Branch: refs/heads/main
Home: https://github.com/WebKit/WebKit
Commit: 4837c2fac3c3e5de7278b9b7e2b651ba92f6bfd5
https://github.com/WebKit/WebKit/commit/4837c2fac3c3e5de7278b9b7e2b651ba92f6bfd5
Author: Antti Koivisto <[email protected]>
Date: 2026-10-08 (Thu, 08 Oct 2026)
Changed paths:
M LayoutTests/TestExpectations
M Source/WebCore/css/parser/CSSPropertyParserConsumer+Content.cpp
Log Message:
-----------
[css-values-5 attr()] Don't take legacy path in simple cases
https://bugs.webkit.org/show_bug.cgi?id=326777
rdar://problem/189485680
Reviewed by Alan Baradlay.
We were accidentally taking the legacy path in simple attr() cases even with
the flag enabled.
consumeStyleProperty() tries the property grammar before checking for
substitution functions.
The legacy content grammar accepts attr(<ident>) and attr(<ident>, <string>) so
the common forms
never reached CSSSubstitutionParser. The legacy path looks the attribute up on
the styled element
instead of the originating element, so ::part()::before and
::details-content::after got the wrong value.
Reject attr() in the content grammar when the setting is enabled so it is
always parsed as
an arbitrary substitution function.
* LayoutTests/TestExpectations:
imported/w3c/web-platform-tests/html/rendering/the-details-element/details-content-pseudo-attr.html
now passes.
* Source/WebCore/css/parser/CSSPropertyParserConsumer+Content.cpp:
(WebCore::CSSPropertyParserHelpers::consumeUnresolvedContentLegacyAttrFunction):
Canonical link: https://commits.webkit.org/323098@main
To unsubscribe from these emails, change your notification settings at
https://github.com/WebKit/WebKit/settings/notifications
