Two "variants" of `color-mix()` ??

[svgeesus]

In CSS Values 5 I was astonished to read 5.3. Interpolated Color Values: the color-mix() notation

This specification extends the color-mix() functional notation as a mix notation accepting the following syntaxes: ``` = color-mix( && ?, , ) | color-mix( , [ && ?]#{2} ) ``` The used value of the first mix notation variant is equivalent to assigning the <progress> value, as a <percentage>, to the of the second <color> argument in the second variant. That is, color-mix(progress, color1, color2) is equivalent to color-mix(color1, color2 progress). See CSS Color 5 § 3 Mixing Colors: the color-mix() Function for the normative definition of the second variant.

Why is there a second "variant" of color-mix() defined, with zero discussion that I can recall, which is merely a permutation of the parameter order, and has zero tests in WPT, unlike the other "variant" which is widely implemented, with 1510 tests in WPT and part of Interop 2023?

[SebastianZ]

This was added in https://github.com/w3c/csswg-drafts/commit/0d8967a097ba06aeaba3375d636004b8e9575b08, which references the related discussions, especially https://github.com/w3c/csswg-drafts/issues/6245#issuecomment-1813010786. @fantasai can surely explain more. Though it looks like it was added as an initial draft for generalized interpolation functions while the discussion is still going on.

Sebastian

[tabatkins]

Thanks for digging up the ref, @SebastianZ! Yes, we discussed this last year; the reasoning is to give all the mixing functions a consistent grammar form so they can all benefit from all the mixing features, even if they have a legacy grammar that looks a little different.

[svgeesus]

Thanks, @SebastianZ and @tabatkins

So, are any browsers implementing this, or expressing interest?

[tabatkins]

At the moment, all of the mix functions are not yet implemented. But the mix functions in general have implementor interest; we needed them to solve font palette mixing, for example.