|
| 1 | +import os |
| 2 | +import re |
| 3 | +import urllib.request |
| 4 | +import urllib.error |
| 5 | +import ssl |
| 6 | +import concurrent.futures |
| 7 | + |
| 8 | +# Configuration |
| 9 | +ROOT_DIR = os.getcwd() |
| 10 | +MAX_THREADS = 20 |
| 11 | +SOFT_404_KEYWORDS = [ |
| 12 | + 'id="pageNotFound"', |
| 13 | + "Oops! That page can't be found.", |
| 14 | + "alt=\"Page Not Found\"", |
| 15 | + "illus-404.png", |
| 16 | +] |
| 17 | + |
| 18 | + |
| 19 | +# Context for https requests |
| 20 | +ctx = ssl.create_default_context() |
| 21 | +ctx.check_hostname = False |
| 22 | +ctx.verify_mode = ssl.CERT_NONE |
| 23 | + |
| 24 | +def find_md_files(root_dir): |
| 25 | + md_files = [] |
| 26 | + for root, dirs, files in os.walk(root_dir): |
| 27 | + if '.git' in dirs: dirs.remove('.git') |
| 28 | + if '_site' in dirs: dirs.remove('_site') |
| 29 | + |
| 30 | + for file in files: |
| 31 | + if file.endswith('.md'): |
| 32 | + md_files.append(os.path.join(root, file)) |
| 33 | + return md_files |
| 34 | + |
| 35 | +DS_SOFT_404_MARKERS = [ |
| 36 | + 'id="pagenotfound"', # DOM id（你 lower() 了，所以用小写） |
| 37 | + 'illus-404.png', # 404 图片 |
| 38 | + "oops! that page can't be found.", # 404 H1 文案 |
| 39 | + 'alt="page not found"', # 图片 alt |
| 40 | +] |
| 41 | + |
| 42 | +def is_soft_404(content_lower: str) -> str | None: |
| 43 | + # 返回命中的 marker，便于 debug |
| 44 | + for m in DS_SOFT_404_MARKERS: |
| 45 | + if m in content_lower: |
| 46 | + return m |
| 47 | + return None |
| 48 | + |
| 49 | +def check_url(url): |
| 50 | + try: |
| 51 | + req = urllib.request.Request( |
| 52 | + url, |
| 53 | + headers={'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'} |
| 54 | + ) |
| 55 | + with urllib.request.urlopen(req, timeout=15, context=ctx) as response: |
| 56 | + code = response.getcode() |
| 57 | + |
| 58 | + # 3xx 也算 OK（已允许重定向则一般拿到最终 200，这里保留逻辑） |
| 59 | + if 200 <= code < 400: |
| 60 | + try: |
| 61 | + # 建议读多一点，Dynamsoft 的 pageNotFound 区块可能不在前 15KB |
| 62 | + content = response.read(50000).decode('utf-8', errors='ignore').lower() |
| 63 | + |
| 64 | + # ✅ 先做“强特征” soft-404 判断（更准） |
| 65 | + hit = is_soft_404(content) |
| 66 | + if hit: |
| 67 | + return False, code, f"Soft 404 (marker: {hit})" |
| 68 | + |
| 69 | + # ✅ 再做你原来的 title/h1 关键词检测（兜底） |
| 70 | + if "<title>" in content: |
| 71 | + title_part = content.split("<title>", 1)[1].split("</title>", 1)[0] |
| 72 | + for keyword in SOFT_404_KEYWORDS: |
| 73 | + if keyword in title_part: |
| 74 | + return False, code, f"Soft 404 (Title: '{keyword}')" |
| 75 | + |
| 76 | + # h1 可能有属性，例如 <h1 class="...">，所以用更宽松的方式找 |
| 77 | + if "<h1" in content and "</h1>" in content: |
| 78 | + h1_block = content.split("<h1", 1)[1].split("</h1>", 1)[0] |
| 79 | + for keyword in SOFT_404_KEYWORDS: |
| 80 | + if keyword in h1_block: |
| 81 | + return False, code, f"Soft 404 (H1: '{keyword}')" |
| 82 | + |
| 83 | + except Exception: |
| 84 | + pass |
| 85 | + |
| 86 | + return True, code, "OK" |
| 87 | + else: |
| 88 | + return False, code, f"Status {code}" |
| 89 | + |
| 90 | + except urllib.error.HTTPError as e: |
| 91 | + return False, e.code, f"HTTP Error {e.code}" |
| 92 | + except urllib.error.URLError as e: |
