[webkit-reviews] review granted: [Bug 28724] [ES5] Implement getOwnPropertyDescriptor : [Attachment 38580] WebCore portion

[bugzilla-daemon at webkit.org]

Gavin Barraclough <barraclough at apple.com> has granted Oliver Hunt
<oliver at apple.com>'s request for review:
Bug 28724: [ES5] Implement getOwnPropertyDescriptor
https://bugs.webkit.org/show_bug.cgi?id=28724

Attachment 38580: WebCore portion
https://bugs.webkit.org/attachment.cgi?id=38580&action=review

------- Additional Comments from Gavin Barraclough <barraclough at apple.com>
"This requires a custom implementation of getOwnPropertyDescriptor for every
class with a custom implementation of getOwnPropertyDescriptor."

Well, none do right now ... so, nothing to be added then? ;-)
I guess you mean "with a custom implementation of getOwnPropertySlot"?

 87	if (JSGlobalObject::getOwnPropertyDescriptor(exec, propertyName,
descriptor))
 88	    return true;
 89	return false;

Um, isn't that	"return JSGlobalObject::getOwnPropertyDescriptor(exec,
propertyName, descriptor);" ?

Could you rewrite the CodeGeneratorJS.pm script in a language other then Perl
please? - perl sucks.

So, I share Geoff's concern that this duplicates all the getOwnPropertySlot
methods.  I understand that they could not be used to implement
getOwnPropertyDescriptor, since there is a loss of information re getters (they
have already been resolved), but I wonder if we could structure it the other
way around – only implement the full complex logic for
getOwnPropertyDescriptor, and make gopSlot call this.  If the slot returned by
getOwnPropertyDescriptor describes a getter, then gopSlot could invoke it. 
Please cogitate and ruminate on this suggestion, and if you can't make it
better, r+.