Support the `x` resolution unit

• Component: CSS
• Status: RESOLVED FIXED

[Sam Sneddon [:gsnedders]]

https://drafts.csswg.org/css-values-4/#x

This is identical to dppx, but allows for nicer statements like `resolution: 2x`.

[Sam Sneddon [:gsnedders]]

Created attachment 437335 [details]
Patch

This might depend on bug 229776, not quite sure if it'll apply cleanly without

[Sam Sneddon [:gsnedders]]

Ah yes, the updates to LayoutTests/imported/w3c/web-platform-tests/css/mediaqueries/test_media_queries-expected.txt depend on bug 229776. Could in principle land this without, but the gain is much smaller.

[Radar WebKit Bug Importer]

<rdar://problem/82897690>

[EWS]

Committed r282396 (241657@main): <https://commits.webkit.org/241657@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 437335 [details].

[Darin Adler]

Comment on attachment 437335 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=437335&action=review

> Source/WebCore/css/DeprecatedCSSOMPrimitiveValue.cpp:94
> +    case CSSUnitType::CSS_X:                            return 37;

Please don’t add this. We don’t need to expose modern units in the legacy CSS primitive value object model. Also, this patch has no test coverage for this. Just return CSS_UNKNOWN for tis case.

If you have a different strategy in mind, please add new test cases demonstrating what we intend, and consider adding a constant to the IDL file.

> Source/WebCore/css/DeprecatedCSSOMPrimitiveValue.cpp:136
> +    case 37:                return m_value->getFloatValue(CSSUnitType::CSS_X);

Please don’t add this. Same reason as above.

[Brent Fulgham]

(In reply to Darin Adler from comment #5)
> Comment on attachment 437335 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=437335&action=review
> 
> > Source/WebCore/css/DeprecatedCSSOMPrimitiveValue.cpp:94
> > +    case CSSUnitType::CSS_X:                            return 37;
> 
> Please don’t add this. We don’t need to expose modern units in the legacy
> CSS primitive value object model. Also, this patch has no test coverage for
> this. Just return CSS_UNKNOWN for tis case.
> 
> If you have a different strategy in mind, please add new test cases
> demonstrating what we intend, and consider adding a constant to the IDL file.
> 
> > Source/WebCore/css/DeprecatedCSSOMPrimitiveValue.cpp:136
> > +    case 37:                return m_value->getFloatValue(CSSUnitType::CSS_X);
> 
> Please don’t add this. Same reason as above.

These comments will be addressed in Bug 233651.