What is the fastest way to open urls in new tabs v

Question

I want to create a python script to open up a lot tabs

import os
import selenium
from selenium import webdriver


driver =webdriver.Chrome('/usr/local/bin/chromedriver')
driver.execute_script("window.open('http://www.msn.com');")
driver.execute_script("window.open('http://www.cnn.com');")
driver.execute_script("window.open('http://www.yahoo.com');")
driver.execute_script("window.open('https://www.apple.com');")
driver.execute_script("window.open('https://www.google.com');")
driver.execute_script("window.open('https://www.stackoverflow.com');")

# driver.quit()

When I run I get

Is what I have right now is the fastest way?

---

I used to have this

# -*- coding: utf-8 -*-

import os
import selenium
from selenium import webdriver
from selenium.webdriver.common.keys import Keys

driver =webdriver.Chrome('/usr/local/bin/chromedriver')

#1
driver.get("http://msn.com")

#2
driver.find_element_by_tag_name('body').send_keys(Keys.COMMAND + 't')
driver.switch_to.window(driver.window_handles[-1])
driver.get("http://cnn.com")

#3
driver.find_element_by_tag_name('body').send_keys(Keys.COMMAND + 't')
driver.switch_to.window(driver.window_handles[-1])
driver.get("http://www.yahoo.com")

#4
driver.find_element_by_tag_name('body').send_keys(Keys.COMMAND + 't')
driver.switch_to.window(driver.window_handles[-1])
driver.get("https://www.apple.com")

#5
driver.find_element_by_tag_name('body').send_keys(Keys.COMMAND + 't')
driver.switch_to.window(driver.window_handles[-1])
driver.get("https://www.google.com")

#6
driver.find_element_by_tag_name('body').send_keys(Keys.COMMAND + 't')
driver.switch_to.window(driver.window_handles[-1])
driver.get("https://www.stackoverflow.com")

It works but it is painfully slow.

---

I start with 6 now, but I want to load 100 tabs.

Also, how do I get rid of my first weird looking tab? I am even sure why it is there.

Answer 1

So we already have have 2 approaches at out disposal to open New TAB with Selenium.

Your previous approach with send_keys(Keys.COMMAND + 't') is bound to consume more time as we switch_to.window() and get("http://www.url.com") too. Hence slower.

Your current approach with execute_script is bound to be faster as we are simply injecting Java Scripts to open new TAB s with URL s.

Now, the reason you see a Blank Window is because once you initialized the browser through :

driver =webdriver.Chrome('/usr/local/bin/chromedriver')

After this, we haven't invoked the get() method to open any URL. Rather we have straight taken help of JavascriptExecutor to open NEW TAB s. Hence the first TAB remains unused.

---

Update :

To bring the unused TAB into the play, we can access the first of the URL s through the first TAB invoking get() method as follows :

driver =webdriver.Chrome('/usr/local/bin/chromedriver')
driver.get("http://www.msn.com")
driver.execute_script("window.open('http://www.cnn.com');")
driver.execute_script("window.open('http://www.yahoo.com');")
driver.execute_script("window.open('https://www.apple.com');")
driver.execute_script("window.open('https://www.google.com');")
driver.execute_script("window.open('https://www.stackoverflow.com');")

---

Summary :

To open a New Blank TAB you can use the following line of code :

driver.execute_script("window.open('','_blank');");

To open a New TAB with url you can use the following line of code :

driver.execute_script("window.open('http://facebook.com/');");

Answer 2

Upgrade the chromedriver(>2.25)/chrome browser(>55.0) on your MAC to remove the empty data; tab. You can use threading or multiprocessing to speed up the process.

from multiprocessing import Process
import os
import selenium
from selenium import webdriver
from selenium.webdriver.common.keys import Keys

driver =webdriver.Chrome('/usr/local/bin/chromedriver')
driver.get("http://msn.com")
def func1():
  print 'launching: MSN'
 driver.execute_script("window.open('http://www.msn.com');")

def func2():
  print 'launching: CNN'
  driver.execute_script("window.open('http://www.cnn.com');")

if __name__ == '__main__':
  p1 = Process(target=func1)
  p1.start()
  p2 = Process(target=func2)
  p2.start()
  p1.join()
  p2.join()

def runInParallel(*fns):
  proc = []
  for fn in fns:
    p = Process(target=fn)
    p.start()
    proc.append(p)
  for p in proc:
    p.join()

runInParallel(func1, func2)

Depending on how many CPUs you have, the load of the machine, the timing of many things happening in the computer will have an influence on the time the threads/process start. Also, since the processes are started right after creation, the overhead of creating a process also has to be calculated in the time difference you see.