Develop Reference

Bypass Yahoo reCAPTCHA by 2captcha API

I'm trying to create yahoo account using python selenium, and while creating i have to bypass a recaptcha. I'm using 2Captcha API to automate solving captchas
My issue is i can't solve recaptcha
based on my tests, i noticed that yahoo is using entreprise captcha, not sure if it's a V2 or V3
Here the API Documentation : https://2captcha.com/2captcha-api
here is my code :
import os
import time
from selenium.webdriver.chrome.options import Options
from selenium import webdriver
import random
from twocaptcha import TwoCaptcha
opt = Options()
opt.add_argument("--disable-infobars")
opt.add_argument("start-maximized")
# Pass the argument 1 to allow and 2 to block
opt.add_experimental_option("excludeSwitches", ["enable-logging"])
opt.add_experimental_option("prefs", {
"profile.default_content_setting_values.media_stream_mic": 2,
"profile.default_content_setting_values.media_stream_camera": 2,
"profile.default_content_setting_values.geolocation": 2,
"profile.default_content_setting_values.notifications": 2
})
executable_path = r'chromedriver'
os.environ["webdriver.chrome.driver"] = executable_path
global driver
driver = webdriver.Chrome(r'chromedriver', options=opt)
time.sleep(5)
driver.get("https://login.yahoo.com/account/create")
# Fname and Lname
time.sleep(6)
driver.find_element_by_xpath("//input[#name='firstName']").send_keys("fname")
time.sleep(3)
driver.find_element_by_xpath("//input[#name='lastName']").send_keys("lname")
# Email
time.sleep(3)
numberid = random.randint(100000, 900000)
driver.find_element_by_xpath("//input[#name='yid']").send_keys("fname" + str(numberid) + "lname")
# Password
time.sleep(3)
driver.find_element_by_xpath("//input[#name='password']").send_keys("TestEPWD.")
######## number region +
FC = '(+212)'
option_el = driver.find_element_by_xpath("//option[contains(text(),'%s')]" % FC)
option_el.click()
driver.find_element_by_xpath("//input[#name='phone']").send_keys('684838340')
# Choose date
Month = random.randint(1, 12)
Months = "//option[#value='{}']".format(Month)
monthselect = driver.find_element_by_xpath(Months)
monthselect.click()
time.sleep(3)
Day = random.randint(1, 27)
driver.find_element_by_xpath("//input[#name='dd']").send_keys(Day)
time.sleep(3)
Year = random.randint(1975, 2000)
driver.find_element_by_xpath("//input[#name='yyyy']").send_keys(Year)
time.sleep(3)
list = ["Man", "Woman"]
item = random.choice(list)
driver.find_element_by_xpath("//input[#name='freeformGender']").send_keys(item)
time.sleep(3)
driver.find_element_by_xpath("//button[#name='signup']").click()
time.sleep(5)
# CAPTCHA PART :
api_key = os.getenv('APIKEY_2CAPTCHA', 'mycaptchaAPI')
solver = TwoCaptcha(api_key)
yy = driver.current_url
try:
result = solver.recaptcha(
sitekey="6LeGXAkbAAAAAAMGHQaxwylqpyvtF2jJMrvJff1h",
url=yy,
entreprise=1,
version='v3',
score=0.2
)
except Exception as e:
print(e)
else:
print('result: ' + str(result))

Python Selenium Multiple Webdrivers

So I have this python selenium code and I want it to run multiple times at the same time. So when I activate it, it opens multiple webdrivers and execute this script at the same time. How can I do this?
driver.get(base_url)
password_id = driver.find_element_by_id('password')
password = input("Password: ")
password_id.send_keys(password)
password_id.send_keys(Keys.ENTER)
email_id1 = EC.presence_of_element_located((By.ID, 'email'))
WebDriverWait(driver, 100).until(email_id1)
email_id = driver.find_element_by_id('email')
email_id.send_keys(user_email)
start = time.time()
print(Fore.WHITE + "STATUS:" + Fore.LIGHTYELLOW_EX + " Email Filled!")
name_id = driver.find_element_by_id('name')
name_id.send_keys(user_name)
print(Fore.WHITE + "STATUS:" + Fore.LIGHTYELLOW_EX + " Name Filled!")
button_id = driver.find_element_by_id('purchase')
end = time.time()
button_id.click()
print(Fore.WHITE + "STATUS:" + Fore.LIGHTGREEN_EX + " Processing order...")
sleep(10)
timeresult = end - start
speed = (str(timeresult))
checkout_done = driver.current_url

