Skip to content

Instantly share code, notes, and snippets.

View dimitryzub's full-sized avatar
🇺🇦
Grateful

Dmitiry Zub☀️ dimitryzub

🇺🇦
Grateful
45 followers · 12 following
View GitHub Profile
@dimitryzub
dimitryzub / bing_bs4_scrape_first_10_video_results.py
Created July 17, 2021 07:40
from bs4 import BeautifulSoup
import requests, lxml
headers = {
"User-Agent":
"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.114 Safari/537.36"
}
params = {
"q": "somebody toucha my spaghet",
@dimitryzub
dimitryzub / scrape_all_bing_video_results.py
Created July 17, 2021 07:37
import time
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
driver = webdriver.Chrome(executable_path='path/to/chromedriver.exe')
driver.get('https://www.bing.com/videos/search?q=somebody+toucha+my+spaghet&FORM=HDRSC3&cc=us')
time.sleep(1)
# scrolls until "more videos" button is located
while True:
@dimitryzub
dimitryzub / duckduckgo_scrape_inline_images.py
Created July 16, 2021 14:27
from selenium import webdriver
import re, urllib.parse
driver = webdriver.Chrome(executable_path='path/to/chromedriver.exe')
driver.get('https://duckduckgo.com/?q=elon musk dogecoin&kl=us-en&ia=web')
for result in driver.find_elements_by_css_selector('.js-images-link'):
title = result.find_element_by_css_selector('.js-images-link a img').get_attribute('alt')
link = result.find_element_by_css_selector('.js-images-link a').get_attribute('href')
thumbnail_encoded = result.find_element_by_css_selector('.js-images-link a img').get_attribute('src')
@dimitryzub
dimitryzub / duckduckgo_scrape_inline_videos.py
Last active August 4, 2021 06:38
from selenium import webdriver
import re, urllib.parse
driver = webdriver.Chrome(executable_path='path/to/chromedriver.exe')
driver.get('https://duckduckgo.com/?q=elon musk dogecoin&kl=us-en&ia=web')
for result in driver.find_elements_by_css_selector('#m3-0 .has-image'):
title = result.find_element_by_css_selector('#m3-0 .js-carousel-item-title').text
link = result.find_element_by_css_selector('#m3-0 .module--carousel__body a').get_attribute('href')
@dimitryzub
dimitryzub / scrape_duckduckgo_organic_news_results.py
Last active August 4, 2021 08:38
from selenium import webdriver
import urllib.parse, re
driver = webdriver.Chrome(executable_path='path/tochromedriver.exe')
driver.get('https://duckduckgo.com/?q=elon musk&kl=us-en&ia=web')
for result in driver.find_elements_by_css_selector('#m1-0 .has-image'):
title = result.find_element_by_css_selector('#m1-0 .js-carousel-item-title').text.strip()
link = result.find_element_by_css_selector('#m1-0 .js-carousel-item-title').get_attribute('href')
source = result.find_element_by_css_selector('#m1-0 .result__url').text
@dimitryzub
dimitryzub / scrape_duckduckgo_ad_results.py
Last active August 6, 2021 08:52
DuckDuckGo Scrape Ad Results
from selenium import webdriver
driver = webdriver.Chrome(executable_path='path/to/chromedriver.exe')
driver.get('https://duckduckgo.com/?q=rtx 3080&kl=us-en&ia=web')
for result in driver.find_elements_by_css_selector('.results--ads .result__body.links_main.links_deep'):
title = result.find_element_by_css_selector('.js-result-title-link').text
link = result.find_element_by_css_selector('.js-result-title-link').get_attribute('href')
source = result.find_element_by_css_selector('.js-result-extras-url').text
snippet = result.find_element_by_css_selector('.js-result-snippet > at').text
@dimitryzub
dimitryzub / bing_related_question.py
Created July 5, 2021 07:10
bing_related_question
@dimitryzub
dimitryzub / serpapi_scrape_bing_related_searches.py
Created June 26, 2021 09:22
serpapi_scrape_bing_related_searches
@dimitryzub
dimitryzub / python_scrape_bing_related_searches.py
Created June 26, 2021 09:22
python_scrape_bing_related_searches
@dimitryzub
dimitryzub / python_scrape_bing_related_search_results.py
Created June 26, 2021 09:21
python_scrape_bing_related_search_results