scripts/validate_docs_sync.py

@@ -0,0 +1,134 @@
+#!/usr/bin/env python3
+"""Validate top-level docs synchronization invariants."""
+
+from __future__ import annotations
+
+import re
+import sys
+from pathlib import Path
+
+REPO_ROOT = Path(".")
+REQUIRED_FILES = {
+    "README.md": REPO_ROOT / "README.md",
+    "CLAUDE.md": REPO_ROOT / "CLAUDE.md",
+    "commands": REPO_ROOT / "docs" / "commands.md",
+    "testing": REPO_ROOT / "docs" / "testing.md",
+    "workflow": REPO_ROOT / "docs" / "workflow.md",
+}
+
+LINK_PATTERN = re.compile(r"\[[^\]]+\]\((?P<link>[^)]+)\)")
+ENDPOINT_ROW_PATTERN = re.compile(
+    r"^\|\s*`(?P<endpoint>(?:GET|POST|PUT|PATCH|DELETE)\s+/[^`]*)`\s*\|"
+)
+
+
+def _read(path: Path) -> str:
+    return path.read_text(encoding="utf-8")
+
+
+def validate_required_files_exist(errors: list[str]) -> None:
+    for name, path in REQUIRED_FILES.items():
+        if not path.exists():
+            errors.append(f"missing required doc file ({name}): {path}")
+
+
+def validate_links_resolve(doc_path: Path, text: str, errors: list[str]) -> None:
+    for match in LINK_PATTERN.finditer(text):
+        raw_link = match.group("link").strip()
+        if not raw_link or raw_link.startswith("#") or raw_link.startswith("http"):
+            continue
+        link_path = raw_link.split("#", 1)[0].strip()
+        if not link_path:
+            continue
+        if link_path.startswith("/"):
+            errors.append(f"{doc_path}: absolute link is forbidden -> {raw_link}")
+            continue
+        target = (doc_path.parent / link_path).resolve()
+        if not target.exists():
+            errors.append(f"{doc_path}: broken link -> {raw_link}")
+
+
+def validate_summary_docs_reference_core_docs(errors: list[str]) -> None:
+    required_links = {
+        "README.md": ("docs/workflow.md", "docs/commands.md", "docs/testing.md"),
+        "CLAUDE.md": ("docs/workflow.md", "docs/commands.md"),
+    }
+    for file_name, links in required_links.items():
+        doc_path = (
+            REQUIRED_FILES["README.md"]
+            if file_name == "README.md"
+            else REQUIRED_FILES["CLAUDE.md"]
+        )
+        text = _read(doc_path)
+        for link in links:
+            if link not in text:
+                errors.append(f"{doc_path}: missing core doc link reference -> {link}")
+
+
+def collect_command_endpoints(text: str) -> list[str]:
+    endpoints: list[str] = []
+    for line in text.splitlines():
+        match = ENDPOINT_ROW_PATTERN.match(line.strip())
+        if match:
+            endpoints.append(match.group("endpoint"))
+    return endpoints
+
+
+def validate_commands_endpoint_duplicates(errors: list[str]) -> None:
+    text = _read(REQUIRED_FILES["commands"])
+    endpoints = collect_command_endpoints(text)
+    seen: set[str] = set()
+    duplicates: set[str] = set()
+    for endpoint in endpoints:
+        if endpoint in seen:
+            duplicates.add(endpoint)
+        seen.add(endpoint)
+    for endpoint in sorted(duplicates):
+        errors.append(f"docs/commands.md: duplicated API endpoint row -> {endpoint}")
+
+
+def validate_testing_doc_has_dynamic_count_guidance(errors: list[str]) -> None:
+    text = _read(REQUIRED_FILES["testing"])
+    if "pytest --collect-only -q" not in text:
+        errors.append(
+            "docs/testing.md: missing dynamic test count guidance "
+            "(pytest --collect-only -q)"
+        )
+
+
+def main() -> int:
+    errors: list[str] = []
+
+    validate_required_files_exist(errors)
+    if errors:
+        print("[FAIL] docs sync validation failed")
+        for err in errors:
+            print(f"- {err}")
+        return 1
+
+    readme_text = _read(REQUIRED_FILES["README.md"])
+    claude_text = _read(REQUIRED_FILES["CLAUDE.md"])
+    validate_links_resolve(REQUIRED_FILES["README.md"], readme_text, errors)
+    validate_links_resolve(REQUIRED_FILES["CLAUDE.md"], claude_text, errors)
+    validate_links_resolve(REQUIRED_FILES["commands"], _read(REQUIRED_FILES["commands"]), errors)
+    validate_links_resolve(REQUIRED_FILES["testing"], _read(REQUIRED_FILES["testing"]), errors)
+
+    validate_summary_docs_reference_core_docs(errors)
+    validate_commands_endpoint_duplicates(errors)
+    validate_testing_doc_has_dynamic_count_guidance(errors)
+
+    if errors:
+        print("[FAIL] docs sync validation failed")
+        for err in errors:
+            print(f"- {err}")
+        return 1
+
+    print("[OK] docs sync validated")
+    print("[OK] summary docs link to core docs and links resolve")
+    print("[OK] commands endpoint rows have no duplicates")
+    print("[OK] testing doc includes dynamic count guidance")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

