Implement CSS attr() substitution with type() syntax, number, and unit types

[CGQAQ]

d47bc4a

Implement CSS attr() substitution with type() syntax, number, and unit types
https://bugs.webkit.org/show_bug.cgi?id=203322
https://bugs.webkit.org/show_bug.cgi?id=203323
https://bugs.webkit.org/show_bug.cgi?id=203324
https://bugs.webkit.org/show_bug.cgi?id=203326
https://bugs.webkit.org/show_bug.cgi?id=203327
https://bugs.webkit.org/show_bug.cgi?id=203328
https://bugs.webkit.org/show_bug.cgi?id=203329

Reviewed by NOBODY (OOPS!).

Extend the CSS variable reference substitution pipeline to handle
attr() as a substitution function per CSS Values 5 specification.
This allows attr() with type(<syntax>), raw-string, number, and
<attr-unit> to work in any CSS property, not just the content property.

When an attr() is encountered during parsing, it is detected as a
variable reference via classifyBlock() and stored as a
CSSVariableReferenceValue. During style resolution, resolveAttrReference()
reads the element attribute, validates it against the specified type,
and substitutes the resolved tokens (or the fallback value).

Key behaviors implemented:
- type(<syntax>): validates attribute value against CSS syntax
- raw-string / omitted type: substitutes as CSS string token
- number: parses attribute as a number token
- <attr-unit> (px, em, %, etc.): parses as number then converts to dimension
- Fallback defaults per spec (empty string when type omitted, guaranteed-invalid otherwise)
- Cycle detection for nested attr() references

FIXME: Substitution of var()/attr()/env() within attribute values
(spec step 6) is not yet implemented to avoid crashes from recursive
custom property resolution.
FIXME: attr()-taint security is not yet implemented.

* LayoutTests/TestExpectations:
* LayoutTests/imported/w3c/web-platform-tests/css/css-content/parsing/content-invalid-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-content/parsing/content-invalid.html: update it the latest WPT one
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-IACVT-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-all-types-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-container-style-query-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-cycle-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-invalidation-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-length-specified-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-null-namespace-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-pseudo-elem-invalidation-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-revert-rule-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/attr-security-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/if-conditionals-expected.txt:
* LayoutTests/imported/w3c/web-platform-tests/css/css-values/if-cycle-expected.txt:
* Source/WebCore/css/CSSVariableReferenceValue.cpp:
(WebCore::parseAttrTypeSyntax):
(WebCore::isValidAttrUnit):
(WebCore::parseAttrValueAsNumber):
(WebCore::CSSVariableReferenceValue::resolveAttrReference const):
(WebCore::CSSVariableReferenceValue::resolveTokenRange const):
* Source/WebCore/css/CSSVariableReferenceValue.h:
* Source/WebCore/css/parser/CSSVariableParser.cpp:
(WebCore::classifyBlock):
(WebCore::isValidAttrUnit):
(WebCore::isValidAttrReference):

d47bc4a

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 win
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	❌ 🧪 wpe-wk2	❌ 🧪 win-tests
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe
	❌ 🧪 ios-wk2-wpt	✅ 🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🛠 ios-safer-cpp	❌ 🧪 mac-wk2	✅ 🧪 gtk-wk2
	✅ 🛠 vision	❌ 🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	🛠 playstation
	✅ 🧪 vision-wk2	❌ 🧪 mac-intel-wk2
	✅ 🛠 tv	✅ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[webkit-early-warning-system]

EWS run on previous version of this PR (hash 0d6fa52)

Details

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 win
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	❌ 🧪 wpe-wk2	❌ 🧪 win-tests
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe
	🧪 ios-wk2-wpt	✅ 🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	❌ 🛠 ios-safer-cpp	❌ 🧪 mac-wk2	❌ 🧪 gtk-wk2
	✅ 🛠 vision	🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	🛠 playstation
	✅ 🧪 vision-wk2	❌ 🧪 mac-intel-wk2
	✅ 🛠 tv	❌ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[webkit-ews-buildbot]

iOS Safer C++ Build #4538 (0d6fa52)

