#!/usr/bin/env python3
"""
站点 YAML 配置结构校验（2026-11-02 教训驱动）

为什么需要：Python 默认 yaml.safe_load 对重复键是"后者覆盖前者"的静默行为，
收编时用它验收会漏。本包实测 public/admin/config.yml 里 typography 被插进双语
字段的 name: 与 widget: object 之间 → 同一 mapping 出现重复键 widget/fields，
后台双语编辑器退化成字符串、排版选择器字段被劫持。本脚本用自定义构造器在
解析阶段对重复键直接报错，并附带后台 hint 覆盖率体检。

用法：python3 scripts/build-perf/config-check.py [repo根=.]
退出码：0 通过；1 有失败项
"""
import sys
from pathlib import Path

try:
    import yaml
except ImportError:
    print('[config-check] 缺少 pyyaml：pip install pyyaml')
    sys.exit(1)

FILES = [
    'public/admin/config.yml',
    'content/site/home.yaml',
    'content/site/about.yaml',
    'content/site/opensource.yaml',
    'content/site/theme.yaml',
]


class StrictLoader(yaml.SafeLoader):
    """对重复键报错的 SafeLoader（默认行为是静默覆盖）。"""


def _construct_mapping(loader, node, deep=False):
    seen = set()
    for key_node, _value_node in node.value:
        key = loader.construct_object(key_node, deep=deep)
        if key in seen:
            raise yaml.YAMLError(
                f'duplicate key {key!r} at line {key_node.start_mark.line + 1}'
            )
        seen.add(key)
    return yaml.SafeLoader.construct_mapping(loader, node, deep=deep)


StrictLoader.add_constructor(
    yaml.resolver.BaseResolver.DEFAULT_MAPPING_TAG, _construct_mapping
)


def main() -> int:
    root = Path(sys.argv[1] if len(sys.argv) > 1 else '.')
    failed = 0
    loaded = {}

    for rel in FILES:
        path = root / rel
        if not path.exists():
            print(f'[config-check] SKIP {rel}（不存在）')
            continue
        try:
            loaded[rel] = yaml.load(path.read_text(encoding='utf-8'), Loader=StrictLoader)
            print(f'[config-check] OK   {rel}')
        except yaml.YAMLError as exc:
            failed += 1
            first = str(exc).splitlines()[0]
            print(f'[config-check] FAIL {rel}\n  {first}')

    admin = loaded.get('public/admin/config.yml')
    if admin:
        # 结构断言（2026-11-02 教训）：每个集合必须二选一——folder 集合或
        # files 文件集合；files 项必须齐 file/label/fields，且 file 指向真实文件。
        # 把 file 直接写在集合层级会被 Sveltia 判非法（后台整体报错）。
        for coll in admin.get('collections', []):
            name = coll.get('name', '<unnamed>')
            has_folder, has_files = 'folder' in coll, 'files' in coll
            if not has_folder and not has_files:
                failed += 1
                print(f'[config-check] FAIL 集合 {name}: 既无 folder 也无 files（Sveltia 会拒绝加载）')
            if has_files:
                for entry in coll.get('files') or []:
                    miss = [k for k in ('file', 'label', 'fields') if k not in entry]
                    if miss:
                        failed += 1
                        print(f'[config-check] FAIL 集合 {name} 的文件项缺 {miss}')
                        continue
                    if not (root / entry['file']).exists():
                        failed += 1
                        print(f"[config-check] FAIL 集合 {name}: 文件不存在 {entry['file']}")

    if admin:
        fields = hints = 0

        def walk(node):
            nonlocal fields, hints
            if isinstance(node, dict):
                if 'name' in node:
                    fields += 1
                if 'hint' in node:
                    hints += 1
                for value in node.values():
                    walk(value)
            elif isinstance(node, list):
                for value in node:
                    walk(value)

        walk(admin)
        print(f'[config-check] 后台字段 {fields} 个，其中带 hint {hints} 个')
        if hints < 20:
            failed += 1
            print('[config-check] FAIL hint 覆盖不足（<20），后台字段缺少中文说明')

    if failed:
        print(f'[config-check] {failed} 项未通过')
        return 1
    print('[config-check] 全部通过')
    return 0


if __name__ == '__main__':
    raise SystemExit(main())