Bug 30719 – Fix the Chromium build after IsloatedWorld change (r49963)

• Component: WebKit Misc.
• Status: RESOLVED FIXED

[Tony Chang]

see summary

[Tony Chang]

Created attachment 41732 [details]
Patch v1

[Tony Chang]

Comment on attachment 41732 [details]
Patch v1

need changelog

[Tony Chang]

Created attachment 41734 [details]
Patch v1

[Tony Chang]

(In reply to comment #3)
> Created an attachment (id=41734) [details]
> Patch v1

I haven't tested the compile for webkit mac, let me do that before landing this.

[David Levin]

Comment on attachment 41734 [details]
Patch v1

> diff --git a/WebCore/bindings/v8/ScriptController.cpp b/WebCore/bindings/v8/ScriptController.cpp

> +// FIXME

Would be nice to say what is to be fixed.

> +DOMWrapperWorld* mainThreadNormalWorld()
> +{
> +    return 0;
> +}


> diff --git a/WebCore/bindings/v8/ScriptController.h b/WebCore/bindings/v8/ScriptController.h

> +    // FIXME: Stub method so we compile.  Currently called from
> +    // FrameLoader.cpp.
> +    DOMWrapperWorld* mainThreadNormalWorld();

Why not move this comment inside of the cpp file and just not have it here?
(Is there anything to fix with the prototype?)


> diff --git a/WebCore/dom/Document.cpp b/WebCore/dom/Document.cpp
> +#if USE(JSC)
>      world->rememberDocument(this);
Can this method be implemented (to do nothing) for V8 and then remove this ifdef in the code?

> +#endif

[Tony Chang]

Created attachment 41736 [details]
v2

[Tony Chang]

(In reply to comment #5)
> (From update of attachment 41734 [details])
> > diff --git a/WebCore/bindings/v8/ScriptController.h b/WebCore/bindings/v8/ScriptController.h
> 
> > +    // FIXME: Stub method so we compile.  Currently called from
> > +    // FrameLoader.cpp.
> > +    DOMWrapperWorld* mainThreadNormalWorld();
> 
> Why not move this comment inside of the cpp file and just not have it here?
> (Is there anything to fix with the prototype?)

Done.

> > diff --git a/WebCore/dom/Document.cpp b/WebCore/dom/Document.cpp
> > +#if USE(JSC)
> >      world->rememberDocument(this);
> Can this method be implemented (to do nothing) for V8 and then remove this
> ifdef in the code?
> 
> > +#endif

That would involve implementing |world| (currently, we get by because it's just a pointer that gets passed around).  abarth volunteered to do some cleanup after this.

[Adam Barth]

(In reply to comment #7)
> That would involve implementing |world| (currently, we get by because it's just
> a pointer that gets passed around).  abarth volunteered to do some cleanup
> after this.

We'll implement |world| eventually.  At the moment, the object we use for that has a different lifetime than in JSC, so it will require more surgery than we want to do in this patch.

[Tony Chang]

http://trac.webkit.org/changeset/49992