❌ Found 1 failing file with 2 issues. Please address these issues before landing. See WebKit Guidelines for Safer C++ Programming.
(cc @rniwa)

[webkit-ews-buildbot]

macOS Safer C++ Build #86182 (0d6fa52)

❌ Found 1 failing file with 2 issues. Please address these issues before landing. See WebKit Guidelines for Safer C++ Programming.
(cc @rniwa)

[webkit-early-warning-system]

EWS run on previous version of this PR (hash def5a0a)

Details

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 win
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	🧪 wpe-wk2	🧪 win-tests
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe
	🧪 ios-wk2-wpt	✅ 🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	❌ 🛠 ios-safer-cpp	🧪 mac-wk2	🧪 gtk-wk2
	✅ 🛠 vision	🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	🛠 playstation
	✅ 🧪 vision-wk2	🧪 mac-intel-wk2
	✅ 🛠 tv	❌ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[webkit-early-warning-system]

EWS run on current version of this PR (hash d47bc4a)

Details

Misc	iOS, visionOS, tvOS & watchOS	macOS	Linux	Windows
✅ 🧪 style	✅ 🛠 ios	✅ 🛠 mac	✅ 🛠 wpe	✅ 🛠 win
✅ 🧪 bindings	✅ 🛠 ios-sim	✅ 🛠 mac-AS-debug	❌ 🧪 wpe-wk2	❌ 🧪 win-tests
✅ 🧪 webkitperl	✅ 🧪 ios-wk2	✅ 🧪 api-mac	✅ 🧪 api-wpe
	❌ 🧪 ios-wk2-wpt	✅ 🧪 api-mac-debug	✅ 🛠 gtk3-libwebrtc
	✅ 🧪 api-ios	✅ 🧪 mac-wk1	✅ 🛠 gtk
	✅ 🛠 ios-safer-cpp	❌ 🧪 mac-wk2	✅ 🧪 gtk-wk2
	✅ 🛠 vision	❌ 🧪 mac-AS-debug-wk2	✅ 🧪 api-gtk
	✅ 🛠 vision-sim	✅ 🧪 mac-wk2-stress	🛠 playstation
	✅ 🧪 vision-wk2	❌ 🧪 mac-intel-wk2
	✅ 🛠 tv	✅ 🛠 mac-safer-cpp
	✅ 🛠 tv-sim
	✅ 🛠 watch
	✅ 🛠 watch-sim

[webkit-ews-buildbot]

macOS Safer C++ Build #86189 (def5a0a)

❌ Found 1 failing file with 2 issues. Please address these issues before landing. See WebKit Guidelines for Safer C++ Programming.
(cc @rniwa)

[webkit-ews-buildbot]

iOS Safer C++ Build #4545 (def5a0a)

❌ Found 1 failing file with 2 issues. Please address these issues before landing. See WebKit Guidelines for Safer C++ Programming.
(cc @rniwa)

[weinig]

This needs to be behind a new preference flag.

[weinig]

Can you also add a bit more (either here in the PR comments or in the commit message) about the overall plan for this feature? e.g. how does it interact with the existing attr() support that is specific to the content property? How will the equivalent functionality that Style::BuilderState::registerContentAttribute() (and related) currently provide, including how invalidation will work?

It's not a huge problem for this to be in classes called CSSVariableParser/CSSVariableReferenceValue. This is an existing problem, as they don't just support var(), but with each additional feature we add, it gets less ideal.

[CGQAQ]

Can you also add a bit more (either here in the PR comments or in the commit message) about the overall plan for this feature?

I'm just poking around to see what's happening. But I guess I need to split the PR when things are sorted out.

e.g. how does it interact with the existing attr() support that is specific to the content property?

If I understand it correctly, the new, modern behavior should eventually completely override the old one?