使用lxml.html.clean.Cleaner清洗html

文檔: https://lxml.de/lxmlhtml.html#cleaning-up-html 代碼示例

# -*- coding: utf-8 -*-

from lxml.html.clean import Cleaner

html = """
<p cms-style="font-L">　　
    <strong>鐵打的騰訊</strong>
    <a href="//n.sinaimg.cn/tech/crawl/115/w550h365/20200326/963a-irkazzv3237667.jpg" class="keyword f_st" target="_blank">
        <img src="//n.sinaimg.cn/tech/crawl/115/w550h365/20200326/963a-irkazzv3237667.jpg" alt="">
    </a>
</p>
"""

# 儲存新聞的時候，很多屬性不需要儲存，不然會占用硬碟資源，是以隻保留圖檔标簽的src屬性就行
safe_attrs = frozenset(['src'])

# a标簽也不要，隻保留裡邊的内容
remove_tags = frozenset([
    'a'
])

cleaner = Cleaner(safe_attrs=safe_attrs, remove_tags=remove_tags)
cleaned_html = cleaner.clean_html(html)
print(cleaned_html)

'''
<p>　　
    <strong>鐵打的騰訊</strong>
    
        <img src="//n.sinaimg.cn/tech/crawl/115/w550h365/20200326/963a-irkazzv3237667.jpg">
    
</p>
'''      

清洗之後内容簡潔多了