Move color CSS properties to the new StyleBuilder

[Chris Dumez]

Move color CSS properties to the new StyleBuilder.

[Chris Dumez]

Created attachment 243665 [details]
Patch

[Antti Koivisto]

Comment on attachment 243665 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=243665&action=review

> Source/WebCore/css/CSSPropertyNames.in:69
> +// * ColorProperty:
> +// Indicates that this CSS property is a color property with a
> +// "setVisitedLinkXXX()" setter on RenderStyle to be called when
> +// StyleResolver::applyPropertyToVisitedLinkStyle() return true.
> +// The regular setter on RenderStyle will only be called if
> +// StyleResolver::applyPropertyToRegularStyle() returns true.

I think this should have a name that indicates that the property works for visited links. VisitedLinkSupportingColorProperty or something.

[Chris Dumez]

Created attachment 243677 [details]
Patch

[Chris Dumez]

(In reply to comment #2)
> Comment on attachment 243665 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=243665&action=review
> 
> > Source/WebCore/css/CSSPropertyNames.in:69
> > +// * ColorProperty:
> > +// Indicates that this CSS property is a color property with a
> > +// "setVisitedLinkXXX()" setter on RenderStyle to be called when
> > +// StyleResolver::applyPropertyToVisitedLinkStyle() return true.
> > +// The regular setter on RenderStyle will only be called if
> > +// StyleResolver::applyPropertyToRegularStyle() returns true.
> 
> I think this should have a name that indicates that the property works for
> visited links. VisitedLinkSupportingColorProperty or something.

I used "VisitedLinkColorSupport" in the latest patch.

[WebKit Commit Bot]

Comment on attachment 243677 [details]
Patch

Clearing flags on attachment: 243677

Committed r177687: <http://trac.webkit.org/changeset/177687>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.

[Darin Adler]

Comment on attachment 243665 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=243665&action=review

> Source/WebCore/rendering/style/RenderStyle.h:2086
> +    static Color invalidColor() { static Color invalid; return invalid; }

I just noticed this pointless global variable. How about just this?

    static Color invalidColor() { return Color(); }

[Chris Dumez]

Reopening to attach new patch.

[Chris Dumez]

Created attachment 243689 [details]
Patch

[Chris Dumez]

(In reply to comment #7)
> Comment on attachment 243665 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=243665&action=review
> 
> > Source/WebCore/rendering/style/RenderStyle.h:2086
> > +    static Color invalidColor() { static Color invalid; return invalid; }
> 
> I just noticed this pointless global variable. How about just this?
> 
>     static Color invalidColor() { return Color(); }

My guess is that the idea was to avoid calling the Color constructor every time. However, the default Color constructor is really cheap so this is likely not worth it.

I uploaded another patch to fix this here.

[WebKit Commit Bot]

Comment on attachment 243689 [details]
Patch

Clearing flags on attachment: 243689

Committed r177720: <http://trac.webkit.org/changeset/177720>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.