From 94a74e1fd537f9d5618ec83a29dd09700e6c2334 Mon Sep 17 00:00:00 2001
From: tangwz <tangwz.com@gmail.com>
Date: Fri, 15 Aug 2025 16:03:36 +0800
Subject: [PATCH] chore: revert README to previous version

---
 scripts/update_readme.py | 139 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 139 insertions(+)
 create mode 100755 scripts/update_readme.py

diff --git a/scripts/update_readme.py b/scripts/update_readme.py
new file mode 100755
index 0000000..7e58fdd
--- /dev/null
+++ b/scripts/update_readme.py
@@ -0,0 +1,139 @@
+#!/usr/bin/env python3
+"""
+Fetch latest posts from https://mysql.taobao.org/monthly/ (or a mirror) and
+update README.md with new entries grouped by database categories. The script
+parses month pages and appends unseen articles to the corresponding sections in
+README.md. If the default site is unreachable, set the ``MONTHLY_BASE_URL``
+environment variable to point to an accessible mirror.
+"""
+import os
+import re
+import sys
+from collections import defaultdict
+from pathlib import Path
+from html.parser import HTMLParser
+import urllib.request
+
+BASE_URL = os.environ.get("MONTHLY_BASE_URL", "https://mysql.taobao.org/monthly")
+
+
+def fetch(url: str) -> str:
+    """Fetch URL and return decoded text."""
+    req = urllib.request.Request(url, headers={"User-Agent": "Mozilla/5.0"})
+    proxy = os.environ.get("MONTHLY_PROXY")
+    if proxy:
+        handler = urllib.request.ProxyHandler({"http": proxy, "https": proxy})
+    else:
+        handler = urllib.request.ProxyHandler({})  # disable env proxies
+    opener = urllib.request.build_opener(handler)
+    try:
+        with opener.open(req, timeout=10) as resp:  # pragma: no cover - network I/O
+            return resp.read().decode("utf-8", errors="ignore")
+    except Exception as exc:  # pragma: no cover - network I/O
+        raise RuntimeError(f"failed to fetch {url}: {exc}")
+
+
+def get_months(html: str):
+    """Extract all (year, month) tuples from index html."""
+    pattern = re.compile(r"/monthly/(\d{4})/(\d{2})/")
+    months = {(int(y), int(m)) for y, m in pattern.findall(html)}
+    return sorted(months)
+
+
+class MonthParser(HTMLParser):
+    """Parse monthly page into articles grouped by category."""
+
+    def __init__(self):
+        super().__init__()
+        self.category = None
+        self.in_h2 = False
+        self.in_li = False
+        self.link = None
+        self.text_parts = []
+        self.articles = defaultdict(list)
+
+    def handle_starttag(self, tag, attrs):
+        if tag == "h2":
+            self.in_h2 = True
+        elif tag == "li":
+            self.in_li = True
+            self.link = None
+            self.text_parts = []
+        elif tag == "a" and self.in_li:
+            attrs = dict(attrs)
+            self.link = attrs.get("href")
+
+    def handle_endtag(self, tag):
+        if tag == "h2":
+            self.in_h2 = False
+        elif tag == "li":
+            if self.category and self.link and self.text_parts:
+                text = "".join(self.text_parts).strip()
+                m = re.match(r"\[(.*?)\]\s*(.*)", text)
+                if m:
+                    typ, title = m.groups()
+                else:
+                    parts = text.split(None, 1)
+                    typ = parts[0] if len(parts) == 2 else ""
+                    title = parts[-1]
+                self.articles[self.category].append((typ, title, self.link))
+            self.in_li = False
+            self.link = None
+            self.text_parts = []
+
+    def handle_data(self, data):
+        if self.in_h2:
+            self.category = data.strip()
+        elif self.in_li:
+            self.text_parts.append(data)
+
+
+def parse_month(year: int, month: int):
+    html = fetch(f"{BASE_URL}/{year:04d}/{month:02d}/")
+    parser = MonthParser()
+    parser.feed(html)
+    return parser.articles
+
+
+def update_readme(new_articles):
+    readme_path = Path(__file__).resolve().parent.parent / "README.md"
+    content = readme_path.read_text(encoding="utf-8")
+
+    for category, items in new_articles.items():
+        if not items:
+            continue
+        table_header = f"# {category}\n| 分类 | 标题  |\n|---|---|"
+        if table_header not in content:
+            # skip categories not present
+            continue
+        lines = [f"| {typ} | [{title}]({link}) |" for typ, title, link in items]
+        content = content.replace(table_header, table_header + "\n" + "\n".join(lines))
+
+    readme_path.write_text(content, encoding="utf-8")
+
+
+def main():  # pragma: no cover - integration logic
+    index_html = fetch(BASE_URL + "/")
+    months = get_months(index_html)
+    readme = Path(__file__).resolve().parent.parent / "README.md"
+    existing = readme.read_text(encoding="utf-8")
+
+    updates = defaultdict(list)
+    for y, m in months:
+        url = f"{BASE_URL}/{y:04d}/{m:02d}/"
+        if url in existing:
+            continue
+        try:
+            articles = parse_month(y, m)
+        except Exception as exc:
+            print(f"Failed to fetch {y}-{m:02d}: {exc}", file=sys.stderr)
+            continue
+        for cat, items in articles.items():
+            updates[cat].extend(items)
+
+    if updates:
+        update_readme(updates)
+
+
+if __name__ == "__main__":
+    main()