chore: revert README to previous version

[tangwz]

Summary

• roll back README.md to its prior content

Testing

• python scripts/update_readme.py (fails: RuntimeError: failed to fetch https://mysql.taobao.org/monthly/: <urlopen error [Errno 101] Network is unreachable>)

---

https://chatgpt.com/codex/tasks/task_e_689ed80fd5808330b5cd142dc02a4729

[Copilot]

Pull Request Overview

This pull request adds a new Python script that fetches latest posts from MySQL Taobao monthly archives and updates the README.md file with new database-related articles. The script includes functionality to parse HTML content, categorize articles, and append new entries to the README in markdown table format.

• Adds a complete web scraping script for MySQL monthly archives
• Implements HTML parsing to extract articles by database categories
• Provides README.md update functionality with new article entries

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[Copilot]

The hardcoded timeout value of 10 seconds should be configurable. Consider adding a TIMEOUT environment variable or making it a module-level constant to allow customization for different network conditions.

Suggested change

	with opener.open(req, timeout=10) as resp: # pragma: no cover - network I/O
	with opener.open(req, timeout=TIMEOUT) as resp: # pragma: no cover - network I/O

[Copilot]

Using string replace for content insertion is fragile and could cause incorrect replacements if the table header appears multiple times in the README. Consider using regex with word boundaries or a more precise matching approach to ensure only the intended table is updated.

Suggested change

	content = content.replace(table_header, table_header + "\n" + "\n".join(lines))
	# Use regex to match the table header and following table rows
	pattern = re.compile(
	re.escape(table_header) + r"(?:\n(?:\|.*\|))*",
	re.MULTILINE
	)
	replacement = table_header + "\n" + "\n".join(lines)
	content, count = pattern.subn(replacement, content, count=1)
	if count == 0:
	# fallback: skip if not matched (should not happen due to earlier check)
	continue

[Copilot]

Using string containment check on the entire README content for each URL is inefficient. Consider extracting existing URLs once at the beginning using regex and storing them in a set for O(1) lookup instead of O(n) string searches.

Suggested change

	if url in existing:
	# Extract all URLs from the README content and store in a set for O(1) lookup
	existing_urls = set(re.findall(r"https?://[^\s\)]+", existing))
	updates = defaultdict(list)
	for y, m in months:
	url = f"{BASE_URL}/{y:04d}/{m:02d}/"
	if url in existing_urls: