>Comment By: Marc Guillemot (mguillem)
Date: 2012-10-04 23:55

Message:
Fixed in SVN. Thanks for reporting.

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

Comment By: RBRi (rbri)
Date: 2012-08-29 23:16

Message:
FF 3.6
abbr, acronym, cite, code, dfn, kbd, label, samp, var

FF 10 ESR
-

IE 9
-

IE 8
-

IE 6
abbr

So it is ok forme, if Neko does not support IE6 and FF3.6 because there is
hope that theese browsers are gone (soner or later)

    rbri

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

Comment By: RBRi (rbri)
Date: 2012-08-29 10:54

Message:
Sorry, 

IE 6
abrr

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

Comment By: RBRi (rbri)
Date: 2012-08-29 10:52

Message:
IE 6
-

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

Comment By: RBRi (rbri)
Date: 2012-08-29 10:50

Message:
did some tests with the attached 'tableClosesInlineTag.html'

FF 3.6
abbr, acronym, cite, code, dfn, kbd, label, samp, var

FF 10 ESR
-

IE 9
-

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

Comment By: https://www.google.com/accounts ()
Date: 2012-08-21 13:36

Message:
@rbri are you suggesting NekoHTML parser should match old unsupported
browsers?

I would like NekoHTML to conform to the standard:
http://www.whatwg.org/specs/web-apps/current-work/#parsing
like current browsers do.

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

Comment By: RBRi (rbri)
Date: 2012-08-21 11:24

Message:
Marc,

if my memories are correct, FF3 and FF3_6 are producing different results.
Here you can find my test source

http://htmlunit.svn.sourceforge.net/viewvc/htmlunit/trunk/htmlunit/src/test/java/com/gargoylesoftware/htmlunit/html/HTMLParser2Test.java?view=markup#l144



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

Comment By: https://www.google.com/accounts ()
Date: 2012-08-21 08:49

Message:
Now that browsers conform to the standard:
http://www.whatwg.org/specs/web-apps/current-work/#parsing
if the tree is different in a browser or NekoHTML then there is a bug in
either of them.
If a test in NekoHTML differs from Gecko and WebKit parsers I would
consider that a wrong test.

In this particular case it seems there is a bug in NekoHTML as both Gecko
and WebKit parsers differ from NekoHTML.

@rbri what browser did you find producing a different tree from Gecko and
WebKit?

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

Comment By: RBRi (rbri)
Date: 2012-05-21 08:44

Message:
As far as i understand the spec, table has to close all inline tags. So at
least from the spec (and if we balance xhtml content) i think the behavior
is correct.
And again the browsers are different. FF3 and FF3.6 are closing some tags
and some not (already added an NYI test to HtmlUnit).

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

Comment By: Marc Guillemot (mguillem)
Date: 2012-05-21 06:12

Message:
Should table close any "inline" at all? Ok, some unit tests explicitly test
for this but it seems to me that these tests should be considered as wrong.

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=952178&aid=3527659&group_id=195122