Format at-rule preludes with structured parsing instead of regexes

[bartveneman]

Summary

• format_atrule_prelude reformatted @media/@supports/@container/@layer/@scope/@import preludes with 13 hand-written regexes applied to raw text. This missed cases the regex whitelist didn't cover — e.g. @container STYLE(--foo: bar) never got lowercased, since style wasn't in the hardcoded function-name list — and any new at-rule syntax needs a new regex to work correctly.
• @projectwallace/css-parser (already used for everything else in this library) ships a real structured parser for this that was previously disabled (parse_atrule_preludes: false). Flipped it on and added a new internal printer, used only by print_atrule, that dispatches on each prelude child's node type instead of pattern-matching text.
• No public API changes. format_atrule_prelude — the exported partial formatter documented in the README — is completely untouched: same signature, same regex implementation, same tests. It's now also reused internally as the safety-net fallback whenever the structured parser can't represent something losslessly (an at-rule it doesn't recognize, a comment inside a prelude, or the standalone @layer name; statement form).
• While building this I found and worked around several real bugs in the dependency's structured prelude parser (dotted @layer names getting split at the dot and corrupting cascade semantics if naively reprinted, calc()/env()/other function calls silently dropped from media-feature and @supports condition values, @supports selector(...) returning nothing at all, and a few smaller ones). Full writeup with repro snippets in PARSER_ISSUES.md, since these are worth fixing upstream.
• Verified via differential testing against the previous implementation across ~170 hand-picked and combinatorial cases (in addition to the existing test suite) to catch regressions the unit tests alone wouldn't — only one divergence found on genuinely invalid CSS input, and two on cases where the old regex output was actually wrong (e.g. @namespace svg url(http://...) — the old regex incorrectly inserted a space inside http://, which the new structured path fixes automatically).

Test plan

• vitest run — all 269 tests pass (259 existing, unchanged, + 10 new covering the previously-uncovered cases this rework specifically fixes: @container style() lowercasing, function calls surviving in feature/condition values, dotted @layer names, unrecognized at-rules not losing their prelude, nested @supports boolean groups)
• tsc --noEmit — no type errors
• oxlint / oxfmt --check — clean
• Differential fuzz testing against the pre-change implementation (~170 cases) — see PR description above
• Rough perf sanity check on an at-rule-heavy stylesheet: ~6% slower (structured parsing now happens for every prelude instead of just capturing raw text), not alarming for a correctness fix

---

Generated by Claude Code

[github-actions]

⚠️ Duplicate Dependencies (found: 22, threshold: 1)

📦 Package	📋 Versions
@babel/parser	2 versions root@ @vitest/coverage-v8@4.1.9 magicast@0.5.3 @babel/parser@7.29.7 root@ tsdown@0.22.3 rolldown-plugin-dts@0.26.0 @babel/generator@8.0.0 @babel/parser@8.0.0
@babel/types	2 versions root@ @vitest/coverage-v8@4.1.9 magicast@0.5.3 @babel/parser@7.29.7 @babel/types@7.29.7 root@ tsdown@0.22.3 rolldown-plugin-dts@0.26.0 @babel/generator@8.0.0 @babel/parser@8.0.0 @babel/types@8.0.0
@babel/helper-string-parser	2 versions root@ @vitest/coverage-v8@4.1.9 magicast@0.5.3 @babel/parser@7.29.7 @babel/types@7.29.7 @babel/helper-string-parser@7.29.7 root@ tsdown@0.22.3 ... @babel/parser@8.0.0 @babel/types@8.0.0 @babel/helper-string-parser@8.0.0
@babel/helper-validator-identifier	2 versions root@ @vitest/coverage-v8@4.1.9 magicast@0.5.3 @babel/parser@7.29.7 @babel/types@7.29.7 @babel/helper-validator-identifier@7.29.7 root@ tsdown@0.22.3 ... @babel/parser@8.0.0 @babel/types@8.0.0 @babel/helper-validator-identifier@8.0.2
@rolldown/binding-wasm32-wasi	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-wasm32-wasi@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-wasm32-wasi@1.1.5
picomatch	2 versions root@ @vitest/coverage-v8@4.1.9 ... tinyglobby@0.2.17 fdir@6.5.0 picomatch@4.0.4 root@ @vitest/coverage-v8@4.1.9 vitest@4.1.9 @vitest/mocker@4.1.9 vite@8.1.4 picomatch@4.0.5
rolldown	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 root@ @vitest/coverage-v8@4.1.9 vitest@4.1.9 @vitest/mocker@4.1.9 vite@8.1.4 rolldown@1.1.5
@oxc-project/types	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @oxc-project/types@0.137.0 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @oxc-project/types@0.139.0
@rolldown/binding-android-arm64	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-android-arm64@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-android-arm64@1.1.5
@rolldown/binding-darwin-arm64	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-darwin-arm64@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-darwin-arm64@1.1.5
@rolldown/binding-darwin-x64	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-darwin-x64@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-darwin-x64@1.1.5
@rolldown/binding-freebsd-x64	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-freebsd-x64@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-freebsd-x64@1.1.5
@rolldown/binding-linux-arm-gnueabihf	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-linux-arm-gnueabihf@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-linux-arm-gnueabihf@1.1.5
@rolldown/binding-linux-arm64-gnu	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-linux-arm64-gnu@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-linux-arm64-gnu@1.1.5
@rolldown/binding-linux-arm64-musl	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-linux-arm64-musl@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-linux-arm64-musl@1.1.5
@rolldown/binding-linux-ppc64-gnu	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-linux-ppc64-gnu@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-linux-ppc64-gnu@1.1.5
@rolldown/binding-linux-s390x-gnu	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-linux-s390x-gnu@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-linux-s390x-gnu@1.1.5
@rolldown/binding-linux-x64-gnu	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-linux-x64-gnu@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-linux-x64-gnu@1.1.5
@rolldown/binding-linux-x64-musl	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-linux-x64-musl@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-linux-x64-musl@1.1.5
@rolldown/binding-openharmony-arm64	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-openharmony-arm64@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-openharmony-arm64@1.1.5
@rolldown/binding-win32-arm64-msvc	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-win32-arm64-msvc@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-win32-arm64-msvc@1.1.5
@rolldown/binding-win32-x64-msvc	2 versions root@ tsdown@0.22.3 rolldown@1.1.3 @rolldown/binding-win32-x64-msvc@1.1.3 root@ @vitest/coverage-v8@4.1.9 ... vite@8.1.4 rolldown@1.1.5 @rolldown/binding-win32-x64-msvc@1.1.5

💡 To find out what depends on a specific package, run: pnpm -r why example-package

⚠️ Package Size Increase

📦 Package	📏 Base Size	📏 Source Size	📈 Size Change
@projectwallace/format-css	8.6 kB	11.8 kB	+3.2 kB