[webkit-reviews] review requested: [Bug 31933] Get rid of JSObject::getPropertyAttributes() and all usage of it : [Attachment 43949] Proposed patch

[bugzilla-daemon at webkit.org]

Kent Hansen <kent.hansen at nokia.com> has asked  for review:
Bug 31933: Get rid of JSObject::getPropertyAttributes() and all usage of it
https://bugs.webkit.org/show_bug.cgi?id=31933

Attachment 43949: Proposed patch
https://bugs.webkit.org/attachment.cgi?id=43949&action=review

------- Additional Comments from Kent Hansen <kent.hansen at nokia.com>
Not sure what to do with the JSVariableObject::getPropertyAttributes()
implementation. Is it really dead code after the move to
getOwnPropertyDescriptor()? The tests that were added in r31225 still pass, but
I'm wondering about the change to set the DontDelete attribute (part of r33979,
SquirrelFish merge). Should that be done in
JSGlobalObject::getOwnPropertyDescriptor() now?

The alternative I see would be to reimplement getOwnPropertyDescriptor() in
JSVariableObject, and set the DontDelete attribute on the result (to mimic
getPropertyAttributes()). But then, really, JSActivationObject should
reimplement getOwnPropertySlot() too, i.e. both those functions (and
hasOwnPropertyForWrite()? and put()? and defineGetter()/Setter()?) should be
moved from JSGlobalObject to JSVariableObject. JSVariableObject only
implementing "half" the API is error-prone.