通过BeautifulSoup获取属性值

Question

I want to get all data-js attribute values from the content by BeautifulSoup.

Input:

<p data-js="1, 2, 3">some text..</p><p data-js="5">some 1 text</p><p data-js="4"> some 2 text. </p>

Output:

['1, 2, 3', '5', '4']

I've done it with lxml:

>>> content = """<p data-js="1, 2, 3">some text..</p><p data-js="5">some 1 text</p><p data-js="4"> some 2 text. </p>"""
>>> import lxml.html as PARSER
>>> root = PARSER.fromstring(content)
>>> root.xpath("//*/@data-js")
['1, 2, 3', '5', '4']

I want the above result via BeautifulSoup.

Answer 1

The idea would to find all elements having data-js attributes and collect them in a list:

from bs4 import BeautifulSoup


data = """
<p data-js="1, 2, 3">some text..</p><p data-js="5">some 1 text</p><p data-js="4"> some 2 text. </p>
"""

soup = BeautifulSoup(data)
print [elm['data-js'] for elm in soup.find_all(attrs={"data-js": True})]

Prints ['1, 2, 3', '5', '4'] .

Answer 2

May a faster method with map without list comprehension.

from bs4 import BeautifulSoup
d = "..."
# create a soup instance
soup = BeautifulSoup(d)
# find all p-elements containing an data-js attribute
p = soup.find_all('p', attrs={"data-js": True})
# unpack data-js attribute from p-elements and map to new list
print map(lambda x: x['data-js'], p)

http://www.crummy.com/software/BeautifulSoup/bs4/doc/#find-all

Answer 3

You can use find_all() for this, but you have to put the attribute name in a dictionary, because it can't be used as a keyword argument by itself.

html = BeautifulSoup(content)
data = html.find_all(attrs={'data-js': True})

See here for more explanation.