fix: don't count keyframe selectors in selector-complexity

[ump45nose]

What does this change?

css/selector-complexity counted the from, to and percentage selectors of an @keyframes rule toward maxTypes, because css-tree parses them as TypeSelector nodes even though they name no element. A stylesheet whose only "type selectors" were keyframe selectors therefore failed with { maxTypes: 0 }:

/* eslint css/selector-complexity: ["error", { "maxTypes": 0 }] */

@keyframes fade {
	from { opacity: 0; }
	to { opacity: 1; }
}

4:2  Exceeded maximum type selector. Only 0 allowed.
8:2  Exceeded maximum type selector. Only 0 allowed.

The fix tracks whether the selector sits inside a keyframes rule and skips those. It uses the same technique as the existing no-duplicate-keyframe-selectors rule, because @eslint/css's language exposes no parent links on AST nodes (I first tried walking node.parent and confirmed the property is undefined).

Vendor-prefixed keyframes rules are handled by the same regex that rule uses.

Fixes #582

Testing

Three cases added to tests/rules/selector-complexity.test.js:

• @keyframes fade { from {...} to {...} } with { maxTypes: 0 } — the reported repro.
• @keyframes slide { 0% {...} 100% {...} } with { maxTypes: 0 } — percentage selectors.
• @keyframes followed by a .foo rule, and @keyframes followed by div span with { maxTypes: 1 } which must still report — proving the flag is cleared on the way out, not just set.

Verified

• The two keyframe valid cases fail on main with exactly the reported messages (Should have no errors but had 2: ... Exceeded maximum type selector) and pass with the fix — confirmed by stashing the rule and re-running.
• Full suite npx mocha "tests/**/*.test.js": 1236 passing, 0 failing (was 1233 before, +3 from this change).
• npm run build (which also runs the rule-docs regeneration) exits 0.
• npm run lint (the repo's own eslint) and prettier --check are clean.

Not verified

• A browser/editor run — no visual change is involved; the rule's behaviour is fully covered by the RuleTester cases.

Docs

Added a short note plus example to docs/rules/selector-complexity.md under the existing maxTypes section. The rule description and options are unchanged.

Summary by CodeRabbit

• Bug Fixes

  • Keyframe selectors such as from, to, and percentages no longer count toward the type-selector limit. Type selectors outside keyframes continue to be checked as before.

• Documentation

  • Added an example showing how keyframe selectors interact with the maxTypes setting.

[coderabbitai]

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

• Configuration used: defaults
• Review profile: CHILL
• Plan: Advanced
• Run ID: 0a07adfb-8a60-45fb-8e5d-29c123e8233e

📥 Commits

Reviewing files that changed from the base of the PR and between 2e80de0 and 1996e31.

📒 Files selected for processing (3)

• docs/rules/selector-complexity.md
• src/rules/selector-complexity.js
• tests/rules/selector-complexity.test.js

Included review availability: This review used your included allowance. Your plan provides up to 2 included reviews per hour; 1 remain after this review.

---

📝 Walkthrough

Walkthrough

The selector-complexity rule now excludes type selectors inside standard and vendor-prefixed keyframes. Tests cover keyframe selectors and selectors after keyframes. The rule documentation includes an example.

Changes

Keyframe type-selector counting

Layer / File(s)	Summary
Track keyframes and exclude their type selectors src/rules/selector-complexity.js, tests/rules/selector-complexity.test.js, docs/rules/selector-complexity.md	The rule tracks entry into and exit from standard and vendor-prefixed keyframes, and excludes type selectors inside them. Tests cover keyframe selectors and selectors after keyframes. The documentation includes a maxTypes: 0 example.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~10 minutes

Change: Bug fix · Severity of issue fixed: Low

Suggested reviewers: pixel998

Merge Risk: ⚪ Minimal · up to 1996e

The change excludes keyframe step type selectors while continuing to check ordinary selectors afterward. No actionable issue is established, so the PR appears mergeable subject to normal project checks.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: excluding keyframe selectors from selector-complexity counts.
Linked Issues check	✅ Passed	Issue #582 requires keyframe selectors to not count toward maxTypes. The rule tracks entry into and exit from standard and supported vendor-prefixed @keyframes at-rules, then excludes `TypeSelecto…
Out of Scope Changes check	✅ Passed	The source change, tests, and documentation all support issue #582. The tests also verify that selectors after a keyframes rule remain subject to maxTypes. No unrelated changes appear in the pull re…
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 2…

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create a new PR

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[DMartens]

The referenced issue is neither accepted nor assigned to you.
In the future please follow our contribution guidelines.