Created
November 11, 2017 08:23
-
-
Save mqu/dfd78ea45dad2072da6e68d3158ce0d1 to your computer and use it in GitHub Desktop.
selenium and youtube player
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/ruby | |
require "selenium-webdriver" | |
require 'pp' | |
# install : | |
# sudo apt-get install ruby ruby-dev chromium-chromedriver firefoxdriver | |
# sudo gem install selenium-webdriver | |
# sudo ln -s /usr/lib/chromium-browser/chromedriver /usr/local/bin/ | |
# usefull links : | |
# https://gist.github.com/kenrett/7553278 | |
# http://www.assertselenium.com/webdriver/getting-started-with-ruby-webdriver/ | |
# python script : https://gist.github.com/sweetmoniker/58c63c6c107384bb642524f0ae1c9a48 | |
# here, select one youtube video URL | |
url='https://www.youtube.com/watch?v=XXXXXXX' | |
# xpath (and css) selectors to get information on HTML content in page. | |
path={} | |
# view counter | |
path[:counter]='/html//yt-view-count-renderer' | |
path[:title]='/html//ytd-video-primary-info-renderer/div/h1' | |
path[:likes]='ytd-toggle-button-renderer.style-scope:nth-child(1) > a:nth-child(1) > yt-formatted-string:nth-child(2)' | |
path[:duration]='/html/body/ytd-app/div[1]/ytd-page-manager/ytd-watch/div[1]/div[1]/div[1]/div/div[20]/div[2]/div[1]/div/span[3]' | |
# title | |
# /html/body/ytd-app/div[1]/ytd-page-manager/ytd-watch/div[1]/div[2]/div/div[6]/div[2]/ytd-video-primary-info-renderer/div/h1 | |
# likes (css): | |
# ytd-toggle-button-renderer.style-scope:nth-child(1) > a:nth-child(1) > yt-formatted-string:nth-child(2) | |
# duration : | |
# /html/body/ytd-app/div[1]/ytd-page-manager/ytd-watch/div[1]/div[1]/div[1]/div/div[20]/div[2]/div[1]/div/span[3] | |
# configure the driver to run in headless mode | |
options = Selenium::WebDriver::Chrome::Options.new | |
options.add_argument('--headless') | |
options.add_argument('--incognito') | |
options.add_argument('--mute-audio') | |
driver = Selenium::WebDriver.for :chrome, options: options | |
#Loading the assertselenium URL | |
driver.navigate.to url | |
# Timeout = 15 sec | |
wait = Selenium::WebDriver::Wait.new(:timeout => 15) | |
wait.until { | |
/https:\/\/www.youtube.com\//.match(driver.page_source) | |
views=driver.find_elements(:xpath, path[:counter]).first.text | |
title=driver.find_elements(:xpath, path[:title]).first.text | |
likes=driver.find_elements(:css, path[:likes]).first.text | |
duration=driver.find_elements(:xpath, path[:duration]).first.text | |
puts "views: #{views} ; likes: #{likes} ; len:#{duration} - #{title}" | |
#read for some time video | |
sleep 15 | |
#Quitting the browser | |
driver.quit | |
exit 0 | |
} | |
throw exception "could not open url : #{ur}" | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment