trim-ghost.py

from pathlib import Path
import re
import html

posts = list(Path('_posts').glob('*.md'))

for post_p in posts:
    content = post_p.read_text(encoding='utf-8')
    content = content.replace('<pre><code>', '<pre><code class="language-text">')
    content = content.replace('&quot;', '"')
    
    res = re.findall(r'<pre><code class="language-(.*?)">(.*?)</code></pre>', content, re.DOTALL)

    for lang_type, text in res:
        target_text = html.unescape(text)
        target_text = f'\n\n<!--程序替换头 <pre><code>-->\n```{lang_type}\n{target_text}\n```\n<!--程序替换尾 <pre><code>-->\n'

        content = content.replace(f'<pre><code class="language-{lang_type}">{text}</code></pre>', target_text)


    res = re.findall(r'<blockquote>(.+?)</blockquote>', content, re.DOTALL)
    for text in res:
        target_text = html.unescape(text)
        target_text = f'\n\n<!--程序替换头 quote-->\n\n> {target_text}\n\n<!--程序替换尾 quote-->\n\n'

        content = content.replace(f'<blockquote>{text}</blockquote>', target_text)
    post_p.write_text(content, encoding='utf-8')