數據庫存儲的富文本包含絕對圖片 URL (https://192.168.1.1:888/m/getimg?filename=xxx.jpg),需批量替換為相對路徑 (/m/getimg?filename=xxx.jpg) 以適配移動端。
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import pymongo, re
from StringIO import StringIO
conn = pymongo.Connection()
db = conn.test
coll = db.test
# 匹配 src 或 href 屬性中的絕對 URL,捕獲 /m/getimg 及後續參數
regex = re.compile(r'(href|src)=["\']https://.*?(/m/getimg\?.*?)["\']')
def replace_url():
for doc in coll.find():
content = doc.get('content')
if not content: continue
buf_in = StringIO(content)
buf_out = StringIO()
for line in buf_in:
# 替換為相對路徑,保持屬性名
line = regex.sub(r'\1="\2"', line)
buf_out.write(line)
new_content = buf_out.getvalue()
if new_content != content:
coll.update({'_id': doc['_id']}, {'$set': {'content': new_content}})
if __name__ == '__main__':
replace_url()
(href|src):捕獲屬性名(分組 1)。=["\']:匹配等號與引號。https://.*?:非貪婪匹配協議與域名。(/m/getimg\?.*?):捕獲目標相對路徑(分組 2)。["\']:結束引號。r'\1="\2"':保持屬性名,雙引號包裹相對路徑。BeautifulSoup 解析 DOM 更可靠,避免正則誤匹配。bulk_write 提高性能。