<p>这是一个很好的起点:</p>
<pre><code>In [18]: for a in response.css('.extraUserInfo'):
...: print(a.css('*::text').extract())
...: print('\n\n\n')
...:
['\n', '\n', '\n', '\n'] # < this (and other outputs like this) is because there is an extra `extraUserInfo` class block above the desired info block if the user has a user group picture/avatar below their username
['\n', '\n', 'Joined:', '\n', 'Mar 24, 2013', '\n', '\n', '\n', 'Messages:', '\n', '6,747', '\n', '\n']
['\n', '\n', '\n', '\n']
['\n', '\n', 'Joined:', '\n', 'Mar 24, 2013', '\n', '\n', '\n', 'Messages:', '\n', '6,747', '\n', '\n']
['\n', '\n', 'Joined:', '\n', 'Apr 14, 2013', '\n', '\n', '\n', 'Messages:', '\n', '580', '\n', '\n', '\n', 'Location:', '\n', '\n', 'Fort Pierce, FL', '\n', '\n', '\n']
['\n', '\n', 'Joined:', '\n', 'Oct 20, 2012', '\n', '\n', '\n', 'Messages:', '\n', '2,476', '\n', '\n', '\n', 'Location:', '\n', '\n', 'Philadelphia, PA', '\n', '\n', '\n']
['\n', '\n', 'Joined:', '\n', 'Dec 11, 2012', '\n', '\n', '\n', 'Messages:', '\n', '2,938', '\n', '\n', '\n', 'Location:', '\n', '\n', 'Colorado', '\n', '\n', '\n']
['\n', '\n', 'Joined:', '\n', 'Sep 30, 2016', '\n', '\n', '\n', 'Messages:', '\n', '833', '\n', '\n', '\n', 'Location:', '\n', '\n', 'Indiana', '\n', '\n', '\n']
...
</code></pre>
<p>有很多方法可以做到这一点。稍微摆弄一下就会得到你喜欢的格式。上面的方法只是一个很好的起点,因为有很多行只有换行符列表作为输出,这是因为(看起来)用户信息块在用户有用户组图像的地方(比如亚利桑那州的特斯拉),然后<code>extraUserInfo</code>类也被用来对html块进行分组。会有更好的方法来分组</p>
<p>基本上,response.css('.extraUserInfo')将使用类<code>extraUserInfo</code>聚合所有块,这些块似乎是保存您要查找的用户信息的块。
从那里用<code>::text</code>伪选择器提取所有底层文本并解析数组</p>
<p>如果您仔细查看html结构,这样提取它的方式会减少以后的处理工作,那么肯定有更好的方法来实现这一点,但这应该会让您走上正确的轨道。CSS选择器或xpath文档应该很有帮助</p>