Closed
Bug 542620
Opened 14 years ago
Closed 14 years ago
table sometimes ignore background-color of column
Categories
(Core :: Layout: Tables, defect)
Tracking
()
People
(Reporter: runoutoftime, Assigned: roc)
References
Details
(Keywords: regression, testcase, verified1.9.2)
Attachments
(4 files, 1 obsolete file)
545 bytes,
text/html
|
Details | |
4.15 KB,
patch
|
dbaron
:
review+
beltzner
:
approval1.9.2.2+
|
Details | Diff | Splinter Review |
2.47 KB,
patch
|
bzbarsky
:
review+
beltzner
:
approval1.9.2.2+
|
Details | Diff | Splinter Review |
9.25 KB,
image/png
|
Details |
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; he; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729) Build Identifier: Mozilla/5.0 (Windows; U; Windows NT 5.1; he; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729) <table> ignores background-color style of <col>. However, background-color of <col> applies when I add a <tr> with another background-color. Two tables (good and bad) are attached. Reproducible: Always Steps to Reproduce: <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <head> <title>title</title> </head> <body> <table> <col style="background-color: green"></col> <tr style="background-color: blue"><td>blue</td></tr> <tr><td>green</td></tr> </table> <table> <col style="background-color: green"></col> <tr><td>THIS SHOULD BE IN GREEN</td></tr> </table> </body> </html> Actual Results: First table is ok (blue and green). Second table is broken (not green) Expected Results: Second table should be green I believe it worked in FF 3.5
Comment 1•14 years ago
|
||
Comment 2•14 years ago
|
||
Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.3a1pre) Gecko/20100127 Minefield/3.7a1pre Confirmed, this worked indeed fine with Firefox 3.5 but is broken on trunk and with Firefox 3.6.
Status: UNCONFIRMED → NEW
Ever confirmed: true
Version: unspecified → 3.6 Branch
Updated•14 years ago
|
Component: General → Layout: Tables
Product: Firefox → Core
QA Contact: general → layout.tables
Version: 3.6 Branch → 1.9.2 Branch
Comment 3•14 years ago
|
||
works Gecko/20091212 Minefield/3.7a1pre http://hg.mozilla.org/mozilla-central/rev/d379a17cbf8f fails Gecko/20091213 Minefield/3.7a1pre http://hg.mozilla.org/mozilla-central/rev/3ff17b03644e http://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=d379a17cbf8f&tochange=3ff17b03644e two bugs stand out: bug 530686 and bug 531461
Blocks: 530686
blocking2.0: --- → beta1
AnyTablePartHasBorderOrBackground needs to look at column frames, which are stored on a separate child list.
Assignee | ||
Updated•14 years ago
|
Assignee: nobody → roc
blocking1.9.2: --- → ?
Assignee | ||
Updated•14 years ago
|
Keywords: regressionwindow-wanted
Assignee | ||
Updated•14 years ago
|
Whiteboard: [needs review]
Comment on attachment 423929 [details] [diff] [review] fix r=dbaron
Attachment #423929 -
Flags: review?(dbaron) → review+
Assignee | ||
Updated•14 years ago
|
Whiteboard: [needs review] → [needs landing]
Assignee | ||
Comment 7•14 years ago
|
||
http://hg.mozilla.org/mozilla-central/rev/5c6c40887584
Assignee | ||
Comment 8•14 years ago
|
||
Comment on attachment 423929 [details] [diff] [review] fix fix for a layout regression
Attachment #423929 -
Flags: approval1.9.2.1?
Comment 9•14 years ago
|
||
That patch would trigger assertions in a build with DEBUG_FRAME_LIST. In particular, using some random non-last-sibling frame as the last frame of an nsFrameList is wrong. The method here probably wants to take an nsFrameList::Slice, not an nsFrameList.
