[source code] Python Programming Tutorial - 25 - How to Make a Web Crawler

+26 Bucky Roberts · September 3, 2014

import requests
from bs4 import BeautifulSoup


def trade_spider(max_pages):
    page = 1
    while page <= max_pages:
        url = "https://buckysroom.org/trade/search.php?page=" + str(page)
        source_code = requests.get(url)
        # just get the code, no headers or anything
        plain_text = source_code.text
        # BeautifulSoup objects can be sorted through easy
        soup = BeautifulSoup(plain_text)
        for link in soup.findAll('a', {'class': 'item-name'}):
            href = "https://buckysroom.org" + link.get('href')
            title = link.string  # just the text, not the HTML
            print(href)
            print(title)
            # get_single_item_data(href)
        page += 1


def get_single_item_data(item_url):
    source_code = requests.get(item_url)
    plain_text = source_code.text
    soup = BeautifulSoup(plain_text)
    # if you want to gather information from that page
    for item_name in soup.findAll('div', {'class': 'i-name'}):
        print(item_name.string)
    # if you want to gather links for a web crawler
    for link in soup.findAll('a'):
        href = "https://buckysroom.org" + link.get('href')
        print(href)


trade_spider(1)

Post a Reply

Replies

- page 2
Oldest  Newest  Rating
0 Gunseli Ozkan · January 28, 2016
Hello Bucky, I do whatever you say but my program prints out nothing. BTW, I guess you closed the webpage for trading and I couldn't find something similar to it. Can you guys help me out?
0 Tanner Hoke · January 29, 2015
import requests


from bs4 import BeautifulSoup

source_code = requests.get('https://www.thenewboston.com/forum/topic.php?id=1610')
plain_text = source_code.text
soup = BeautifulSoup(plain_text)
print(str(soup.find('code')).replace('<code>', '').replace('<br>', '\n').replace('?', '\t').replace('</code>', '').replace('</br>', ''))
0 Steve Gregg · January 5, 2016
I am doing this Python tutorial and decided to do my own web crawler. It gets all the stories from slashdot.org and saves them to an html file so you can read it offline. Any feedback or comments are appreciated


import requests
from bs4 import BeautifulSoup

def story_extractor(href):
source = requests.get(href) #get the page containing the story
plain_text = source.text #save the plain text version of the source code in a variable
soup = BeautifulSoup(plain_text,'lxml') #convert the source to a Beautiful Soup object
paragraph = soup.findAll('div',{'class':'p'}) #find the <div> tag where the class is 'p'
story_text = str(paragraph) #convert the story text found to a string
length = len(story_text) #find the length of the story in characters to use in the next line
return story_text[1:length-1] #return the story minus the first and last characters which were square brackets "[ ]"


def slashdot_crawler(pagesToCrawl): #begin definition of slashdot_crawler function
sd = open('slashdot.html', 'w') #create html file to write stories to
sd.write('<html><head><title>Slashdot stories</title></head><body><h1>The latest Slashdot stories</h1><br>') #write the initial HTML tags to the file to start the page
page = 0 #initialize page to 0 since slashdot starts counting their pages at 0
while page < pagesToCrawl: #keep looping until we get to the defined number of pages to crawl
print("Processing page " + str(page+1) + " of " + str(pagesToCrawl) + "\n") #prompt user on progress
url = "http://slashdot.org/?page=" + str(page) #assign the slashdot url to a variable
source = requests.get(url) #get the source code of the page
plain_text = source.text #save the plain text version of the source code in a variable
soup = BeautifulSoup(plain_text,'lxml') #convert the source to a Beautiful Soup object
for link in soup.findAll('span',{'class':'story-title'}): #loop through all the <span> tags with their class = story-title
href = link.a.get("href") #get the link to the story
title = link.a.string #store the story title
#print(title)
#print('http:' + href)
#print('\n')
sd.write('<p><a href="' + href + '">' + title + '</a><br>') #for each story start a paragraph and create a link to the original story using the story title
story_link = "http:" + href #add "http:" to the beginning of the url since slashdot links start with //
story_contents = story_extractor(story_link) #pass the link to the story extractor function which returns the actual text content of the story
sd.write(story_contents) #write the story content to the html file
sd.write("</p><br>") #close the html paragraph and insert a return
page += 1 #increment page
sd.write('</body></html>') #once the loop exits, no more stories so we can close the html page
sd.close() #close the html file we have been writing to

slashdot_crawler(3) #call the function to crawl slashdot.org
print("Processing complete") #prompt user to show that the program is done

