Bug 232830 – Re-use isCSSWideKeyword in CSSVariableParser.cpp & CSSPropertyParser.cpp

• Component: CSS
• Status: RESOLVED FIXED

[Tim Nguyen (:ntim)]

Sub-set of bug 224816

[Tim Nguyen (:ntim)]

Created attachment 443584 [details]
Patch

[Antti Koivisto]

Comment on attachment 443584 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=443584&action=review

> Source/WebCore/css/parser/CSSPropertyParser.cpp:4568
> -        CSSValueID id = m_range.peek().id();
> -        if (id == CSSValueInherit || id == CSSValueInitial || id == CSSValueRevert)
> +        if (isCSSWideKeyword(m_range.peek().id()))

Can we make a test for this?

[Tim Nguyen (:ntim)]

(In reply to Antti Koivisto from comment #2)
> Comment on attachment 443584 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=443584&action=review
> 
> > Source/WebCore/css/parser/CSSPropertyParser.cpp:4568
> > -        CSSValueID id = m_range.peek().id();
> > -        if (id == CSSValueInherit || id == CSSValueInitial || id == CSSValueRevert)
> > +        if (isCSSWideKeyword(m_range.peek().id()))
> 
> Can we make a test for this?

I don't see any easily extendable test in https://github.com/web-platform-tests/wpt/tree/master/css/css-properties-values-api unfortunately, or similar test for the matter.

[EWS]

Committed r285483 (244007@main): <https://commits.webkit.org/244007@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 443584 [details].

[Radar WebKit Bug Importer]

<rdar://problem/85191933>

[Darin Adler]

Comment on attachment 443584 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=443584&action=review

>>> Source/WebCore/css/parser/CSSPropertyParser.cpp:4568
>>> +        if (isCSSWideKeyword(m_range.peek().id()))
>> 
>> Can we make a test for this?
> 
> I don't see any easily extendable test in https://github.com/web-platform-tests/wpt/tree/master/css/css-properties-values-api unfortunately, or similar test for the matter.

Is this a behavior change, because we now check for "unset" as well? But not testable?