Assignee | ||
Comment 10•14 years ago
|
||
If I use Slice, then to construct a slice that contains just this table frame, I have to call GetParent()->GetChildList(nsnull) to get the framelist the table frame belongs to. (Fortunately we know that the table must be in the main child list of the table-outer.) It's awfully tempting to make AnyTablePartHasBorderOrBackground just take "start, end" nsIFrame* parameters. Would that be wrong? :-)
Attachment #424925 -
Flags: review?(bzbarsky)
Comment 11•14 years ago
|
||
Comment on attachment 424925 [details] [diff] [review] fix nsFrameList usage in previous patch This looks fine. Taking start,end frames is probably ok, since this method never _modifies_ the passed-in list.
Attachment #424925 -
Flags: review?(bzbarsky) → review+
Assignee | ||
Comment 12•14 years ago
|
||
OK, revised to just use two frames. This should be more efficient and robust.
Attachment #423929 -
Attachment is obsolete: true
Attachment #424925 -
Attachment is obsolete: true
Attachment #425120 -
Flags: review?(bzbarsky)
Attachment #423929 -
Flags: approval1.9.2.2?
Updated•14 years ago
|
Attachment #425120 -
Flags: review?(bzbarsky) → review+
Assignee | ||
Updated•14 years ago
|
Whiteboard: [needs landing]
Attachment #425120 -
Flags: approval1.9.2.2?
Attachment #423929 -
Attachment is obsolete: false
Attachment #423929 -
Flags: approval1.9.2.2?
Attachment #425120 -
Attachment description: fix → fix nsFrameList usage in previous patch
Attachment #424925 -
Attachment description: fix → fix nsFrameList usage in previous patch
Assignee | ||
Comment 14•14 years ago
|
||
http://hg.mozilla.org/mozilla-central/rev/9b4fe4d0ca1f
Whiteboard: [needs landing]
Updated•14 years ago
|
blocking1.9.2: ? → needed
status1.9.2:
--- → wanted
This fixes a pretty bad regression in 3.6 from some last-minute changes, and we really shouldn't ship a major update to 3.6 without this patch; I think this would have been a clear release blocker if we'd seen it before the release.
Updated•14 years ago
|
Attachment #425120 -
Flags: approval1.9.2.2? → approval1.9.2.2+
Comment 17•14 years ago
|
||
Comment on attachment 423929 [details] [diff] [review] fix a1922=beltzner, yay tests!
Attachment #423929 -
Flags: approval1.9.2.2? → approval1.9.2.2+
Assignee | ||
Updated•14 years ago
|
Whiteboard: [needs 192 landing]
Assignee | ||
Comment 18•14 years ago
|
||
http://hg.mozilla.org/releases/mozilla-1.9.2/rev/2af75bfe1976 http://hg.mozilla.org/releases/mozilla-1.9.2/rev/16e3ddc501df
Whiteboard: [needs 192 landing]
Comment 19•14 years ago
|
||
I tested this with the attached testcase on Firefox 3.6.0 and my 3.6.2 build. I see no difference in rendering (which appears correct on Firefox 3.6.0). Both render as this attachment does. So, the testcase issue isn't something that we fixed.
What changeset is your 3.6.2 build built on? (See the SourceStamp line in application.ini .)
Comment 21•14 years ago
|
||
I can confirm that the testcase doesn't work with a freshly made build, at http://hg.mozilla.org/releases/mozilla-1.9.2/rev/18ea3b9ac710 when I look at the push log, I see that that those checkins are tagged with 'GECKO1921_20100126_RELBRANCH' http://hg.mozilla.org/releases/mozilla-1.9.2/summary
The previous commit was inadvertently on a relbranch. I backed it out of the relbranch: http://hg.mozilla.org/releases/mozilla-1.9.2/rev/f57f01f6b371 and relanded it on the default branch: http://hg.mozilla.org/releases/mozilla-1.9.2/rev/4f9d4c092dc1 http://hg.mozilla.org/releases/mozilla-1.9.2/rev/ff7fcef1c23e
Comment 23•14 years ago
|
||
verified with Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.3pre) Gecko/20100319 Namoroka/3.6.3pre
Keywords: verified1.9.2
You need to log in
before you can comment on or make changes to this bug.
Description
•