在python中使用lxml解析标签内的所有文本

Question

我正在尝试解析如下所示的HTML文件

<ol>
  <li>
    <div class="c1">
      <span class="s1">hi</span>
      " hello "
      <span class="s2">world!</span>
    </div>
  </li>
  <li>
    <div class="c2">
      <span class="s3">abc</span>
      " def ghijkl "
      <span class="s1">mno</span>
      " pqr!"
    </div>
  </li>
</ol>

我尝试使用以下代码进行解析

tree = html.fromstring(code.content)
sol = tree.xpath('//ol//text()')
for x in sol:
    print x

我得到这样的结果

hi
 hello 
world!
abc
 def ghijkl
mno
 pqr!

我要怎么做才能在一行中的每个<li>标记中获取所有文本。 即我希望输出是

hi hello world!
abc def ghijkl mno pqr!

Answer 1

$ cat a.py
from lxml import etree

xml = """<ol>
  <li>
    <div class="c1">
      <span class="s1">hi</span>
      " hello "
      <span class="s2">world!</span>
    </div>
  </li>
  <li>
    <div class="c2">
      <span class="s3">abc</span>
      " def ghijkl "
      <span class="s1">mno</span>
      " pqr!"
    </div>
  </li>
</ol>"""

tree = etree.fromstring(xml)
sol = tree.xpath('//ol//li')
for a in sol:
   print " ".join([t.strip() for t in a.itertext()]).strip()

$ python a.py
hi " hello " world!
abc " def ghijkl " mno " pqr!"

Answer 2

您可以获取每个li并使用normalize-space ：

from lxml import html
h = """<ol>
  <li>
    <div class="c1">
      <span class="s1">hi</span>
      " hello "
      <span class="s2">world!</span>
    </div>
  </li>
  <li>
    <div class="c2">
      <span class="s3">abc</span>
      " def ghijkl "
      <span class="s1">mno</span>
      " pqr!"
    </div>
  </li>
</ol>"""


tree = html.fromstring(h)

for li in tree.xpath("//ol/li"):
    print(li.xpath("normalize-space(.)"))

这给你：

hi " hello " world!
abc " def ghijkl " mno " pqr!"