|
| 1 | +import os |
| 2 | +import time |
| 3 | +import json |
| 4 | +import re |
| 5 | +from selenium import webdriver |
| 6 | +from selenium.webdriver.support.wait import WebDriverWait |
| 7 | +from selenium.webdriver.common.by import By |
| 8 | +from selenium.webdriver.support import expected_conditions as EC |
| 9 | +from selenium.webdriver.chrome.service import Service |
| 10 | +from selenium.webdriver.chrome.options import Options |
| 11 | +from twocaptcha import TwoCaptcha |
| 12 | + |
| 13 | + |
| 14 | + |
| 15 | +# CONFIGURATION |
| 16 | + |
| 17 | +url = "https://2captcha.com/demo/cloudflare-turnstile-challenge" |
| 18 | +apikey = os.getenv('APIKEY_2CAPTCHA') |
| 19 | + |
| 20 | + |
| 21 | +""" |
| 22 | +When a web page first loads, some JavaScript functions and objects (such as window.turnstile) may already be initialized |
| 23 | +and executed. If the interception script is launched too late, this may lead to the fact that the necessary parameters |
| 24 | +will already be lost, or the script simply will not have time to intercept the right moment. Refreshing the page ensures |
| 25 | +that everything starts from scratch and you trigger the interception at the right time. |
| 26 | +""" |
| 27 | +intercept_script = """ |
| 28 | + console.clear = () => console.log('Console was cleared') |
| 29 | + const i = setInterval(()=>{ |
| 30 | + if (window.turnstile) |
| 31 | + console.log('success!!') |
| 32 | + {clearInterval(i) |
| 33 | + window.turnstile.render = (a,b) => { |
| 34 | + let params = { |
| 35 | + sitekey: b.sitekey, |
| 36 | + pageurl: window.location.href, |
| 37 | + data: b.cData, |
| 38 | + pagedata: b.chlPageData, |
| 39 | + action: b.action, |
| 40 | + userAgent: navigator.userAgent, |
| 41 | + } |
| 42 | + console.log('intercepted-params:' + JSON.stringify(params)) |
| 43 | + window.cfCallback = b.callback |
| 44 | + return } |
| 45 | + } |
| 46 | +},50) |
| 47 | +""" |
| 48 | + |
| 49 | +# LOCATORS |
| 50 | + |
| 51 | +success_message_locator = "//p[contains(@class,'successMessage')]" |
| 52 | + |
| 53 | + |
| 54 | +# GETTERS |
| 55 | + |
| 56 | +def get_element(locator): |
| 57 | + """Waits for an element to be clickable and returns it""" |
| 58 | + return WebDriverWait(browser, 30).until(EC.element_to_be_clickable((By.XPATH, locator))) |
| 59 | + |
| 60 | + |
| 61 | +# ACTIONS |
| 62 | + |
| 63 | +def get_captcha_params(script): |
| 64 | + """ |
| 65 | + Refreshes the page, injects a JavaScript script to intercept Turnstile parameters, and retrieves them. |
| 66 | +
|
| 67 | + Args: |
| 68 | + script (str): The JavaScript code to be injected. |
| 69 | +
|
| 70 | + Returns: |
| 71 | + dict: The intercepted Turnstile parameters as a dictionary. |
| 72 | + """ |
| 73 | + browser.refresh() # Refresh the page to ensure the script is applied correctly |
| 74 | + |
| 75 | + browser.execute_script(script) # Inject the interception script |
| 76 | + |
| 77 | + time.sleep(5) # Allow some time for the script to execute |
| 78 | + |
| 79 | + logs = browser.get_log("browser") # Retrieve the browser logs |
| 80 | + params = None |
| 81 | + for log in logs: |
| 82 | + if "intercepted-params:" in log['message']: |
| 83 | + log_entry = log['message'].encode('utf-8').decode('unicode_escape') |
| 84 | + match = re.search(r'intercepted-params:({.*?})', log_entry) |
| 85 | + if match: |
| 86 | + json_string = match.group(1) |
| 87 | + params = json.loads(json_string) |
| 88 | + break |
| 89 | + print("Parameters received") |
| 90 | + return params |
| 91 | + |
| 92 | +def solver_captcha(apikey, params): |
| 93 | + """ |
| 94 | + Solves the Turnstile captcha using the 2Captcha service. |
| 95 | +
|
| 96 | + Args: |
| 97 | + apikey (str): The 2Captcha API key. |
| 98 | + params (dict): The intercepted Turnstile parameters. |
| 99 | +
|
| 100 | + Returns: |
| 101 | + str: The solved captcha token. |
| 102 | + """ |
| 103 | + solver = TwoCaptcha(apikey) |
| 104 | + try: |
| 105 | + result = solver.turnstile(sitekey=params["sitekey"], |
| 106 | + url=params["pageurl"], |
| 107 | + action=params["action"], |
| 108 | + data=params["data"], |
| 109 | + pagedata=params["pagedata"], |
| 110 | + useragent=params["userAgent"]) |
| 111 | + print(f"Captcha solved") |
| 112 | + return result['code'] |
| 113 | + except Exception as e: |
| 114 | + print(f"An error occurred: {e}") |
| 115 | + return None |
| 116 | + |
| 117 | +def send_token_callback(token): |
| 118 | + """ |
| 119 | + Executes the callback function with the given token. |
| 120 | +
|
| 121 | + Args: |
| 122 | + token (str): The solved captcha token. |
| 123 | + """ |
| 124 | + script = f"cfCallback('{token}')" |
| 125 | + browser.execute_script(script) |
| 126 | + print("The token is sent to the callback function") |
| 127 | + |
| 128 | +def final_message(locator): |
| 129 | + """ |
| 130 | + Retrieves and prints the final success message. |
| 131 | +
|
| 132 | + Args: |
| 133 | + locator (str): The XPath locator of the success message. |
| 134 | + """ |
| 135 | + message = get_element(locator).text |
| 136 | + print(message) |
| 137 | + |
| 138 | + |
| 139 | +# MAIN LOGIC |
| 140 | + |
| 141 | +chrome_options = Options() |
| 142 | +chrome_options.add_argument("user-agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) " |
| 143 | + "Chrome/126.0.0.0 Safari/537.36") |
| 144 | +# Set logging preferences to capture only console logs |
| 145 | +chrome_options.set_capability("goog:loggingPrefs", {"browser": "INFO"}) |
| 146 | + |
| 147 | +with webdriver.Chrome(service=Service(), options=chrome_options) as browser: |
| 148 | + browser.get(url) |
| 149 | + print("Started") |
| 150 | + |
| 151 | + params = get_captcha_params(intercept_script) |
| 152 | + |
| 153 | + if params: |
| 154 | + token = solver_captcha(apikey, params) |
| 155 | + |
| 156 | + if token: |
| 157 | + send_token_callback(token) |
| 158 | + final_message(success_message_locator) |
| 159 | + time.sleep(5) |
| 160 | + print("Finished") |
| 161 | + else: |
| 162 | + print("Failed to solve captcha") |
| 163 | + else: |
| 164 | + print("Failed to intercept parameters") |
0 commit comments