[css-values-5] Should random() and random-item() share indexing?

[fantasai]

This came up in https://bugs.webkit.org/show_bug.cgi?id=319919 / WebKit/WebKit#71594 ... The spec says:

the index of the random function among all the random functions used in the same property value

But it's not clear to me that this should be "among all random functions of any type, grouping random() and random-item() together" vs "among all random functions of a given type, treating random() and random-item() separately".

The goal is for the random functions to be maximally random, but this is more simply accomplished by giving each type of random function its own bucket, i.e. making the function name part of the random caching key. That also avoids complications with these functions resolving at different times.

[tabatkins]

My intention was "together", but the details are unimportant except insofar as we get something specified. I agree that it's probably easier to make them separate buckets.

[astearns]

Does this only apply to property-index-scoped cache keys, or more generally?

[tabatkins]

Just index-scoping, yeah. That's the only thing that cares about the function itself.

[astearns]

The CSSWG will automatically accept this resolution one week from now if no objections are raised here. Anyone can add an emoji to this comment to express support. If you do not support this resolution, please add a new comment.

Proposed Resolution: random() and random-item() get separate buckets for property-index-scoped cache keys