Can I do a findall regular expression like this?

Question

So I need to grab the numbers after lines looking like this

<div class="gridbarvalue color_blue">79</div>

and

<div class="gridbarvalue color_red">79</div>

Is there a way I can do a findAll('div', text=re.recompile('<>)) where I would find tags with gridbarvalue color_<red or blue>?

I'm using beautifulsoup.

Also sorry if I'm not making my question clear, I'm pretty inexperienced with this.

Yeah, like there are multiple lines like the ones I said and I want to grab the numbers. — user3716335, Oct 22 '14 at 03:43
[Check out this part of the documentation](http://www.crummy.com/software/BeautifulSoup/bs3/documentation.html#The%20basic%20find%20method:%20findAll%28name,%20attrs,%20recursive,%20text,%20limit,%20**kwargs%29) — nhahtdh, Oct 22 '14 at 03:56
You should be able to use BeautifulSoup to do this, without regex. See the thousands of questions already posted for reasons why parsing HTML with regex is bad. — John Zwinck, Oct 22 '14 at 04:00
OOH! OOH! I get to talk about [the ponies](http://stackoverflow.com/a/1732454/57191) — cwallenpoole, Oct 22 '14 at 04:11
@cwallenpoole & JohnZwinck - I don't think this is about parsing html with regex. Rather, BeautifulSoup can use regex for matching tag/attribute names. — Peter Gibson, Oct 22 '14 at 04:15

Peter Gibson · Answer 1 · 2014-10-22T04:11:20.193

class is a Python keyword, so BeautifulSoup expects you to put an underscore after it when using it as a keyword parameter

>>> soup.find_all('div', class_=re.compile(r'color_(?:red|blue)'))
[<div class="gridbarvalue color_blue">79</div>, <div class="gridbarvalue color_red">79</div>]

To also match the text, use

>>> soup.find_all('div', class_=re.compile(r'color_(?:red|blue)'), text='79')
[<div class="gridbarvalue color_blue">79</div>, <div class="gridbarvalue color_red">79</div>]

score 0 · Accepted Answer · answered Oct 22 '14 at 04:04

0

import re
elems = soup.findAll(attrs={'class' : re.compile("color_(blue|red)")})
for each e in elems:
    m = re.search(">(\d+)<", str(e))
    print "The number is %s" % m.group(1)

answered Oct 22 '14 at 04:04

Andrew Luo

919
1
5
6

I think you should use `e.strip()` instead of using a regex to extract. – nhahtdh Oct 22 '14 at 04:12

Can I do a findall regular expression like this?

2 Answers2