[Webkit-unassigned] [Bug 107207] [V8] Support selectively wrapping DOM accesses from certain V8 contexts.

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=107207





--- Comment #24 from Ulfar Erlingsson <ulfar.chromium at gmail.com>  2013-03-14 15:39:31 PST ---
(In reply to comment #21)
> (From update of attachment 191857 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=191857&action=review
> 
> > Source/WebCore/bindings/v8/V8DOMCoverWrapping.cpp:81
> > +static void weakReferenceCallback(v8::Isolate* isolate, v8::Persistent<v8::Value> value, void* dataPtr)
> 
> We have a new mechanism for making weak callbacks typesafe.  We should probably use that here now.

what is that mechanism?

> 
> > Source/WebCore/bindings/v8/V8DOMCoverWrapping.cpp:85
> > +    value.Dispose(isolate);
> 
> We usually call value.Clear after disposing the handle.  I'm not sure it actually does much though.

Couldn't tell myself.  Will call clear.

> 
> > Source/WebCore/bindings/v8/V8DOMCoverWrapping.cpp:97
> > +    const V8DOMCoverWrapping::ClosureData dataContents = {0, method, 0, 0};
> > +    V8DOMCoverWrapping::ClosureData* instance = new V8DOMCoverWrapping::ClosureData(dataContents);
> 
> This is a strange way of creating a V8DOMCoverWrapping::ClosureData object.  Why do we have the const one?

Idiomatic way of heap-allocating an initialized C++ POD struct in a manner friendly to compiler-optimization (in the 90's).  Will go away with the changes we discussed.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.