llm_wiki/tools/scripts/validate-frontmatter.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
validate-frontmatter.py — Pre-commit hook: 验证被修改的 wiki 页面 frontmatter

检查项：
- 必须包含 categories（含 [[LLM Wiki]]）
- 必须包含 tags（含 wiki）
- 必须包含 type（合法值列表）
- 必须包含 source
- 如果 status=superseded，必须包含 superseded_by
- relations 的 target 必须指向存在的页面

用法：
    python tools/scripts/validate-frontmatter.py                    # 所有 wiki 页面
    python tools/scripts/validate-frontmatter.py --files file1.md file2.md  # 指定文件
    python tools/scripts/validate-frontmatter.py --git-hook         # 从 git diff 读取
"""
import argparse
import re
import subprocess
import sys
from pathlib import Path

try:
    import yaml
except ImportError:
    yaml = None

WIKI = Path(r"D:\Applications\app\kepano-obsidian-main\wiki")

VALID_TYPES = [
    "concept", "entity", "tool", "reference", "place",
    "institution", "method", "knowledge-card", "synthesis",
    "index", "log", "research-report", "lesson"
]


def validate_one(fp: Path) -> list[str]:
    """验证单个文件，返回错误列表"""
    errors = []
    content = fp.read_text(encoding="utf-8")
    m = re.match(r"^---\n(.*?)\n---", content, re.DOTALL)
    if not m:
        return ["No frontmatter found"]
    if not yaml:
        return ["PyYAML not installed"]
    try:
        front = yaml.safe_load(m.group(1)) or {}
    except yaml.YAMLError as e:
        return [f"YAML parse error: {e}"]

    # categories must contain [[LLM Wiki]]
    cats = front.get("categories", [])
    if isinstance(cats, str):
        cats = [cats]
    if not any(str(c).strip("[]") == "LLM Wiki" for c in cats):
        errors.append("Missing [[LLM Wiki]] in categories")

    # tags must contain wiki
    tags = front.get("tags", [])
    if isinstance(tags, str):
        tags = [tags]
    if "wiki" not in tags:
        errors.append("Missing 'wiki' in tags")

    # type must exist and be valid
    ptype = front.get("type")
    if not ptype:
        errors.append("Missing type")
    elif isinstance(ptype, str) and ptype not in VALID_TYPES:
        errors.append(f"Invalid type: '{ptype}' (valid: {', '.join(VALID_TYPES)})")
    elif isinstance(ptype, list):
        for t in ptype:
            if t not in VALID_TYPES:
                errors.append(f"Invalid type in list: '{t}'")

    # source must exist
    if not front.get("source"):
        errors.append("Missing source")

    # status=superseded must have superseded_by
    if front.get("status") == "superseded" and not front.get("superseded_by"):
        errors.append("status=superseded but missing superseded_by")

    # relations targets must exist
    for rel in front.get("relations", []):
        target = str(rel.get("target", "")).strip("[]")
        if target and not (WIKI / f"{target}.md").exists():
            errors.append(f"Relation target [[{target}]] not found")

    return errors


def get_git_changed_wiki_files() -> list[Path]:
    """获取 git 暂存区中被修改的 wiki .md 文件"""
    try:
        result = subprocess.run(
            ["git", "diff", "--cached", "--name-only", "--diff-filter=ACM"],
            capture_output=True, text=True, check=True, cwd=WIKI.parent
        )
    except (subprocess.CalledProcessError, FileNotFoundError):
        return []
    files = []
    for line in result.stdout.splitlines():
        line = line.strip()
        if line.startswith("wiki/") and line.endswith(".md"):
            fp = WIKI.parent / line
            if fp.exists():
                files.append(fp)
    return files


def main():
    parser = argparse.ArgumentParser(description="Validate wiki page frontmatter")
    parser.add_argument("--files", nargs="+", help="Specific files to check")
    parser.add_argument("--git-hook", action="store_true", help="Read from git diff")
    args = parser.parse_args()

    if args.git_hook:
        files = get_git_changed_wiki_files()
        if not files:
            sys.exit(0)
    elif args.files:
        files = [Path(f) if Path(f).is_absolute() else WIKI / f for f in args.files]
    else:
        files = sorted(WIKI.glob("*.md"))

    all_errors = {}
    for fp in files:
        if fp.name in ("index.md", "log.md"):
            continue
        errors = validate_one(fp)
        if errors:
            all_errors[fp.stem] = errors

    if all_errors:
        print(f"=== Frontmatter Validation Errors ({len(all_errors)} files) ===")
        for name, errs in sorted(all_errors.items()):
            print(f"\n  [[{name}]]:")
            for e in errs:
                print(f"    [ERR] {e}")
        sys.exit(1)
    else:
        print("[OK] All frontmatter valid")
        sys.exit(0)


if __name__ == "__main__":
    main()