"""
文件連結驗證腳本
檢查所有 .md 文件中的內部連結是否正確
"""
import os
import re
from pathlib import Path
from typing import List, Tuple
def find_broken_links() -> List[Tuple[str, str, str]]:
"""
查找所有損壞的內部連結
Returns:
List[Tuple[str, str, str]]: (文件路徑, 損壞的連結, 行號)
"""
broken_links = []
project_root = Path(__file__).parent.parent.parent
# 查找所有 .md 文件
md_files = list(project_root.rglob("*.md"))
# 正則表達式匹配 Markdown 連結
link_pattern = re.compile(r'\[([^\]]+)\]\(([^)]+)\)')
print("🔍 開始檢查文件連結...")
print(f"找到 {len(md_files)} 個 Markdown 文件\n")
for md_file in md_files:
try:
with open(md_file, 'r', encoding='utf-8') as f:
content = f.read()
lines = content.split('\n')
# 逐行檢查連結
for line_num, line in enumerate(lines, 1):
matches = link_pattern.findall(line)
for link_text, link_url in matches:
# 跳過外部連結和錨點
if link_url.startswith(('http://', 'https://', '#', 'mailto:')):
continue
# 移除錨點部分
link_path = link_url.split('#')[0]
if not link_path:
continue
# 解析相對路徑
if link_path.startswith('/'):
# 絕對路徑
target_path = project_root / link_path[1:]
else:
# 相對路徑
target_path = (md_file.parent / link_path).resolve()
# 檢查文件是否存在
if not target_path.exists():
rel_path = md_file.relative_to(project_root)
broken_links.append((
str(rel_path),
link_url,
line_num
))
except Exception as e:
print(f"⚠️ 處理文件時出錯 {md_file}: {e}")
return broken_links
def print_results(broken_links: List[Tuple[str, str, str]]):
"""列印檢查結果"""
print("=" * 70)
if not broken_links:
print("✅ 所有連結檢查通過！沒有發現損壞的連結。")
print("\n📊 檢查統計:")
print(" • 所有內部連結都正確指向目標文件")
print(" • 文件路徑正確")
print(" • 組織結構完整")
else:
print(f"⚠️ 發現 {len(broken_links)} 個損壞的連結：\n")
# 按文件分組顯示
current_file = None
for file_path, link, line_num in sorted(broken_links):
if file_path != current_file:
if current_file is not None:
print()
print(f"📄 {file_path}:")
current_file = file_path
print(f" ❌ Line {line_num}: {link}")
print(f"\n總共發現 {len(broken_links)} 個損壞的連結")
print("=" * 70)
def check_specific_docs_links():
"""檢查 docs/ 資料夾特定的舊路徑"""
print("\n🔎 檢查 docs/ 舊路徑引用...")
project_root = Path(__file__).parent.parent.parent
docs_path = project_root / "docs"
# 需要檢查的舊路徑模式
old_patterns = [
r'docs/PROMPTS_USAGE_GUIDE\.md',
r'docs/PROMPTS_ENHANCEMENT_PLAN\.md',
r'docs/DEVELOPER_GUIDE\.md',
r'docs/INSTALLATION_GUIDE\.md',
r'docs/API\.md',
r'docs/EXAMPLES\.md',
r'docs/PHASE_\d+',
r'docs/PROMPTS_COMPLETE_REFACTORING',
r'docs/PROMPTS_IMPROVEMENT',
r'docs/PROMPTS_DIAGNOSTIC',
]
issues = []
for md_file in project_root.rglob("*.md"):
try:
with open(md_file, 'r', encoding='utf-8') as f:
content = f.read()
for pattern in old_patterns:
if re.search(pattern, content):
rel_path = md_file.relative_to(project_root)
issues.append((str(rel_path), pattern))
except Exception as e:
pass
if issues:
print("\n⚠️ 發現使用舊路徑的引用：\n")
for file_path, pattern in issues:
print(f" ❌ {file_path}: {pattern}")
else:
print(" ✅ 沒有發現舊路徑引用")
if __name__ == "__main__":
print("🔍 開始驗證文件連結...")
print("=" * 70)
# 檢查損壞的連結
broken_links = find_broken_links()
print_results(broken_links)
# 檢查舊路徑
check_specific_docs_links()
print("\n✨ 檢查完成！")
# 返回狀態碼
exit(0 if not broken_links else 1)