0 Daniel Stoica · February 1, 2015
It's kind of hard to use this tutorial, considering that there's nothing on your trade page. :(
0 Ola Berglund · January 4, 2015
I tried to solve the problem Bucky gave us, but i couldn't figured it out so I went to watch these spoilers. Now I'm completely lost. What is this "('<br>',)"? I really need to understand since a webcrawler is basic stuff!
0 Probuddha N · April 17, 2015
I used the same coding (changing the page attributes) to get data from Truelocal.com.au


import requests
from bs4 import BeautifulSoup

def tru_crawler(max_pages):
page = 1
while page <= max_pages:
url = 'http://www.truelocal.com.au/find/car-rental/' + str(page)
code = requests.get(url)
text = code.text
soup = BeautifulSoup(text)
for link in soup.findAll('a', {'class':'name'}):
href = 'http://www.truelocal.com.au' + link.get('href')
title = link.string
# print (href)
# print (title)
get_single_item_data(href)
page += 1

def get_single_item_data(item_url):
code = requests.get(item_url)
text = code.text
soup = BeautifulSoup(text)
for item_name in soup.findAll('h1',{'itemprop':'name'}):
print item_name.string
for link in soup.findAll('a'):
href = 'http://www.truelocal.com.au' + link.get('href')
print (href)


tru_crawler(2)


and after getting the first page title and all the links from that page, I got the following error


Traceback (most recent call last):
  File "C:/Users/PB/PycharmProjects/crawler/new-crawler.py", line 30, in <module>
    tru_crawler(2)
  File "C:/Users/PB/PycharmProjects/crawler/new-crawler.py", line 16, in tru_crawler
    get_single_item_data(href)
  File "C:/Users/PB/PycharmProjects/crawler/new-crawler.py", line 26, in get_single_item_data
    href = 'http://www.truelocal.com.au' + link.get('href')
TypeError: cannot concatenate 'str' and 'NoneType' objects


Please advise
0 Ola Berglund · January 4, 2015
Yeah I had totally forgotten about this, thanks for reminding me that they are in his videos, great help!!
0 sfolje 0 · January 30, 2016
[ Edited 15.2.2016, working then ]

Tip: when you copy&paste, use tool Find&Replace, to replace question marks into tabs. Warning : dont replace them into spaces, because it causes indentation errors. Replace them with nothing (empty string or ""). Then indent all from beginning.


#Description:
#this program goes on website https://thenewboston.com/search.php?type=0&sort=reputation ,and goes on every user's profile, and on that profile, it prints the first few (approx. 20) links of latest photos. To view photos, click on url or copy in web broser.


import requests
from bs4 import BeautifulSoup


def trade_spider(max_pages):         #this function gets links to user's profiles
    page = 1
    while page <= max_pages:
        url = 'https://thenewboston.com/search.php?type=0&sort=reputation&page==' + str(page)
        source_code = requests.get(url, allow_redirects=False)
        plain_text = source_code.text.encode('ascii', 'replace')
        # soup = BeautifulSoup(plain_text) #this line causes error! (uncomment, to learn new things)
        soup = BeautifulSoup(plain_text,'html.parser')
        for link in soup.findAll('a', {'class': 'user-name'}): #all profiles
            print(' <<---BEGINNING OF LINK--->>')
            print('link: ',link)
            href = link.get('href')
            title = link.string
            print('href: ' ,href)
            print('title: ',title)
            get_single_item_data(href)        #comment this for better understanding
            print(' <<---END OF link--->>')
        print('page: ',page)
        page += 1

