正则表达式用于获取组内的匹配项

Question

我不知道下面的情况是否可行。假设我有以下文本：

<ul class="yes">
    <li><img src="whatever1"></li>
    <li><img src="whatever2"></li>
    <li><img src="whatever3"></li>
    <li><img src="whatever4"></li>
</ul>
<ul class="no">
    <li><img src="whatever5"></li>
    <li><img src="whatever6"></li>
    <li><img src="whatever7"></li>
    <li><img src="whatever8"></li>
</ul>

我希望将ul中的每个img的src都与类yes匹配。我希望一个正则表达式返回给我：

whatever1
whatever2
whatever3
whatever4

如何在一个正则表达式中加入两个这样的正则表达式？

<ul class="yes">(.+?)<\/ul>
<img src="(whatever.+?)">

Answer 1

众所周知，Regex很难用于解析类似XML的东西。最好跳过这个想法，转而使用适当的超文本标记语言解析器，例如使用BeautifulSoup4

import bs4

html = """
<ul class="yes">
    <li><img src="whatever1"></li>
    <li><img src="whatever2"></li>
    <li><img src="whatever3"></li>
    <li><img src="whatever4"></li>
</ul>
<ul class="no">
    <li><img src="whatever5"></li>
    <li><img src="whatever6"></li>
    <li><img src="whatever7"></li>
    <li><img src="whatever8"></li>
</ul>
"""

soup = bs4.BeautifulSoup(html)

def match_imgs(tag):
    return tag.name == 'img' \
        and tag.parent.parent.name == 'ul' \
        and tag.parent.parent['class'] == ['yes']

imgs = soup.find_all(match_imgs)
print(imgs)

whatevers = [i['src'] for i in imgs]
print(whatevers)

收益率：

[<img src="whatever1"/>, <img src="whatever2"/>, <img src="whatever3"/>,
<img src="whatever4"/>]

[u'whatever1', u'whatever2', u'whatever3', u'whatever4']