Summary: | Implement Element.all (http://news.vnet.cn : Mouse hover on tabs not working) | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Anantha Keesara <anantha> | ||||||
Component: | DOM | Assignee: | Nobody <webkit-unassigned> | ||||||
Status: | RESOLVED WONTFIX | ||||||||
Severity: | Normal | CC: | andersca, annevk, jwalden+bwo | ||||||
Priority: | P2 | ||||||||
Version: | 528+ (Nightly build) | ||||||||
Hardware: | All | ||||||||
OS: | All | ||||||||
URL: | http://news.vnet.cn | ||||||||
Attachments: |
|
Description
Anantha Keesara
2007-12-02 09:00:07 PST
Created attachment 17647 [details]
reduction
Confirmed. We only have document.all, HTML elements do not implement all() in WebKit. Should Element.all be hidden, like Document.all? I'm no expert in the area. But I would suggest we hide it. There is no sense in advertising our being more like IE. Element.all only needs to exist for compat with sites which only assume IE (and are currently completely broken on all non-IE browsers). We already have the code to make a hidden property, we can just reuse the stuff for Document.all. Catfish_Man asks me to note this here: While I'm thinking about it, I remember hearing that Mozilla got asked to implement Element.all but didn't see the need and were attempting to hold the line on that particular IEism; webkit implementing doesn't help that. :-\ Created attachment 18557 [details]
Patch
The full .all feature is pretty weird, according to Microsoft's documentation. This reuses what we have for document.all, which is enough to make the tabs on news.vnet.cn work.
Comment on attachment 18557 [details]
Patch
So I think this patch is fine. The only thing missing is test cases. :)
r- until we add at least one element.all test case! :) The test that it's undetectable is nice, but doesn't really test the feature. :)
At this point this is no longer needed. |