[webkit-reviews] review denied: [Bug 75762] Pointer Lock: Implement pointer interface : [Attachment 123976] Patch

[bugzilla-daemon at webkit.org]

Adam Barth <abarth at webkit.org> has denied Vincent Scheib
<scheib at chromium.org>'s request for review:
Bug 75762: Pointer Lock: Implement pointer interface
https://bugs.webkit.org/show_bug.cgi?id=75762

Attachment 123976: Patch
https://bugs.webkit.org/attachment.cgi?id=123976&action=review

------- Additional Comments from Adam Barth <abarth at webkit.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=123976&action=review


I think you have some memory safety problems...

> Source/WebCore/page/PointerLock.h:52
> +    PointerLockController* m_controller;

How do you know whether this object hasn't been deallocated?

> Source/WebCore/page/PointerLockController.cpp:83
> +	   callbackToIssue->handleEvent();

How do you know the ScriptExecutionContext for this callback is still active?

> Source/WebKit/chromium/src/WebViewImpl.cpp:1019
> +#if ENABLE(POINTER_LOCK)
> +    requestPointerUnlock();
> +#endif

This doesn't seem right.  Is pointer lock scoped to the lifetime of the WebView
or to the lifetime of a particular document?  How can a document that's not
longer active in a WebView hold the pointer lock?