[webkit-reviews] review denied: [Bug 45145] [Qt] V8 port for QT platform: v8 binding changes : [Attachment 66574] Included plugins changes b/c of dependency

[bugzilla-daemon at webkit.org]

Adam Barth <abarth at webkit.org> has denied Vlad <vladbph at gmail.com>'s request
for review:
Bug 45145: [Qt] V8 port for QT platform: v8 binding changes
https://bugs.webkit.org/show_bug.cgi?id=45145

Attachment 66574: Included plugins changes b/c of dependency 
https://bugs.webkit.org/attachment.cgi?id=66574&action=review

------- Additional Comments from Adam Barth <abarth at webkit.org>
View in context:
https://bugs.webkit.org/attachment.cgi?id=66574&action=prettypatch

> WebCore/bindings/v8/ScriptController.cpp:74
> +#if PLATFORM(QT)
> +PassRefPtr<DOMWrapperWorld> ScriptController::createWorld()
> +{
> +    return DOMWrapperWorld::create();
> +}
> +#endif
This can't be right.  It's not Qt-specific.  We need to integrate V8 with
DOMWrapperWorld in genera.

> WebCore/bindings/v8/V8Binding.cpp:54
> -	   : m_plainString(string)
> +	   : m_plainString(reinterpret_cast<const WTF::String &>(string))
I don't understand this code.  Does Qt use different string types than other
ports?	reinterpret_cast is almost always wrong.

> WebCore/plugins/win/PluginViewWin.cpp:661
> +#if USE(JSC)
>      JSC::JSLock::DropAllLocks dropAllLocks(JSC::SilenceAssertionsOnly);
> +#endif
This ifdefs point to a missing abstraction.  Perhaps a ScriptLock object?