Basically, WebKit fails to recognize many the keyboard inputs. Mainly,
1. I cannot type the URL into the URL bar or the Google search bar.
2. I cannot type into forms or contentEditable.
I can, however, use cmd + l/q/w/r/ family of shortcut keys (as in the browser responds to events associated with these shortcuts).
*** Bug 68569 has been marked as a duplicate of this bug. ***
I cannot reproduce this issue with nightly r95663 on Lion with Safari 5.1, or on Snow Leopard with Safari 5.0.
Not sure what's going on.
This is occurring with Leopard (10.5.8) with the latest WebKit, even as of the 95673 build; Safari 5.0.6 (see comments from "duplicate" bug 68569). In addition, when using bookmarks to navigate pages, there is a residual image that pops up between sessions -- going from Top Sites to a page brings up the "Welcome to WebKit" page (last one seen before accessing Top Sites). Happens each time going forward.
*** Bug 68768 has been marked as a duplicate of this bug. ***
Actually, I can reproduce on another machine - not sure why I couldn't before.
The problem is that -_usesDocumentViews was removed from WebView. It can be easily put back for now, always returning true.
Who is calling _usesDocumentViews? Safari?
Safari 5.0.x, it appears. We've never supported old Safari in nightlies before, so maybe this is OK to WONTFIX.
Alex: Indeed, the problem went away once I updated Safari to 5.1 from 5.0.6 on my Snow Leopard. WONTFIX sounds good to me.
*** Bug 68987 has been marked as a duplicate of this bug. ***
*** Bug 69057 has been marked as a duplicate of this bug. ***
Why you can't fix this and use multi system compatible event handling? There are good reasons not to use recent Safari 5 update from Apple: It will slow down the browser window a lot, so why slowing down a good system?
If it’s really just the single missing method that’s causing the problem then we should just add it back temporarily.
*** Bug 69125 has been marked as a duplicate of this bug. ***
Created attachment 109426 [details]
Well, this does fix the issue indeed. I'm still not sure if it's worth fixing, but any reviewer should feel free to set r+/cq+ (and re-open the bug for landing) if they feel like it.
Comment on attachment 109426 [details]
Clearing flags on attachment: 109426
Committed r96478: <http://trac.webkit.org/changeset/96478>
All reviewed patches have been landed. Closing bug.