Add support for layer declarations ending with semicolon

[samuil-banti-wpenigne]

Semicolon-terminated @layer is valid CSS but was parsed as a block, so later rules were swallowed.

@layer has two forms (CSS Cascade 5):

@layer reset;
@layer base, components;
@layer theme { .button { color: blue; } }

Generic at-rule parsing always searched for {. For @layer reset;, that meant consuming through the next { (e.g. @property), so the rest of the stylesheet was treated as nested contents of a fake layer block.

Fix: stop at { or ;. A ; becomes AtRuleStatement; a { still becomes AtRuleBlockList or AtRuleSet. layer stays in BLOCK_RULES, so @layer theme { … } is unchanged.

[coveralls]

coverage: 75.842% (+0.4%) from 75.487% — samuil-banti-wpenigne:add-support-for-layer-declaration-ending-with-semicolon into MyIntervals:main

[Copilot]

Pull request overview

Adds support for semicolon-terminated at-rules without consuming subsequent CSS rules.

Changes:

• Adds AtRuleStatement parsing and rendering.
• Preserves block-form @layer behavior.
• Adds unit and parser regression tests.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
src/CSSList/CSSList.php	Distinguishes statement and block at-rules.
src/Property/AtRuleStatement.php	Represents semicolon-terminated at-rules.
tests/CSSList/AtRuleStatementTest.php	Tests parsing and rule boundaries.
tests/Unit/Property/AtRuleStatementTest.php	Tests the new representation.
CHANGELOG.md	Documents the fix.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[JakeQZ]

It's the same with {. Fixing it is beyond the scope of this PR. #1625 added.

[JakeQZ]

Wow. Excellent contribution @samuil-banti-wpenigne. Hard to pick any holes in that.

The functional tests for the new class should be in a different location (and namespace), and I don't think we need a changelog entry that tests have been added.

Otherwise, looks perfect. Thank you so much, particularly for taking the time to understand the codebase.

@oliverklee, do you have any other nitpicks?

[JakeQZ]

LGTM now. One of the best PR submissions we've ever had. Thank you.

@oliverklee, anything else to pick up on?

phar.io server is currently down, causing failing checks...

[oliverklee]

@JakeQZ I most probably won't have time to look at this today or tomorrow, and I don't want to block this. So feel free to merge in the meantime.

[JakeQZ]

@JakeQZ I most probably won't have time to look at this today or tomorrow, and I don't want to block this. So feel free to merge in the meantime.

OK. Happy to do so so that people can pick up the -dev branch. The code change looks solid, the new class located with its siblings (which may all be relocated at some point), and the tests good.

The only thing I can think of that might be falling short is whether the tests cover every eventuality and format of at-rules. But they never will - there's always an edge case that has been missed. AI was surprisingly smart at picking up the pre-existing lack of string parsing, which has been logged as #1625. Any other issues found can be resolved as a follow-up.

Thanks again @samuil-banti-wpenigne for your excellent contribution.