How to use BeautifulSoup to parse google search results in Python
Question:
I am trying to parse the first page of google search results. Specifically, the Title and the small Summary that is provided. Here is what I have so far:
from urllib.request import urlretrieve
import urllib.parse
from urllib.parse import urlencode, urlparse, parse_qs
import webbrowser
from bs4 import BeautifulSoup
import requests
address = 'https://google.com/#q='
# Default Google search address start
file = open( "OCR.txt", "rt" )
# Open text document that contains the question
word = file.read()
file.close()
myList = [item for item in word.split('n')]
newString = ' '.join(myList)
# The question is on multiple lines so this joins them together with proper spacing
print(newString)
qstr = urllib.parse.quote_plus(newString)
# Encode the string
newWord = address + qstr
# Combine the base and the encoded query
print(newWord)
source = requests.get(newWord)
soup = BeautifulSoup(source.text, 'lxml')
The part I am stuck on now is going down the HTML path to parse the specific data that I want. Everything I have tried so far has just thrown an error saying that it has no attribute or it just gives back “[]”.
I am new to Python and BeautifulSoup so I am not sure the syntax of how to get to where I want. I have found that these are the individual search results in the page:
Any help on what to add to parse the Title and Summary of each search result would be MASSIVELY appreciated.
Thank you!
Answers:
Your url doesn’t work for me. But with https://google.com/search?q= I get results.
import urllib
from bs4 import BeautifulSoup
import requests
import webbrowser
text = 'hello world'
text = urllib.parse.quote_plus(text)
url = 'https://google.com/search?q=' + text
response = requests.get(url)
#with open('output.html', 'wb') as f:
# f.write(response.content)
#webbrowser.open('output.html')
soup = BeautifulSoup(response.text, 'lxml')
for g in soup.find_all(class_='g'):
print(g.text)
print('-----')
- Default Google search address start doesn’t contain
# symbol. Instead, it should have ? and /search pathname:
---> https://google.com/#q=
---> https://www.google.com/search?q=cake
- Make sure you’re passing
user-agent into HTTP request headers because the default requests user-agent is python-requests and sites could identify that it’s a bot and block the request thus you would receive a different HTML with some sort of an error that contains different elements/selectors which is the reason you were getting an empty result.
Check what’s your user-agent, and a list of user-agents for mobile, tablets, etc.
Pass user-agent in request headers:
headers = {
'User-agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/106.0.0.0 Safari/537.36'
}
requests.get('YOUR_URL', headers=headers)
Code and example in the online IDE:
from bs4 import BeautifulSoup
import requests, json, lxml
headers = {
'User-agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/106.0.0.0 Safari/537.36'
}
# https://requests.readthedocs.io/en/latest/user/quickstart/#passing-parameters-in-urls
params = {
'q': 'tesla', # query
'gl': 'us', # country to search from
'hl': 'en' # language
}
# https://requests.readthedocs.io/en/latest/user/quickstart/#timeouts
html = requests.get("https://www.google.com/search", headers=headers, params=params, timeout=30)
soup = BeautifulSoup(html.text, 'lxml')
data = []
for result in soup.select('.tF2Cxc'):
title = result.select_one('.DKV0Md').text
link = result.select_one('.yuRUbf a')['href']
# sometimes there's no description and we need to handle this exception
try:
snippet = result.select_one('#rso .lyLwlc').text
except: snippet = None
data.append({
'title': title,
'link': link,
'snippet': snippet
})
print(json.dumps(data, indent=2, ensure_ascii=False))
-------------
'''
[
{
"title": "Tesla: Electric Cars, Solar & Clean Energy",
"link": "https://www.tesla.com/",
"snippet": "Tesla is accelerating the world's transition to sustainable energy with electric cars, solar and integrated renewable energy solutions for homes and ..."
