Summary: | Add test cases to check lookup precedence of DOM constructors | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Pam Greene (IRC:pamg) <pam> | ||||||
Component: | DOM | Assignee: | Pam Greene (IRC:pamg) <pam> | ||||||
Status: | RESOLVED FIXED | ||||||||
Severity: | Normal | ||||||||
Priority: | P2 | ||||||||
Version: | 528+ (Nightly build) | ||||||||
Hardware: | All | ||||||||
OS: | All | ||||||||
Attachments: |
|
Description
Pam Greene (IRC:pamg)
2009-03-11 14:01:59 PDT
Created attachment 28490 [details]
Modified test + new results
Comment on attachment 28490 [details] Modified test + new results > + Add test cases to ensure that DOM constructors have higher lookup > + precedence than a document element with the same name. The test uses id, not name - is that intentional? Should it test both? Created attachment 28543 [details] Test for name too ID is the case that had a problem (http://code.google.com/p/v8/issues/detail?id=43), but sure, let's test name too. Comment on attachment 28543 [details]
Test for name too
r=me, assuming that the new results match Firefox.
WebKit and Firefox don't match in their previously tested behavior: WebKit has constructors for several objects that Firefox doesn't. So FF fails some cases in this test, but no new failures show up due to these test-case additions. That is, in those cases where FF has a constructor at all, its lookup precedence matches ours. Landed as r41669. |