Have you tried multithreading?
Code below allows me to open two browsers at once (in separate loops) and control them within functions (first_window(), second_window()).
from selenium import webdriver
from threading import Thread
def first_window():
driver = webdriver.Chrome('chromedriver.exe')
driver.get("https://stackoverflow.com/")
def second_window():
driver = webdriver.Chrome('chromedriver.exe')
driver.get("https://stackoverflow.com/")
if __name__ == '__main__':
Thread(target=first_window).start()
Thread(target=second_window).start()

Python Selenium - Cannot close tab using google shortcuts

I am working on a little selenium project, but I got some issues.
So what I need to do is to click on a link to open it in a new tab, and whenever I have taken the information I need to close that tab and go into the next one. driver.close() does not work as it gives me the error: Message: no such window: target window already closed. So I intstead tried this (saw this while researching):
driver.find_element_by_tag_name('html').send_keys(Keys.CONTROL + 'w'), and I also tried with adding Keys.F4, but nothing worked.
It seems to work for other people, so why not for me?
Code:
def cpuFunc():
i = 0
print("Launching CPU")
cpu = webdriver.Chrome('chromedriver.exe',options=option)
cpu.get('https://www.komplett.se/category/11204/datorutrustning/datorkomponenter/processor')
cpu.find_element_by_xpath('/html/body/div[1]/div[2]/div[1]/div/div/div[2]/form/div/div[1]/button').click()
#while i < 10:
# cpu.find_element_by_tag_name('html').send_keys(Keys.END)
# i += 1
# time.sleep(0.5)
#print("At bottom: CPU")
cpu.find_element_by_tag_name('body').send_keys(Keys.CONTROL + Keys.HOME)
time.sleep(0.5)
link = cpu.find_element_by_xpath(f'/html/body/main/div/div[2]/div[5]/div[2]/form/div[1]/a')
ActionChains(cpu).key_down(Keys.CONTROL).click(link).key_up(Keys.CONTROL).perform()
time.sleep(1)
window = cpu.window_handles[-1]
cpu.switch_to.window(window)
title = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/section/div/section/div[1]/h1/span").text
price = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/section/div/section/div[3]/div[2]/div[1]/div/div/div[1]/div[1]/div[1]/span").text
btn = cpu.find_element_by_xpath('/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/button')
time.sleep(0.5)
cpu.execute_script("arguments[0].click();", btn)
core = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/div/div/div/table[2]/tbody/tr[2]/td").text
thread = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/div/div/div/table[2]/tbody/tr[3]/td").text
cache = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/div/div/div/table[2]/tbody/tr[4]/td").text
clock = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/div/div/div/table[2]/tbody/tr[7]/td").text
turbo = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/div/div/div/table[2]/tbody/tr[8]/td").text
socket = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/div/div/div/table[2]/tbody/tr[9]/td").text
wattage = cpu.find_element_by_xpath("/html/body/div[2]/main/div[2]/div[2]/div[3]/div/div[2]/div/section[2]/div/div/div/table[2]/tbody/tr[10]/td").text
cpu.find_element_by_tag_name('html').send_keys(Keys.CONTROL + 'w') # Here it shall close
time.sleep(60000)
enter code here

You can simply use ActionChains & Keys.
from selenium.webdriver import ActionChains
from selenium.webdriver.common.keys import Keys
If MacOs:
step_1 = ActionChains(cpu)
step_1.send_keys(Keys.COMMAND + 'w')
If Windows:
step_1 = ActionChains(cpu)
step_1.send_keys(Keys.CONTROL + 'w')
Hope it helps, if issues please comment.

