Summary: | deleteCol{1,2,3} and insertCols{1,2,3} will fail in HTML5 parser | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Adam Barth <abarth> | ||||
Component: | New Bugs | Assignee: | Adam Barth <abarth> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | eric, hyatt, ian | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | Other | ||||||
OS: | OS X 10.5 | ||||||
Bug Depends on: | |||||||
Bug Blocks: | 41123, 42346 | ||||||
Attachments: |
|
Description
Adam Barth
2010-07-15 14:01:29 PDT
Created attachment 61716 [details]
Patch
What does IE do? It may be expected that table.removeChild(col) works even if col is not a direct child of table. I feel like I've seen similar hacks like that in the DOM for table support. The tests fail in IE8 as currently written. Let me try the new versions. The new tests seem to fail in IE8 also for layout reasons. However, the DOM operations appear to work as expected. Bottom line: this colgroup business appears to make us more IE-like. so IE adds implicit colgroup elements, but table.removeChild(col) still works? Does table.removeChild(col) remove the colgroup too? (In reply to comment #6) > so IE adds implicit colgroup elements, but table.removeChild(col) still works? No. The test as written fails in IE the same way it will fail with the HTML5 parser. > Does table.removeChild(col) remove the colgroup too? No. It throws an illegal instruction exception, similar to exception we'll throw. Comment on attachment 61716 [details]
Patch
Please update the ChangeLog to explain that you tested in IE8, etc. and explain IE's exact behavior. I think it's important for us to justify this change for future readers of these tests.
I wonder if old FF added implicit colgroup elements or not.
Thank you for looking at this.
Committed r63464: <http://trac.webkit.org/changeset/63464> |