How to select div by text content using Beautiful Soup? - html

Question

Trying to scrape some HTML from something like this. Sometimes the data I need is in div[0], sometimes div[1], etc.
Imagine everyone takes 3-5 classes. One of them is always Biology. Their report card is always alphabetized. I want everybody's Biology grade.
I've already scraped all this HTML into a text, now how to fish out the Biology grades?
<div class = "student">
<div class = "score">Algebra C-</div>
<div class = "score">Biology A+</div>
<div class = "score">Chemistry B</div>
</div>
<div class = "student">
<div class = "score">Biology B</div>
<div class = "score">Chemistry A</div>
</div>
<div class = "student">
<div class = "score">Alchemy D</div>
<div class = "score">Algebra A</div>
<div class = "score">Biology B</div>
</div>
<div class = "student">
<div class = "score">Algebra A</div>
<div class = "score">Biology B</div>
<div class = "score">Chemistry C+</div>
</div>
<div class = "student">
<div class = "score">Alchemy D</div>
<div class = "score">Algebra A</div>
<div class = "score">Bangladeshi History C</div>
<div class = "score">Biology B</div>
</div>
I'm using beautiful soup, and I think I'm going to have to find divs where Text includes "Biology"?
This is only for a quick scrape and I'm open to hard-coding and fiddling in Excel or whatnot. Yes, it's a shoddy website! Yes, they do have an API, and I don't know a thing about WDSL.
Short version: http://www.legis.ga.gov/Legislation/en-US/Search.aspx ,to find the date of last action on every bill, FWIW. It's troublesome because if a bill has no sponsors in the second chamber, instead of a div containing nothing, they just don't have a div there at all. So sometimes the timeline is in div 3, sometimes 2, etc.

Answer 1

(1) To just get the biology grade only, it is almost one liner.
import bs4, re
soup = bs4.BeautifulSoup(html)
scores_string = soup.find_all(text=re.compile('Biology'))
scores = [score_string.split()[-1] for score_string in scores_string]
print scores_string
print scores
The output looks like this:
[u'Biology A+', u'Biology B', u'Biology B', u'Biology B', u'Biology B']
[u'A+', u'B', u'B', u'B', u'B']
(2) You locate the tags and maybe for further tasks, you need to find the parent:
import bs4, re
soup = bs4.BeautifulSoup(html)
scores = soup.find_all(text=re.compile('Biology'))
divs = [score.parent for score in scores]
print divs
Output looks like this:
[<div class="score">Biology A+</div>,
<div class="score">Biology B</div>,
<div class="score">Biology B</div>,
<div class="score">Biology B</div>,
<div class="score">Biology B</div>]
*In conclusion, you can use find_siblings/parent/...etc to move around the HTML tree.*
More information about how to navigate the tree.
And Good luck with your work.

Answer 2

Another way (using css selector) is:
divs = soup.select('div:-soup-contains("Biology")')
EDIT:
BeautifulSoup4 4.7.0+ (SoupSieve) is required

Answer 3

You can extract them searching for any <div> element that has score as class attribute value, and use a regular expression to extract its biology score:
from bs4 import BeautifulSoup
import sys
import re
soup = BeautifulSoup(open(sys.argv[1], 'r'), 'html')
for div in soup.find_all('div', attrs={'class': 'score'}):
t = re.search(r'Biology\s+(\S+)', div.string)
if t: print(t.group(1))
Run it like:
python3 script.py htmlfile
That yields:
A+
B
B
B
B