280618 - say no to foreign colgroup children

Status: RESOLVED FIXED

(Core :: Layout: Tables, defect)

Description

[Bernd]

see https://bugzilla.mozilla.org/show_bug.cgi?id=265867#c28

<DIV STYLE="display:table-column-group;">
  <DIV>Hello</DIV>
</DIV>

the inner div should not create a frame, even if by some code path it does, it
should never be reflown.

patch coming soon

Comment 1

[Boris Zbarsky [:bzbarsky]]

Note that the same is true for table-column frames.

Please make sure to fix ContentAppended and ContentInserted, not just
TableProcessChild...

Comment 2

[Ben Fowler]

Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8b) Gecko/20050201
Firefox/1.0+

The testcase referenced at http://exchangecode.com/crashbugs/265867.html crashes
on the Mac.

Comment 3

[Bernd]

Hixie at w3c-style:

On Mon, 31 Jan 2005, Boris Zbarsky wrote:

>> 
>> Section 17.2.1 seems to miss a few cases of nesting of various types of table
>> elements.  In particular, I am interested in the following cases:
>> 
>> 1) A child T of a table-column-group P when T is not a table-column
>> 2) A child T of a table-column P


No anonymous boxes are created in those cases (since nothing is described 
in that section).



>> Should any sort of anonymous objects be generated in these cases?  If not,
>> what should be the rendering?


# Elements with 'display' set to 'table-column' or 'table-column-group' 
# are not rendered (exactly as if they had 'display: none'), but they are 
# useful, because they may have attributes which induce a certain style 
# for the columns they represent."
 -- http://www.w3.org/TR/CSS21/tables.html#q2

Comment 4

[Bernd]

Attachment: patch

Comment 5

[Bernd]

Attachment: revised patch

Comment 6

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 173781 [details] [diff] [review]
revised patch

r+sr=bzbarsky

Comment 7

[Bernd]

Comment on attachment 173781 [details] [diff] [review]
revised patch

I rtested the patch, it fixes one of these mangleme crashes

Comment 8

[Asa Dotzler [:asa]]

Comment on attachment 173781 [details] [diff] [review]
revised patch

a=asa for 1.8b checkin

Comment 9

[Bernd]

fix checked in

Comment 10

[Andrew Schultz]

*** Bug 282172 has been marked as a duplicate of this bug. ***

Comment 11

[Jesse Ruderman]

in-testsuite+ covered by the crashtest for bug 265867 (in my tree, will push soon).