CODEWITHSUNDEEP
pythonpython-3.xweb-scrapingbeautifulsoup
ByRequest
ByRequest

Reputation: 311

Using Python and BeautifulSoup to scrape list with variable orders and tags based on text strings

Details: MacOS, Python3, BeautifulSoup4

I am new to Python and even newer to BeautifulSoup so please excuse any beginner mistakes here. I am attempting to scrape html pages which do not heavily differentiate their tags by classes or div ids. In other words, I am trying to scrape the middle section of a list. The list will have an unpredictable amount of tags and elements (sometimes they use an unordered list, other times they are using a description list) so what I am scraping is fairly unpredictable, however, I do have two known variables and those would be the header string text I want to START at and the header string text I want to END at.

I have assembled the following example html to test this on:

<div class="panel panel-default">
  <div class="panel-heading">
    <h3 class="panel-title">First Section Title - Known Variable or String</h3>
  </div>
</div>
<div>
  <ul class="unstyled">
    <li>Item1</li>
    <li>Item2</li>
    <li>Empty LI Tags Also Exist</li>
  </ul>
  <dl class="dl-horizontal">
    <dt>Title of some description list</dt>
    <dd>Another item may exist here</dd>
  </dl>
</div>
<div>
  <div class="panel panel-default">
    <div class="panel-heading">
      <h3 class="panel-title">Another Section Title</h3>
    </div>
  </div>
  <ul class="unstyled">
    <li>Item1</li>
    <li></li>
  </ul>
  <dl class="dl-horizontal">
    <dt>Another Description List Title</dt>
    <dd>Another item may exist here</dd>
    <dt>And here</dt>
    <dd>And Here</dd>
  </dl>
</div>
<div>
  <div class="panel panel-default">
    <div class="panel-heading">
      <h3 class="panel-title">Section Title (String) I Wish To Stop At - Known Variable or String</h3>
    </div>
  </div>
</div>

Again, using the above model, I want to start at the first section I listed and end at the known text string of a particular section towards the bottom.

I have listed my Python script below. So far, the following Python is grabbing the correct information, however, I do not believe it will work under all circumstances, and there is probably a more efficient way to go about this. Here are some of the issues I believe are in my script:

  1. My script is rather static - while it appears to start at the correct header, I have pieced out two sections separately as I do not believe my For loop is working the way it should be (I do not think ##Section 2 should be needed if written correctly).

  2. Because my For loop is likely not doing what I probably think it is (I'd like it to iterate through the sections) I never had to define the stopping point (the string of text at the section I wish to stop at).

  3. Since I am not convinced the loop is working correctly, I do not believe this will handle any curveballs I am thrown by the site - for example variable numbers of items on the list and if they add an additional section I would want between the "Beginning section" and "Ending section" defined.

I believe what needs to happen is:

Python:

##Scrape
#import beautifulsoup and requests library
from bs4 import BeautifulSoup
import requests

soup = BeautifulSoup(open("mock.html"), "html.parser")#BeautifulSoup(page.read())
#Begin by grabbing the section
stuff = soup.find_all(class_="panel-heading")
#Search for the first section title text string
next_elem = soup.find(text="First Section Title - Known Variable or String").findNext('li').contents[0]
#Attempt to scan the remainder of the section, starting with the next line item
next_next = next_elem.parent.find_next_sibling()
for item in next_next.findAll('li','dt','dd'):
    if isinstance(item, Tag):
        print(item.text)
print(next_elem)
print(next_next.text)

##Section 2 - I'd like to cut this out
s2_elem = soup.find(text="Another Section Title").findNext('li').contents[0]
s2_nxnx = s2_elem.parent.find_next_sibling()
s2_nxnxnx = s2_nxnx.parent.find_next_sibling()
print(s2_elem)
print(s2_nxnx.text)
print(s2_nxnxnx.text)

Upvotes: 1

Views: 879

Answers (1)

Martin Evans
Martin Evans

Reputation: 46779

You could use a variable to spot when you are between search_start and search_end:

from bs4 import BeautifulSoup, Tag
import requests

search_start = "First Section Title - Known Variable or String"
search_end = "Section Title (String) I Wish To Stop At - Known Variable or String"

soup = BeautifulSoup(open("mock.html"), "html.parser")
start = False

for el in soup.find_all(['li', 'dt', 'dd', 'h3']):
    if el.name == 'h3':
        if el.text == search_start:
            start = True
        elif el.text == search_end:
            break
    elif start and isinstance(el, Tag):
        print(el.text)

This would give you the following output:

Item1
Item2
Empty LI Tags Also Exist
Title of some description list
Another item may exist here
Item1

Another Description List Title
Another item may exist here
And here
And Here

Upvotes: 1

Related Questions