W3C home > Mailing lists > Public > public-css-archive@w3.org > March 2020

Re: [csswg-drafts] [css-media-queries-5] add clarification on applied order for multiple preferences (#4895)

From: Thomas Gossmann via GitHub <sysbot+gh@w3.org>
Date: Thu, 26 Mar 2020 19:22:58 +0000
To: public-css-archive@w3.org
Message-ID: <issue_comment.created-604634707-1585250576-sysbot+gh@w3.org>
> There's no hierarchy; they're independent preferences that can be queried separately or together, as the page author desires.
>

+

> > Can I define contrast and color scheme independently, so the cascade can do its job?
> 
> Not really right now, because dealing with either preference involves assigning to the same properties (`color`, `background-color`, etc), and ideally you want to handle all combinations of those two preferences.

= hierarchy, because to handle all combinations I need to setup my queries that way, e.g. color-scheme and then contrast and results in those matrices of query (+ selectors)

> _That said_, Color 5 is working on a (still _very_ early stages, don't look to its details as particularly informative of a final design yet) [automatic contrast adjuster](https://drafts.csswg.org/css-color-5/#colorcontrast), where you can feed it two colors and a desired contrast ratio and it spits out one of the colors adjusted appropriately. This should, if we do it right, let you do a quick-and-dirty contrast adjuster automatically: set a `--contrast-ratio` property according to `(prefers-contrast)`, and set `--color-*` properties according to `(prefers-color-scheme)`, then actually write out your color properties by combining the two with `contrast-adjust()`. It's a bit of scaffolding, but it avoids needing to define a 3x2 (for now) set of colors for everything; instead you'd just define 2 sets of colors (light and dark) and let the function deal with contrast for you. The results won't be as good as hand-designing all the combinations, but it should be better than nothing, and you can still do hand-authoring when desired, either totally or piecemeal.

This seems like tackling the problem of spawning the matrices. Thanks for pointing out, will keep an eye on this one.


-- 
GitHub Notification of comment by gossi
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/4895#issuecomment-604634707 using your GitHub account
Received on Thursday, 26 March 2020 19:23:00 UTC

This archive was generated by hypermail 2.4.0 : Tuesday, 5 July 2022 06:42:02 UTC