Bug 231464 – [macOS] Add support for accent-color

• Status: RESOLVED
• Resolution: FIXED
• Component: CSS

[Aditya Keerthi]

Add support for accent-color when painting controls on macOS.

[Aditya Keerthi]

rdar://84049511

[Aditya Keerthi]

Created attachment 440697 [details]
Patch

[EWS Watchlist]

This patch modifies the imported WPT tests. Please ensure that any changes on the tests (not coming from a WPT import) are exported to WPT. Please see https://trac.webkit.org/wiki/WPTExportProcess

[Aditya Keerthi]

Created attachment 440698 [details]
Patch

[Aditya Keerthi]

Created attachment 440748 [details]
Patch

[Tim Horton]

Comment on attachment 440748 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=440748&action=review

> Source/WebCore/platform/Theme.cpp:73
> +void Theme::paint(ControlPart, ControlStates&, GraphicsContext&, const FloatRect&, float, ScrollView*, float, float, bool, bool, const Color&)

At what point do we package this up in a struct or something?

> Source/WebCore/rendering/style/RenderStyle.cpp:2182
> +    if (!hasAppleColorFilter())
> +        return colorResolvingCurrentColor(accentColor());

Kinda would expect the weird case (color-filter) to be the early return (but this is fine too).

> LayoutTests/platform/mac/TestExpectations:2350
> +webkit.org/b/199350 fast/css/accent-color/datalist.html [ ImageOnlyFailure ]

Maybe these two failing ones go in their own paragraph that doesn't have a header suggesting these are passing because they're supported?

[Aditya Keerthi]

Created attachment 441002 [details]
Patch for landing

[Aditya Keerthi]

(In reply to Tim Horton from comment #6)
> Comment on attachment 440748 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=440748&action=review
> 
> > Source/WebCore/platform/Theme.cpp:73
> > +void Theme::paint(ControlPart, ControlStates&, GraphicsContext&, const FloatRect&, float, ScrollView*, float, float, bool, bool, const Color&)
> 
> At what point do we package this up in a struct or something?

Filed https://bugs.webkit.org/show_bug.cgi?id=231637.
 
> > Source/WebCore/rendering/style/RenderStyle.cpp:2182
> > +    if (!hasAppleColorFilter())
> > +        return colorResolvingCurrentColor(accentColor());
> 
> Kinda would expect the weird case (color-filter) to be the early return (but
> this is fine too).

Changed.
 
> > LayoutTests/platform/mac/TestExpectations:2350
> > +webkit.org/b/199350 fast/css/accent-color/datalist.html [ ImageOnlyFailure ]
> 
> Maybe these two failing ones go in their own paragraph that doesn't have a
> header suggesting these are passing because they're supported?

Done.

[EWS]

Committed r284115 (242935@main): <https://commits.webkit.org/242935@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 441002 [details].

[Aditya Keerthi]

Note that this is behind a disabled-by-default experimental flag.