[css-color-5] Consider moving colorspace parameter in color-adjust() to beginning to match color-mix() and ease parsing

[weinig]

In the current CSS Color 5 draft's color-adjust() function, https://drafts.csswg.org/css-color-5/#coloradjust, the colorspace is argument is last, which is inconsistent with color-mix(), where it is first, and makes parsing a bit more complicated since we don't know which adjusters are allowed until after we have parsed them.

Please consider moving the colorspace to the beginning, changing the grammar from:

color-adjust() = color-adjust( <color> [ color-adjuster <colorspace>? ]? )

to

color-adjust() = color-adjust(<colorspace>?, <color> [ color-adjuster ]? )

[Myndex]

How about moving the "colorspace" in color-mix to the end? And also, not calling it colorspace as in most cases it is not a different colorspace and is only a different color notation?

[samweinig]

Moving colorspace to the end of color-mix() would make parsing quite a bit more inefficient as you would not know which adjusters were valid as you were parsing them. If adjusters are going to be validated based on colorspace, I think it makes sense for it to come first.

[LeaVerou]

We actually made some changes to the color space argument, mandating an in keyword before it which allows it to be anywhere in the function. Could you take another look?
We also removed adjusters, so that shouldn't be an issue now.

[samweinig]

The current spec (or what I get when I load https://drafts.csswg.org/css-color-5/#coloradjust anyway) has the following grammar now:

color-adjust() = color-adjust( <color> [ color-adjuster [ in <colorspace> ]? ]? )

Which seems to still have adjusters (though it is no longer defined what a color-adjuster production is).

[LeaVerou]

Whoops, I completely missed that this was about color-adjust() and not color-mix().
I should also resurrect the color adjuster definition and move it there. Thanks!

Edit: done

[svgeesus]

Closed by WG resolution to drop color-adjust()