[css-values] Are the codepoints in an indentifier stored in escaped form or not?

[Loirooriol]

Imagine a\ b is parsed as an <ident>. What are the code points of the produced ident?

• Option 1: the indentifier is stored in escaped form
  1. a
  2. \
  4. b
• Option 2: the indentifier is stored raw
  1. a
  3. b

https://drafts.csswg.org/css-values/#css-css-identifier seems to assume option 1:

CSS identifiers, generically denoted by <ident>, consist of a sequence of characters conforming to the <ident-token> grammar.

https://drafts.csswg.org/cssom/#serialize-an-identifier seems to assume option 2, since it escapes manually.

[tabatkins]

The Values text is not meant to imply anything about how things are stored. Syntax determines what the ident token is, and it records only the "raw" form.

[Loirooriol]

CSS identifiers, generically denoted by <ident>, consist of a sequence of characters conforming to the <ident-token> grammar.

The character sequence a, , b doesn't conform to the <ident-token> grammar:

So CSS Values should instead say that the characters are the value of an <ident-token>.