Issue was closed in CSSTree project as plan is to change the syntax in CSS Syntax instead.
This is meant to replace #1331.
Side note: we don't need a patch on CSS Syntax because <an+b> is the name of the construct and not its value, and our tests only check that values can be parsed with CSSTree. Now that I think about it, that's probably an oversight, the guarantee should extend to names as well: if the names can't be parsed, they cannot be used as values in any case. I'll propose to refine the test separately (testing locally, <an+b> would be the only problematic name).
Issue was closed in CSSTree project as plan is to change the syntax in CSS Syntax instead. This is meant to replace #1331.
Side note: we don't need a patch on CSS Syntax because
<an+b>
is thename
of the construct and not itsvalue
, and our tests only check that values can be parsed with CSSTree. Now that I think about it, that's probably an oversight, the guarantee should extend to names as well: if the names can't be parsed, they cannot be used as values in any case. I'll propose to refine the test separately (testing locally,<an+b>
would be the only problematic name).