},
{
"title": "Tesla, Inc. - Wikipedia",
"link": "https://en.wikipedia.org/wiki/Tesla,_Inc.",
"snippet": "Tesla, Inc. is an American electric vehicle and clean energy company based in Palo Alto, California, United States. Tesla designs and manufactures electric ..."
},
{
"title": "Nikola Tesla - Wikipedia",
"link": "https://en.wikipedia.org/wiki/Nikola_Tesla",
"snippet": "Nikola Tesla was a Serbian-American inventor, electrical engineer, mechanical engineer, and futurist best known for his contributions to the design of the ..."
}
]
'''
Alternatively, you can achieve the same thing by using Google Organic Results API from SerpApi. It’s a paid API with a free plan just to test the API.
The difference in your case is that you don’t have to figure out why the output is empty and what causes this to happen, bypass blocks from Google or other search engines, and maintain the parser over time.
Instead, you only need to grab the data from the structured JSON you want.
Example code to integrate:
import os
from serpapi import GoogleSearch
params = {
"engine": "google", # serpapi parsing engine
"q": "tesla", # search query
"hl": "en", # language of the search
"gl": "us", # country from where search initiated
"api_key": os.getenv("API_KEY") # your serpapi API key
}
search = GoogleSearch(params) # data extraction on the SerpApi backend
results = search.get_dict() # JSON -> Python dict
for result in results["organic_results"]:
print(f"Title: {result['title']}nSummary: {result['snippet']}nLink: {result['link']}n")
----------
'''
Title: Tesla: Electric Cars, Solar & Clean Energy
Summary: Tesla is accelerating the world's transition to sustainable energy with electric cars, solar and integrated renewable energy solutions for homes and ...
Link: https://www.tesla.com/
Title: Tesla, Inc. - Wikipedia
Summary: Tesla, Inc. is an American electric vehicle and clean energy company based in Palo Alto, California, United States. Tesla designs and manufactures electric ...
Link: https://en.wikipedia.org/wiki/Tesla,_Inc.
'''
Disclaimer, I work for SerpApi.
I am trying to parse the first page of google search results. Specifically, the Title and the small Summary that is provided. Here is what I have so far:
from urllib.request import urlretrieve
import urllib.parse
from urllib.parse import urlencode, urlparse, parse_qs
import webbrowser
from bs4 import BeautifulSoup
import requests
address = 'https://google.com/#q='
# Default Google search address start
file = open( "OCR.txt", "rt" )
# Open text document that contains the question
word = file.read()
file.close()
myList = [item for item in word.split('n')]
newString = ' '.join(myList)
# The question is on multiple lines so this joins them together with proper spacing
print(newString)
qstr = urllib.parse.quote_plus(newString)
# Encode the string
newWord = address + qstr
# Combine the base and the encoded query
print(newWord)
source = requests.get(newWord)
soup = BeautifulSoup(source.text, 'lxml')
The part I am stuck on now is going down the HTML path to parse the specific data that I want. Everything I have tried so far has just thrown an error saying that it has no attribute or it just gives back “[]”.
I am new to Python and BeautifulSoup so I am not sure the syntax of how to get to where I want. I have found that these are the individual search results in the page:
Any help on what to add to parse the Title and Summary of each search result would be MASSIVELY appreciated.
Thank you!
Your url doesn’t work for me. But with https://google.com/search?q= I get results.
import urllib
from bs4 import BeautifulSoup
import requests
import webbrowser
text = 'hello world'
text = urllib.parse.quote_plus(text)
url = 'https://google.com/search?q=' + text
response = requests.get(url)
#with open('output.html', 'wb') as f:
# f.write(response.content)
#webbrowser.open('output.html')
soup = BeautifulSoup(response.text, 'lxml')
for g in soup.find_all(class_='g'):
print(g.text)
print('-----')
- Default Google search address start doesn’t contain
#symbol. Instead, it should have?and/searchpathname:
---> https://google.com/#q=
---> https://www.google.com/search?q=cake
- Make sure you’re passing
user-agentinto HTTP request headers because the defaultrequestsuser-agentispython-requestsand sites could identify that it’s a bot and block the request thus you would receive a different HTML with some sort of an error that contains different elements/selectors which is the reason you were getting an empty result.
