Bug 15219
| Summary: | fast/dom/replaceChild and clientWidthAfterDocumentIsRemoved have invalid expected output | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Matt Perry <mpComplete> |
| Component: | Tools / Tests | Assignee: | Nobody <webkit-unassigned> |
| Status: | NEW | ||
| Severity: | Normal | CC: | ap |
| Priority: | P2 | Keywords: | LayoutTestFailure |
| Version: | 523.x (Safari 3) | ||
| Hardware: | Mac | ||
| OS: | OS X 10.4 | ||
Matt Perry
Both of these tests remove a child iframe from the document. Inside ContainerNode::removeChild, a call to willRemoveChild dispatches some events that leads to a load-completed event being fired, and the render tree is dumped prematurely (ie, before the child nodes are removed from the tree). The DRT expected output contains the nodes that have been removed, which is clearly incorrect.
The scary thing here is that a load completed event is being fired from within a removeChild call!
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
David Kilzer (:ddkilzer)
Also, LayoutTests/fast/dom/resources/replaceChildHelper.html is not well-formatted (<body> tag does not immediately follow </head> tag).
David Kilzer (:ddkilzer)
Confirmed with a local debug build of WebKit r25545 on Safari 3 Public Beta v. 3.0.3 (522.12.1) on Mac OS X 10.4.10 (8R218).
Viewing the test with the Web Inspector shows no <iframe> element left in the DOM. However, the render tree dump clearly shows an <iframe> element in the DOM.
http://trac.webkit.org/projects/webkit/browser/trunk/LayoutTests/fast/dom/replaceChild-expected.txt#L7
David Kilzer (:ddkilzer)
(In reply to comment #1)
> Also, LayoutTests/fast/dom/resources/replaceChildHelper.html is not
> well-formatted (<body> tag does not immediately follow </head> tag).
Or "well-formed".
http://trac.webkit.org/projects/webkit/browser/trunk/LayoutTests/fast/dom/resources/replaceChildHelper.html#L16