tests/test_validate_docs_sync.py

@@ -0,0 +1,102 @@
+from __future__ import annotations
+
+import importlib.util
+from pathlib import Path
+
+
+def _load_module():
+    script_path = Path(__file__).resolve().parents[1] / "scripts" / "validate_docs_sync.py"
+    spec = importlib.util.spec_from_file_location("validate_docs_sync", script_path)
+    assert spec is not None
+    assert spec.loader is not None
+    module = importlib.util.module_from_spec(spec)
+    spec.loader.exec_module(module)
+    return module
+
+
+def test_collect_command_endpoints_parses_markdown_table_rows() -> None:
+    module = _load_module()
+    text = "\n".join(
+        [
+            "| Endpoint | Description |",
+            "|----------|-------------|",
+            "| `GET /api/status` | status |",
+            "| `POST /api/run` | run |",
+            "| not-a-row | ignored |",
+        ]
+    )
+    endpoints = module.collect_command_endpoints(text)
+    assert endpoints == ["GET /api/status", "POST /api/run"]
+
+
+def test_validate_links_resolve_detects_absolute_and_broken_links(tmp_path) -> None:
+    module = _load_module()
+    doc = tmp_path / "doc.md"
+    existing = tmp_path / "ok.md"
+    existing.write_text("# ok\n", encoding="utf-8")
+    doc.write_text(
+        "\n".join(
+            [
+                "[ok](./ok.md)",
+                "[abs](/tmp/nowhere.md)",
+                "[broken](./missing.md)",
+            ]
+        ),
+        encoding="utf-8",
+    )
+    errors: list[str] = []
+    module.validate_links_resolve(doc, doc.read_text(encoding="utf-8"), errors)
+
+    assert any("absolute link is forbidden" in err for err in errors)
+    assert any("broken link" in err for err in errors)
+
+
+def test_validate_summary_docs_reference_core_docs(monkeypatch) -> None:
+    module = _load_module()
+    errors: list[str] = []
+    fake_docs = {
+        str(module.REQUIRED_FILES["README.md"]): (
+            "docs/workflow.md docs/commands.md docs/testing.md"
+        ),
+        str(module.REQUIRED_FILES["CLAUDE.md"]): "docs/workflow.md docs/commands.md",
+    }
+
+    def fake_read(path: Path) -> str:
+        return fake_docs[str(path)]
+
+    monkeypatch.setattr(module, "_read", fake_read)
+    module.validate_summary_docs_reference_core_docs(errors)
+    assert errors == []
+
+
+def test_validate_commands_endpoint_duplicates_reports_duplicates(monkeypatch) -> None:
+    module = _load_module()
+    errors: list[str] = []
+    text = "\n".join(
+        [
+            "| `GET /api/status` | status |",
+            "| `GET /api/status` | duplicate |",
+        ]
+    )
+
+    def fake_read(path: Path) -> str:
+        assert path == module.REQUIRED_FILES["commands"]
+        return text
+
+    monkeypatch.setattr(module, "_read", fake_read)
+    module.validate_commands_endpoint_duplicates(errors)
+    assert errors
+    assert "duplicated API endpoint row -> GET /api/status" in errors[0]
+
+
+def test_validate_testing_doc_has_dynamic_count_guidance(monkeypatch) -> None:
+    module = _load_module()
+    errors: list[str] = []
+
+    def fake_read(path: Path) -> str:
+        assert path == module.REQUIRED_FILES["testing"]
+        return "Use pytest --collect-only -q for dynamic counts."
+
+    monkeypatch.setattr(module, "_read", fake_read)
+    module.validate_testing_doc_has_dynamic_count_guidance(errors)
+    assert errors == []