[css-transforms-2] Potentially invalid syntax for `perspective()`

[pyoor]

The CSS Transforms 2 spec currently defines the perspective() function as:
perspective( <length [0,∞]> | none )

In the syntax above, the functional notation accepts one of two arguments (using the single bar operator). Now, I can't find anything specifically in css-values-5 that forbids this, but I believe that these arguments should be wrapped in a group as in:
perspective( [ <length [0,∞]> | none ] )

Could some please provide clarification on if this actually is an issue?

csswg-drafts/css-transforms-2/Overview.bs

Line 1003 in 77f40b2

: <span class='prod'><dfn>perspective()</dfn> = perspective( <<length [0,∞]>> | <dfn id="valdef-perspective-func-none" dfn-type="value" for="perspective()">none</dfn> )</span>

[tabatkins]

Yup, that's an error.

[cdoublev]

Is it a syntax style error, or more? Where is this style defined?

• CSS Overflow 4 has fade( <length> | <percentage> )
• Filter Effects 1 has drop-shadow( <color>? && <length>{2,3} )

[tabatkins]

The Syntax extension of the CSS grammar defines how at-rules and functions (and other things) are written in the grammar. It doesn't have any special rules for functions (or anything else) to automatically create groups in the grammar.

Perhaps it should, tho? Hm.

[cdoublev]

It doesn't have any special rules for functions (or anything else) to automatically create groups in the grammar.

What are the cases (other than function arguments) where a group should be created automatically? Do these rules also explain why a group is created for the types specific to font-variant-alternates in the value definition of font-variant?

I guess the reason for creating a group that is not strictly required, is to make a grammar parser produce a nested list with the corresponding value(s). From this perspective, it would have been nice to create groups for the types specific to font-variant-ligatures, font-variant-numeric, and font-variant-east-asian. It would make shorthand expansion easier, imo.

I'm sorry for this noise on a closed issue. Please tell me if I should create a new issue.

Related: #2921

  normal 
| none 
| [
       # font-variant-ligatures
       <common-lig-values> || <discretionary-lig-values> || <historical-lig-values> || <contextual-alt-values> 
       # font-variant-caps
    || [small-caps | all-small-caps | petite-caps | all-petite-caps | unicase | titling-caps] 
       # font-variant-alternates
    || [
            stylistic(<feature-value-name>) 
         || historical-forms 
         || styleset(<feature-value-name>#) 
         || character-variant (<feature-value-name>#) 
         || swash(<feature-value-name>) 
         || ornaments(<feature-value-name>) 
         || annotation(<feature-value-name>)
       ] 
       # font-variant-numeric
    || <numeric-figure-values> || <numeric-spacing-values> || <numeric-fraction-values> || ordinal || slashed-zero 
       # font-variant-east-asian
    || <east-asian-variant-values> || <east-asian-width-values> || ruby 
       # font-variant-position
    || [sub | super] 
       # font-variant-emoji
    || [text | emoji | unicode] 
 ]