[webkit-reviews] review granted: [Bug 62248] WKResponderChainSink mechanism isn't robust against some types of responder chain manipulation : [Attachment 96326] Patch to make [WKResponderChainSink detach] robust against some kinds of responder chain manipulations

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Tue Jun 7 16:39:15 PDT 2011


Darin Adler <darin at apple.com> has granted John Sullivan <sullivan at apple.com>'s
request for review:
Bug 62248: WKResponderChainSink mechanism isn't robust against some types of
responder chain manipulation
https://bugs.webkit.org/show_bug.cgi?id=62248

Attachment 96326: Patch to make [WKResponderChainSink detach] robust against
some kinds of responder chain manipulations
https://bugs.webkit.org/attachment.cgi?id=96326&action=review

------- Additional Comments from Darin Adler <darin at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=96326&action=review

> Source/WebKit2/UIProcess/API/mac/WKView.mm:2527
> +    // This assumes that the responder chain was either unmodified since
> +    // -initWithResponderChain: was called, or was modified in such a way
> +    // that _lastResponderInChain is still in the chain, and self was not
> +    // moved earlier in the chain than _lastResponderInChain.

Another technique would be to start with -[NSWindow firstResponder] and search
for self in that chain. We could do that instead of saving
_lastResponderInChain. Or we could do both to make sure we’re removed if either
technique finds us.


More information about the webkit-reviews mailing list