使用selenium解决验证码问题

Question

我下面的代码是不断解决不同的验证码！请纠正我的错误，因为我不知道是什么原因造成的！

from selenium import webdriver
from python3_anticaptcha import ImageToTextTask, CallbackClient
import time
import requests

browser = webdriver.Firefox()

url = 'https://urlmased.com/'
browser.get(url)
time.sleep(10)
username = browser.find_element_by_id("masked")
username.send_keys("testuser")
password = browser.find_element_by_id("masked")
password.send_keys("testpass")

image_link = browser.find_element_by_xpath(
    '//*[@id="masked"]').get_attribute('src')
pic = requests.get(image_link)
if pic.status_code == 200:
    with open("image.png", 'wb') as f:
        f.write(pic.content)
ANTICAPTCHA_KEY = 'masked'
captcha_file = "image.png"
result = ImageToTextTask.ImageToTextTask(
    anticaptcha_key=ANTICAPTCHA_KEY).captcha_handler(captcha_file=captcha_file)

captcha = browser.find_element_by_id("masked")
captcha.send_keys(result['solution']['text'])
login = browser.find_element_by_id("yw2")

请注意，该 API 已激活，因此您可以使用它，直到找到解决方案。然后我会改变它。

求解准确率也是100%

Answer 1

这里的问题是页面源中的验证码 URL 不是实际的图像 URL。它是一个动态生成验证码图像的脚本，当您使用验证码求解器 API 时，您正在求解与浏览器加载的图像不同的图像。为了解决这个问题，我们必须保存浏览器加载的相同图像。我跟踪图像请求并发现，它使用浏览器加载页面时生成的唯一 cookie。

使用硒：

from selenium import webdriver
from python3_anticaptcha import ImageToTextTask, CallbackClient
from time import sleep
import requests



def GetImageCookies():
    print('Extracting Browser Cookies')
    image_cookies = ''
    for cookie in browser.get_cookies():
        if cookie['name'] == 'ssc':
            image_cookies += 'ssc={};'.format(cookie['value'])
        elif cookie['name'] == 'ghsdfkjlksssalk35bbr':
            image_cookies += 'ghsdfkjlksssalk35bbr={};'.format(cookie['value'])
    # print(image_cookies)
    return image_cookies

def SaveImage(captcha_file = "master.jpg"):
    print('Saving the captcha image')
    header = {
    'Accept': 'image/webp,image/apng,image/*,*/*;q=0.8',
    'Accept-Encoding': 'gzip, deflate, br',
    'Accept-Language': 'en,en-US;q=0.9,ar;q=0.8',
    'Cookie': GetImageCookies(),
    'Host': 'masked',
    'Referer': 'masked',
    'Sec-Fetch-Mode': 'no-cors',
    'Sec-Fetch-Site': 'same-origin',
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:70.0) Gecko/20100101 Firefox/70.0'}


    pic = requests.get('https://masked/site/captcha/v/',verify=False,headers = header)
    if pic.status_code == 200:
        with open(captcha_file, 'wb') as f:
            f.write(pic.content)

def SolveCapcha(captcha_file = "master.jpg"):
    print('Solving the captcha image')
    ANTICAPTCHA_KEY = 'masked'  
    result = ImageToTextTask.ImageToTextTask(
        anticaptcha_key=ANTICAPTCHA_KEY).captcha_handler(captcha_file=captcha_file)
    captcha_text = result['solution']['text']
    print('Captcha text is :',captcha_text)
    return captcha_text


browser = webdriver.Firefox()
url = 'https://masked/'
browser.get(url)
def Login():
    SaveImage()
    sleep(5)
    username = browser.find_element_by_id("masked_username")
    username.clear()
    username.send_keys("testuser")
    password = browser.find_element_by_id("masked")
    password.clear()
    password.send_keys("testpass")
    captcha = browser.find_element_by_id("masked")
    captcha.clear()
    captcha_text = SolveCapcha()
    captcha.send_keys(captcha_text)
    login = browser.find_element_by_id("masked").click()
    sleep(5)
    err_message = browser.find_elements_by_id('masked')
    if err_message :
        if err_message[0].text == 'The verification code is incorrect.':
            print(err_message[0].text)
            return False
    return True


"""The logic here is that the image gets downloaded using the cookies but sometimes
the letters are hard to be solved so each time we download the same image with the
same cookies the content of the image will be the same but how it's written is different
So we keep trying till we get it right """
while Login() == False:
    Login()

使用请求和美丽的汤：以下是不确定它是否有效的想法，您必须自己测试：

from bs4 import BeautifulSoup
def SaveImage(captcha_file = "master.jpg"):
    print('Saving the captcha image')
    header = {
    'Accept': 'image/webp,image/apng,image/*,*/*;q=0.8',
    'Accept-Encoding': 'gzip, deflate, br',
    'Accept-Language': 'en,en-US;q=0.9,ar;q=0.8',
    'Host': 'masked',
    'Referer': 'https://masked/',
    'Sec-Fetch-Mode': 'no-cors',
    'Sec-Fetch-Site': 'same-origin',
    'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:70.0) Gecko/20100101 Firefox/70.0'}


    pic = session.get('https://masked/site/captcha/v/',verify=False,headers = header)
    if pic.status_code == 200:
        with open(captcha_file, 'wb') as f:
            f.write(pic.content)

with requests.Session() as session:
    source      = session.get(url = 'https://masked/',verify=False) # To get the itial cookies  
    soup        = BeautifulSoup(source.text, 'html.parser')  
    token       = soup.find('input', {'name': 'masked'}).get('value')
    SaveImage()
    captcha_text = SolveCapcha()
    post_data={"masked": token,
                'masked[username]': 'testuser',
                'masked[password]': 'testpass',
                'masked[captcha]': captcha_text,
                'masked':''}
    session.post('https://masked/', data=post_data,verify=False)