fix(ci): accept pull request bodies with CRLF line endings

The contribution validator matched `^## <heading>\n`, so a body sent with
CRLF line endings reported every required section as missing. Correct
submissions were rejected based on the client that composed them.

Match `\r?\n` instead, and add tests for the validator to the lint job.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: nibzard

.github/workflows/lint.yaml

@@ -21,6 +21,8 @@ jobs:
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
+      - name: Test the validator
+        run: python3 -m unittest discover -s scripts -p "test_*.py"
       - name: Validate contribution policy
         run: python3 scripts/validate_contribution.py
 

scripts/test_validate_contribution.py

@@ -0,0 +1,67 @@
+# ABOUTME: Tests for the contribution policy validator used by the lint workflow.
+# ABOUTME: Run with `python3 -m unittest discover -s scripts -p "test_*.py"`.
+
+from __future__ import annotations
+
+import unittest
+
+from validate_contribution import REQUIRED_BODY_SECTIONS, extract_body_sections, normalize_section
+
+
+BODY = """## Summary
+
+Adds an item.
+
+## Section
+
+Dev Tools
+
+## Why this belongs
+
+It controls a browser.
+
+## Primary documented web-agent use case
+
+https://example.com/docs
+
+## Public reference
+
+https://example.com
+
+## Affiliation disclosure
+
+I maintain the project.
+
+## Checklist
+
+- [x] This PR adds exactly one item.
+"""
+
+
+class ExtractBodySectionsTest(unittest.TestCase):
+    def test_reads_every_required_section(self) -> None:
+        sections = extract_body_sections(BODY)
+        for heading in REQUIRED_BODY_SECTIONS:
+            self.assertTrue(sections.get(heading), f'missing "{heading}"')
+
+    def test_reads_every_required_section_with_carriage_returns(self) -> None:
+        # Some clients send the pull request body with CRLF line endings.
+        sections = extract_body_sections(BODY.replace("\n", "\r\n"))
+        for heading in REQUIRED_BODY_SECTIONS:
+            self.assertTrue(sections.get(heading), f'missing "{heading}"')
+
+    def test_section_name_has_no_carriage_return(self) -> None:
+        sections = extract_body_sections(BODY.replace("\n", "\r\n"))
+        self.assertEqual(normalize_section(sections["Section"]), "Dev Tools")
+
+    def test_absent_section_is_not_reported(self) -> None:
+        sections = extract_body_sections(BODY.replace("## Public reference", "## Other"))
+        self.assertIsNone(sections.get("Public reference"))
+
+    def test_empty_section_is_not_reported(self) -> None:
+        sections = extract_body_sections(BODY.replace("https://example.com\n\n## Affiliation", "\n## Affiliation"))
+        self.assertFalse(sections.get("Public reference"))
+
+
+if __name__ == "__main__":
+    unittest.main()

scripts/validate_contribution.py

@@ -75,8 +75,9 @@ def added_item_lines(base_sha: str) -> list[tuple[str, str]]:
 def extract_body_sections(body: str) -> dict[str, str]:
     sections: dict[str, str] = {}
     for heading in REQUIRED_BODY_SECTIONS:
+        # Some clients send the pull request body with CRLF line endings.
         pattern = re.compile(
-            rf"(?ms)^## {re.escape(heading)}\n+(.*?)(?=^## |\Z)"
+            rf"(?ms)^## {re.escape(heading)}\r?\n+(.*?)(?=^## |\Z)"
         )
         match = pattern.search(body)
         if match: