Complex Beautiful Soup query
Here is a snippet of an HTML file I\'m exploring with Beautiful Soup.
-
>>> BeautifulSoup.BeautifulSoup("""<html><td width="50%">
... <strong class="sans"><a href="http:/website">Site</a></strong> <br />
... </html>""" )
<html><td width="50%">
<strong class="sans"><a href="http:/website">Site</a></strong> <br />
</td></html>
>>> [ a for a in strong.findAll("a")
for strong in tr.findAll("strong", attrs = {"class": "sans"})
for tr in soup.findAll("td", width = "50%")]
[<a href="http:/website">Site</a>]
讨论(0)
-
from bs4 import BeautifulSoup
html_doc = """<td width="50%">
<strong class="sans"><a href="http:/website">Site</a></strong> <br />
"""
soup = BeautifulSoup(html_doc, 'html.parser')
soup.select('td[width="50%"] .sans [href]')
# Out[24]: [<a href="http:/website">Site</a>]
Documentation
讨论(0)
-
BeautifulSoup's search mechanisms accept a callable, which the docs appear to recommend for your case: "If you need to impose complex or interlocking restrictions on a tag's attributes, pass in a callable object for name,...". (ok... they're talking about attributes specifically, but the advice reflects an underlying spirit to the BeautifulSoup API).
If you want a one-liner:
soup.findAll(lambda tag: tag.name == 'a' and \
tag.findParent('strong', 'sans') and \
tag.findParent('strong', 'sans').findParent('td', attrs={'width':'50%'}))
I've used a lambda in this example, but in practice you may want to define a callable function if you have multiple chained requirements as this lambda has to make two findParent('strong', 'sans') calls to avoid raising an exception if an <a> tag has no strong parent. Using a proper function, you could make the test more efficient.
讨论(0)
- 热议问题
加载中...