def get_single_item_data(item_url):                    #I am now on user's profile and I now print links to users's photos

    print(' <<--- BEGINNING OF get_single_item_data() --->>')
    source_code = requests.get(item_url)
    plain_text = source_code.text
    # soup = BeautifulSoup(plain_text) #this causes error!! (uncomment, to learn new things)
    soup = BeautifulSoup(plain_text,"lxml") #use this line, to avoid error!!
    for item_name in soup.findAll('img', {'class': 'img-responsive'}): # all photos of the user
        print('item_name :',item_name)
        photo='https://thenewboston.com'+item_name.get('src')
        print('Click the link to open the photo: ', photo)    
    print(' <<--- END OF get_single_item_data() --->>')

trade_spider(1)

use this copypaste code to avoid future indentation errors:


#Description:
#this program goes on website https://thenewboston.com/search.php?type=0&sort=reputation ,and goes on every user's profile, and on that profile, it prints the first few (approx. 20) links of latest photos. To view photos, click on url or copy in web broser.


import requests
from bs4 import BeautifulSoup


def trade_spider(max_pages):#this function gets links to user's profiles
page = 1
while page <= max_pages:
url = 'https://thenewboston.com/search.php?type=0&sort=reputation&page==' + str(page)
source_code = requests.get(url, allow_redirects=False)
plain_text = source_code.text.encode('ascii', 'replace')
# soup = BeautifulSoup(plain_text) #this line causes error! (uncomment, to learn new things)
soup = BeautifulSoup(plain_text,'html.parser')
for link in soup.findAll('a', {'class': 'user-name'}):# all profiles
print(' <<---BEGINNING OF LINK--->>')
print('link: ',link)
href = link.get('href')
title = link.string
print('href: ' ,href)
print('title: ',title)
get_single_item_data(href)# comment this for better understanding
print(' <<---END OF link--->>')
print('page: ',page)
page += 1

def get_single_item_data(item_url):# I am now on user's profile and I now print links to users's photos

print(' <<--- BEGINNING OF get_single_item_data() --->>')
source_code = requests.get(item_url)
plain_text = source_code.text
# soup = BeautifulSoup(plain_text)# this causes error!! (uncomment, to learn new things)
soup = BeautifulSoup(plain_text,"lxml")#use this line, to avoid error!!
for item_name in soup.findAll('img', {'class': 'img-responsive'}):# all photos of the user
print('item_name :',item_name)
photo='https://thenewboston.com'+item_name.get('src')
print('Click the link to open the photo: ', photo)
print(' <<--- END OF get_single_item_data() --->>')

trade_spider(1)



I hope this code helps. It prints alot of stuff, to see what is going on with the code - what are our variables being used. Please give me a feedback if it is too messy, if it is helpful or whatever.
0 Carlos G. · July 13, 2016
Hi Bucky. Thanks for these useful tutorials. I have two questions. Thanks for your reply:

1) How would you set a different "user-agent" in this script so you won't be banned when scraping another website?

2) How can you test that your "fake" user-agent is really working (how can you retrieve its value) before running the script on a website?

Thanks!
0 Dimitar Vasilev · January 30, 2016
import requests
from bs4 import BeautifulSoup

the_url = 'https://www.thenewboston.com/forum/topic.php?id=1610'
def getting_the_info(url):
the_real_url=url
response=requests.get(the_real_url)
plain_text=response.text
soup=BeautifulSoup(plain_text, 'html.parser')
for line in soup.find_all('code',):
line = str(line).replace('<br>','\n')
if soup.find_all('td',{'class':'third-column'}):
break
print(line)

# print(plain_text)
getting_the_info(the_url)

Python

107,300 followers
About

This section is all about snakes! Just kidding.

Links
Moderators
Bucky Roberts Administrator