REGRESSION (r102262): iAd Producer relies on CSSStyleDeclaration property setters respecting '!important'

• Component: New Bugs
• Status: RESOLVED FIXED

[Andy Estes]

REGRESSION (r102262): iAd Producer relies on CSSStyleDeclaration property setters respecting '!important'

[Andy Estes]

Created attachment 136995 [details]
Patch

[Alexey Proskuryakov]

Comment on attachment 136995 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=136995&action=review

> Source/WebCore/WebCore.exp.in:687
> +__ZN7WebCore24applicationIsIAdProducerEv

It's quite weird that we go back and forth between WebCore and WebKit like this. Is this really how we always do this? Can't this function just be in WebKit?

[Andy Estes]

(In reply to comment #2)
> (From update of attachment 136995 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=136995&action=review
> 
> > Source/WebCore/WebCore.exp.in:687
> > +__ZN7WebCore24applicationIsIAdProducerEv
> 
> It's quite weird that we go back and forth between WebCore and WebKit like this. Is this really how we always do this? Can't this function just be in WebKit?

I might be abusing a file meant for functions called both in WebCore and WebKit. I'll move it into WebKit before landing.

[Andy Estes]

Committed r114048: <http://trac.webkit.org/changeset/114048>

[Eric Seidel (no email)]

Comment on attachment 136995 [details]
Patch

Feels a bit odd to flag this off of USE(JSC), since it's really only PLATFORM(MAC), but OK.

[Andy Estes]

(In reply to comment #5)
> (From update of attachment 136995 [details])
> Feels a bit odd to flag this off of USE(JSC), since it's really only PLATFORM(MAC), but OK.

USE(JSC) seems appropriate since the hack is specific to the JSC bindings. The fact that this is only enabled on WebKit1/Mac at the moment seems orthogonal to how we make the setting available to ports.

[Andy Estes]

<rdar://problem/11131241>