Check what’s your user-agent, and a list of user-agents for mobile, tablets, etc.
Pass user-agent in request headers:
headers = {
'User-agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/106.0.0.0 Safari/537.36'
}
requests.get('YOUR_URL', headers=headers)
Code and example in the online IDE:
from bs4 import BeautifulSoup
import requests, json, lxml
headers = {
'User-agent':
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/106.0.0.0 Safari/537.36'
}
# https://requests.readthedocs.io/en/latest/user/quickstart/#passing-parameters-in-urls
params = {
'q': 'tesla', # query
'gl': 'us', # country to search from
'hl': 'en' # language
}
# https://requests.readthedocs.io/en/latest/user/quickstart/#timeouts
html = requests.get("https://www.google.com/search", headers=headers, params=params, timeout=30)
soup = BeautifulSoup(html.text, 'lxml')
data = []
for result in soup.select('.tF2Cxc'):
title = result.select_one('.DKV0Md').text
link = result.select_one('.yuRUbf a')['href']
# sometimes there's no description and we need to handle this exception
try:
snippet = result.select_one('#rso .lyLwlc').text
except: snippet = None
data.append({
'title': title,
'link': link,
'snippet': snippet
})
print(json.dumps(data, indent=2, ensure_ascii=False))
-------------
'''
[
{
"title": "Tesla: Electric Cars, Solar & Clean Energy",
"link": "https://www.tesla.com/",
"snippet": "Tesla is accelerating the world's transition to sustainable energy with electric cars, solar and integrated renewable energy solutions for homes and ..."
},
{
"title": "Tesla, Inc. - Wikipedia",
"link": "https://en.wikipedia.org/wiki/Tesla,_Inc.",
"snippet": "Tesla, Inc. is an American electric vehicle and clean energy company based in Palo Alto, California, United States. Tesla designs and manufactures electric ..."
},
{
"title": "Nikola Tesla - Wikipedia",
"link": "https://en.wikipedia.org/wiki/Nikola_Tesla",
"snippet": "Nikola Tesla was a Serbian-American inventor, electrical engineer, mechanical engineer, and futurist best known for his contributions to the design of the ..."
}
]
'''
Alternatively, you can achieve the same thing by using Google Organic Results API from SerpApi. It’s a paid API with a free plan just to test the API.
The difference in your case is that you don’t have to figure out why the output is empty and what causes this to happen, bypass blocks from Google or other search engines, and maintain the parser over time.
Instead, you only need to grab the data from the structured JSON you want.
Example code to integrate:
import os
from serpapi import GoogleSearch
params = {
"engine": "google", # serpapi parsing engine
"q": "tesla", # search query
"hl": "en", # language of the search
"gl": "us", # country from where search initiated
"api_key": os.getenv("API_KEY") # your serpapi API key
}
search = GoogleSearch(params) # data extraction on the SerpApi backend
results = search.get_dict() # JSON -> Python dict
for result in results["organic_results"]:
print(f"Title: {result['title']}nSummary: {result['snippet']}nLink: {result['link']}n")
----------
'''
Title: Tesla: Electric Cars, Solar & Clean Energy
Summary: Tesla is accelerating the world's transition to sustainable energy with electric cars, solar and integrated renewable energy solutions for homes and ...
Link: https://www.tesla.com/
Title: Tesla, Inc. - Wikipedia
Summary: Tesla, Inc. is an American electric vehicle and clean energy company based in Palo Alto, California, United States. Tesla designs and manufactures electric ...
Link: https://en.wikipedia.org/wiki/Tesla,_Inc.
'''
Disclaimer, I work for SerpApi.