color(srgb / 0.5) seems to parse on safari with the components zeroed out, but that seems wrong per spec. https://w3c.github.io/csswg-drafts/css-color-4/#funcdef-color only says the alpha is optional.
https://bugs.chromium.org/p/chromium/issues/detail?id=1410200 is the relevant chromium bug, for reference. I found this while reviewing patches to implement this in Firefox.
After: document.body.style.color = color(srgb / 0.5) document.body.style.color WebKit 'color(srgb 0 0 0 / 0.5)' Gecko '' Blink 'color(srgb 0 0 0 / 0.5)'
<rdar://problem/104679823>
This used to be correct per spec, long ago when color() also did custom color spaces and it was reasonable to omit some components and have them auto-filled with zero. Now custom color spaces are in CSS Color 5, and clearly distinguished with a dashed-ident; while predefined color spaces are in CSS Color 4 and are wither an RGB space or an XYZ space. Both take 3 components, as the grammar makes clear. There was some leftover prose about variable number of components, not up to date with the grammar change, which I just corrected.
Pull request: https://github.com/WebKit/WebKit/pull/11940
Committed 262098@main (77b199a5f044): <https://commits.webkit.org/262098@main> Reviewed commits have been landed. Closing PR #11940 and removing active labels.