Theme: fix color ramp generation to avoid unmet contrasts #73331
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
jsnajdr
added
[Type] Enhancement
|
The following accounts have interacted with this PR and/or linked issues. I will continue to update these lists as activity occurs. You can also manually ask me to refresh this list by adding the If you're merging code through a pull request on GitHub, copy and paste the following into the bottom of the merge commit message. To understand the WordPress project's expectations around crediting contributors, please review the Contributor Attribution page in the Core Handbook. |
jsnajdr
changed the title
|
Size Change: -15 B (0%) Total Size: 2.49 MB
ℹ️ View Unchanged
|
|
Flaky tests detected in e8bce5d. 🔍 Workflow run URL: https://github.com/WordPress/gutenberg/actions/runs/19411403142
|
aduth
approved these changes
Nov 17, 2025
Member
aduth
left a comment
aduth
left a comment
There was a problem hiding this comment.
LGTM 👍
Couple side-observations:
- The "Sample Combinations" story shows some warnings, though it doesn't appear to have regressed (or made any better) with the changes here
- On your point about P3 -> hex conversions, would these potential inaccuracies extend to any of the internal P3 usage?
Member
Author
The two questions are related. Why are there warnings in the sample combinations story? One of the color ramps that fail is background ramp for WP Blue, 
What does this chart show?
Now, this seed color has insufficient contrast against white. It's just 6.63, while We could get a darker color by changing (lowering) the chroma C, by making the color less blue and more gray/black. That's what the
The only other internal usage of P3 is in |
bph
added
the
Design System
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Several tweaks to the color ramp generation algorithm that ensure that contrast targets are always reliably met.
@aduth reported that the Storybook color scales story reports unmet contrast targets:
The real contrasts are not 4.5 and 3.0 but only 4.47 and 2.99, respectively. I'm addressing the reason why these slight mismatches happen.
I'm also making adjustments to the
stroke1andstroke2target contrasts that @jameskoster proposed recently. They are quite unrelated, but not entirely: this PR changes the generated colors in three distinct ways, and I wanted to see the entire effect of all three changes together.Change number 1: change
clampToGamutto use the sRGB color space instead of P3. When we generate a color, we eventually store it in the RGB hex format, like#c8ba80. As part of this formatting, colors that are outside the sRGB gamut are automatically mapped to sRGB, otherwise they couldn't be expressed in the hex format. Here it can happen that two colors that previously had sufficient contrast in a wide-gamut space (like P3), have a slightly lower contrast after the sRGB conversion. That's one way how the contrast target could be missed. WithclampToGamutconverting to sRGB, we use sRGB colors during the actual color search and its contrast checks.I'm also moving
clampToGamutto the color-utils.ts file, as that's a nicer place for the function.Change number 2: Increase the "contrast topup" from 0.012 to 0.02. It's a safer value for edge cases when the lighter color is rounded down, to a slightly darker color, and the darker color is rounded up, to a slightly lighter color. Then the contrast can have as big as a 0.016 difference. This was another reason why target contrasts were missed sometimes.
Change number 3: Apply @jameskoster's suggestion to change contrast targets for
stroke1andstroke2, so that the two colors are less prominent against the background surface. If these changes were applied alone, then for the default background ramp (#f8f8f8seed) thestroke1andstroke2colors were#e1e1e1and#d8d8d8. When all three tweaks in this PR are applied together, the colors are#e0e0e0and#d8d8d8. A minimal difference. This difference I what I wanted to see, and verify that the other two changes don't change the colors substantially.How to test: In Storybook, look at the
/story/design-system-theme-provider-color-scales--defaultstory, and verify that the contrast mismatch warnings are gone.