Re: [csswg-drafts] [css-values-5] Can ident() bypass dashed-ident requirements? (#12206)

I don't think `ident()` should be privileged compered to verbatim idents, the whole idea is that it behaves identically. Implementations should not have to care where the ident came from, and whether an `ident()` was used to construct it.

First, the unknown is due to a variable, we already have established rules about how that resolves, so we should focus this issue around checking when we *don't* have a variable, such as the example where `ident(ident("myname" sibling-index()))` is used somewhere where only a `<dashed-ident>` is allowed. Conceptually, this should be always be detectable at parse time.

I think the mock eval makes sense, but it's a bit awkward and I'd rather not encode it in the spec. Perhaps the normative text could be more high-level, leaving UAs flexibility about _how_ they do the checking. Perhaps the mock-eval idea by @andruud could be a non-normative note or something.

-- 
GitHub Notification of comment by LeaVerou
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/12206#issuecomment-3998559631 using your GitHub account


-- 
Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config

Received on Wednesday, 4 March 2026 16:13:21 UTC