[css-color-adjust-1] User origin !important vs. forced colors mode

[andruud]

So let's say I have a user stylesheet:

color: red !important;

In forced colors mode, what applies? The spec says revert !important is added to the cascade at the user level (when?), which may or may not win over my red !important.

I don't think it's enough to state that the value is added at the user level. Clearly we need revert !very important.

cc @fantasai, @lilles

[lilles]

I don't know what the intention was here. I would guess applied after other user stylesheets.

[AmeliaBR]

The intention is to revert author styles. I don't think we're trying to revert user !Important styles. So maybe it should say "user level important styles, before any user style sheets".

[andruud]

If the intent is to revert author styles, then we must revert at the author origin, in which case we get the same problem for an author-level declaration color: red !important.

[fantasai]

Yeah, you're right, this isn't specified quite correctly. It reverts everything in the author origin as intended, but also reverts everything in the user origin as well. For it to work ideally (preserving user styles), we'd need to have an origin between user and author.

We kindof already have this, in the form of the "non-CSS presentation hints" layer. If we turn that into a real origin and give it a corresponding !important level, then we can inject these rules into that origin and they should work as intended.

Agenda+ to discuss.

[andruud]

@fantasai That makes sense. But let's definitely not make the new origin something that can be revert-ed to (only from). revert is already complicated and expensive enough, so revert in the author origin should still revert to the user origin IMO.

[emilio]

FWIW the way Firefox implements this is just treating author-level declarations for those properties as revert: https://searchfox.org/mozilla-central/rev/557a0e222dd104c5d805ba344c45d6abc27d3db0/servo/components/style/properties/cascade.rs#406

[emilio]

I think the spec mostly wants to say that any declaration for those properties in the author origin is treated as revert, or something of that sort.

[css-meeting-bot]

The CSS Working Group just discussed [css-color-adjust-1] User origin !important vs. forced colors mode, and agreed to the following:

• RESOLVED: Option 3- Rewrite all author rules for affected properties to specify 'revert' instead

The full IRC log of that discussion

<dael> Topic: [css-color-adjust-1] User origin !important vs. forced colors mode
<dael> github: https://github.com//issues/4020
<dael> fantasai: The spec says we add revert !important at user level to wipe out author level. Also reverts entire user level of cascade
<dael> fantasai: I think intent was leave user styles.
<dael> fantasai: If we want to do that correctly need origin between user and author. Sort of have this which is non-css presentation layer. It's beginning of author and has no spec
<dbaron> could these rules just be in the author level (at the top)?
<dael> fantasai: We could define there's a new UA control layer called presentation-hints and importance is between user and author and then it works as intended. Or we can do something else. Currently spec doesn't work as intended
<dael> AmeliaBR: If we can explain behavior using source order; say this is inserted at origin before any other rules at origin; I think that's easier than new cascade origin. If that doesn't work then define it explicitly
<dael> fantasai: emilio says they treat all author level as revert so we can spec that way.
<dael> astearns: Is this something where we have more tools to solve when we get to custom origins prop?
<dael> fantasai: No b/c need to revert...revert in author needs to revert non-css presentaiton hints. So we can create a new origin or do what emilio says FF is doing
<dael> astearns: Set of properties?
<dael> fantasai: [missed]
<dael> emilio: I don't understand why revert !important wouldn't work at end of author. Might be easier to spec that way. In practice I think the obserables are same as FF.
<dael> fantasai: I'm happy to put in FF behavior
<dael> astearns: Any concerns with specify FF behavior?
<dael> astearns: Reading emilio comment: Treat author level declarations with a set of properties as revert
<dael> fantasai: 3 solutions. Creat new origin
<fantasai> 1. Create new origin between author / user origins
<fantasai> 2. Insert 'revert !important' rules in author orgin with highest possible specificity
<dael> fantasai: I think 3 is what emilio is saying FF does
<fantasai> 3. Rewrite all author rules for affected properties to specify 'revert' instead
<dael> fantasai: All of those would work. I don't have an opinion between.
<dael> astearns: I think 3 is easiest to explain to authors. Just my personal opinion
<dael> fantasai: That's a vote for easy to explain and a vote from FF it's easy to impl
<dael> Rossen_: I think #3 makes sense
<dael> astearns: AmeliaBR okay to you?
<dael> AmeliaBR: Yeah. So long as clearly defined.
<dael> astearns: Prop: Option 3- Rewrite all author rules for affected properties to specify 'revert' instead
<dael> astearns: Objections?
<dael> RESOLVED: Option 3- Rewrite all author rules for affected properties to specify 'revert' instead
<dael> astearns: Thanks emilio for adding FF behavior