| 93 | + return False, 0, f"URL Error: {e.reason}" |
| 94 | + except Exception as e: |
| 95 | + return False, 0, f"Error: {str(e)}" |
| 96 | + |
| 97 | +def scan_file_for_https_links(file_path): |
| 98 | + links = [] |
| 99 | + try: |
| 100 | + with open(file_path, 'r', encoding='utf-8') as f: |
| 101 | + content = f.read() |
| 102 | + |
| 103 | + md_links = re.findall(r'\[.*?\]\((https://.*?)\)', content) |
| 104 | + html_srcs = re.findall(r'src=["\'](https://.*?)["\']', content) |
| 105 | + |
| 106 | + for link in md_links + html_srcs: |
| 107 | + clean_link = link.split()[0].strip('()') |
| 108 | + links.append(clean_link) |
| 109 | + except Exception as e: |
| 110 | + print(f"Error reading {file_path}: {e}") |
| 111 | + return links |
| 112 | + |
| 113 | +def main(): |
| 114 | + print(f"Scanning for HTTPS links in: {ROOT_DIR}") |
| 115 | + files = find_md_files(ROOT_DIR) |
| 116 | + |
| 117 | + file_links_map = {} |
| 118 | + all_urls = set() |
| 119 | + |
| 120 | + for file_path in files: |
| 121 | + links = scan_file_for_https_links(file_path) |
| 122 | + if links: |
| 123 | + file_links_map[file_path] = links |
| 124 | + all_urls.update(links) |
| 125 | + |
| 126 | + print(f"Found {len(all_urls)} unique HTTPS links to check.") |
| 127 | + print(f"Checking links (Concurrent Status + Soft 404 Check) ...") |
| 128 | + |
| 129 | + url_status = {} |
| 130 | + |
| 131 | + with concurrent.futures.ThreadPoolExecutor(max_workers=MAX_THREADS) as executor: |
| 132 | + future_to_url = {executor.submit(check_url, url): url for url in all_urls} |
| 133 | + completed = 0 |
| 134 | + total = len(all_urls) |
| 135 | + |
| 136 | + for future in concurrent.futures.as_completed(future_to_url): |
| 137 | + url = future_to_url[future] |
| 138 | + try: |
| 139 | + result = future.result() |
| 140 | + url_status[url] = result |
| 141 | + except Exception as e: |
| 142 | + url_status[url] = (False, 0, str(e)) |
| 143 | + |
| 144 | + completed += 1 |
| 145 | + if completed % 10 == 0 or completed == total: |
| 146 | + print(f"Progress: {completed}/{total}", end='\r') |
| 147 | + |
| 148 | + print(f"\nProgress: {total}/{total}") |
| 149 | + print("\n--- Broken or Soft 404 Links Report ---\n") |
| 150 | + |
| 151 | + broken_count = 0 |
| 152 | + files_with_issues = 0 |
| 153 | + |
| 154 | + for file_path, links in file_links_map.items(): |
| 155 | + issues_in_file = [] |
| 156 | + for link in links: |
| 157 | + if link in url_status: |
| 158 | + is_valid, code, msg = url_status[link] |
| 159 | + if not is_valid: |
| 160 | + issues_in_file.append((link, code, msg)) |
| 161 | + |
| 162 | + if issues_in_file: |
| 163 | + files_with_issues += 1 |
| 164 | + rel_path = os.path.relpath(file_path, ROOT_DIR) |
| 165 | + print(f"📄 {rel_path}") |
| 166 | + for link, code, msg in issues_in_file: |
| 167 | + print(f" ❌ [{code}] {link} -> {msg}") |
| 168 | + broken_count += 1 |
| 169 | + print("") |
| 170 | + |
| 171 | + if broken_count == 0: |
| 172 | + print("✅ No broken links found!") |
| 173 | + else: |
| 174 | + print(f"❌ Found {broken_count} issues in {files_with_issues} files.") |
| 175 | + |
| 176 | +if __name__ == "__main__": |
| 177 | + main() |
0 commit comments