Driver.close() worked for me after also fixed a lot of those full xpaths which are easily breakable and added webdriver waits for stability in finding elements.
wait = WebDriverWait(cpu, 10)
cpu.get('https://www.komplett.se/category/11204/datorutrustning/datorkomponenter/processor')
wait.until(EC.element_to_be_clickable((By.XPATH, "//*[#class='btn-large primary'][#type='submit']"))).click()
link = wait.until(EC.element_to_be_clickable((By.CSS_SELECTOR, "form > div:nth-child(1) > a")))
ActionChains(cpu).key_down(Keys.CONTROL).click(link).key_up(Keys.CONTROL).perform()
time.sleep(1)
window = cpu.window_handles[-1]
cpu.switch_to.window(window)
title = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, "div.product-main-info__info > h1 > span"))).text
price = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, "div.product-price > span"))).text
print(title,price)
btn = wait.until(EC.element_to_be_clickable((By.CSS_SELECTOR, "section.product-section.technical-details.col-xs-12 > button")))
btn.click()
table2 = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, "table:nth-child(2) > tbody")))
core = table2.find_element_by_xpath("./tr[2]/td").text
thread = table2.find_element_by_xpath("./tr[3]/td").text
cache = table2.find_element_by_xpath("./tr[4]/td").text
clock = table2.find_element_by_xpath("./tr[7]/td").text
turbo = table2.find_element_by_xpath("./tr[8]/td").text
socket = table2.find_element_by_xpath("./tr[9]/td").text
wattage = table2.find_element_by_xpath("./tr[10]/td").text
cpu.close()
print(core,thread,cache,clock,turbo,socket,wattage)
Import
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC

Why ActionChains(driver).move_to_element(elem).click().perform() twice

I try to crawl the wechat public accounts includes the key word through "http://weixin.sogou.com/"
But i find i must use twice ActionChains(driver).move_to_element(nextpage).click().perform()，it can still work,and go to the next page !
who can tell me why and how to fix ! Thank you!
The source code are as follow , and sorry the comments are in the Chinese .
# coding=utf-8
import time
from selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
key = u"江南大学" #搜索的关键词
driver = webdriver.Chrome()
driver.get("http://weixin.sogou.com/")
assert u'搜狗微信' in driver.title
elem = driver.find_element_by_id("upquery")
elem.clear()
elem.send_keys(key)
button = driver.find_element_by_class_name("swz2") #搜索公众号
button.click()
WebDriverWait(driver,10).until(
EC.title_contains(key)
)
count = 0
while True:
for i in range(10):
try:
wechat_name = driver.find_element_by_xpath("//*[#id=\"sogou_vr_11002301_box_{}\"]/div[2]/h3".format(i)).text
print wechat_name
wechat_id = driver.find_element_by_xpath("//*[#id=\"sogou_vr_11002301_box_{}\"]/div[2]/h4/span/label".format(i)).text
print wechat_id
wechat_intro = driver.find_element_by_xpath("//*[#id=\"sogou_vr_11002301_box_{}\"]/div[2]/p[1]/span[2]".format(i)).text
print wechat_intro
print "*************************"
count += 1
except:
pass
try:
nextpage = driver.find_element_by_xpath("//*[#id=\"sogou_next\"]") #下一页的按钮
actions = ActionChains(driver)
actions.move_to_element(nextpage)
actions.click().
actions.perform()
actions = ActionChains(driver)
actions.move_to_element(nextpage)
actions.click().
actions.perform()
except Exception,e:
print e
break
driver.quit()
print count

You can chain your action, so no need to do perform after each action.
actions = ActionChains(driver)
actions.move_to_element(nextpage)
actions.click(nextpage)
actions.perform()
OR
actions = ActionChains(driver)
actions.move_to_element(nextpage)
actions.click(nextpage).perform()

How to scrape this site for topics that I answer in

Question
How can I modify my script to successfully show me the number of answers that I have made by topics.
Code
This was a script that I tried
import time
from selenium import webdriver
from selenium.webdriver.common.keys import Keys
def get_topics('Juan-Gallardo'):
url = "http://www.quora.com/" + 'Juan-Gallardo' + "/topics"
browser = webdriver.Chrome()
browser.get(url)
time.sleep(2)
bod = browser.find_element_by_tag_name("body")
no_of_pagedowns = 40
while no_of_pagedowns:
bod.send_keys(Keys.PAGE_DOWN)
time.sleep(0.3)
no_of_pagedowns-=1
topics = [t.text.encode('ascii', 'replace') for t in browser.find_elements_by_class_name("name_text")]
counts = [c.text.encode('ascii', 'replace').split(' ')[0] for c in browser.find_elements_by_class_name("name_meta")]
li = [[topics[i], int(counts[i])] for i in xrange(len(topics)) if counts[i] != '']
browser.quit()
return li
Errors

You need to define an argument for the get_topics() function:
def get_topics(user):
url = "http://www.quora.com/" + user + "/topics"
...
Then, call the function this way:
get_topics('Juan-Gallardo')