0

我正在编写一个 python 程序，该程序需要能够解决网站上的验证码。我想使用 2captcha。我已经使用 selenium 编写了一个 python 脚本，它将完成我需要做的所有事情，除了解决验证码。当我点击 2captcha.com 的“API”选项卡时，这个（连同其他参数）就是显示的内容：

您可以上传两种可用格式的验证码：

多部分和 Base64.:

多板示例：

<form method="post" action="http://2captcha.com/in.php" enctype="multipart/form-data">
<input type="hidden" name="method" value="post">
Your key:
<input type="text" name="key" value="YOUR_APIKEY">
The CAPTCHA file:
<input type="file" name="file">
<input type="submit" value="download and get the ID">
</form>

YOUR_APIKEY - 是您的 32 个符号长度的密钥。

Base64 示例：

<form method="post" action="http://2captcha.com/in.php">
<input type="hidden" name="method" value="base64">
Your key:
<input type="text" name="key" value="YOUR_APIKEY">
The CAPTCHA file body in base64 format:
<textarea name="body">BASE64_FILE</textarea>
<input type="submit" value="download and get the ID">
</form>

YOUR_APIKEY - 是您的 32 个符号长度的密钥。

BASE64_FILE - 是 base 64 编码的图像主体。

我很了解python，以及它的大部分科学和数学模块，但我对网络相关的编程有点陌生。上面的代码看起来像 html。我如何让 python 程序执行上面的 html 指令？

4

2 回答 2

5

我来自 2captcha 团队，我们有示例 python。您可以使用以下代码：

"""

这是如何通过 google recaptcha v2 的示例。我使用 python + selenium + phantomJS 来做到这一点。Phantom JS 是用于自动化 Web 测试的无头浏览器。我用它来报废页面并绕过谷歌验证码。要使用它，您应该安装 phantomJS（并在 linux 系统案例中从源代码构建）并安装 selenium python 模块

Google 验证码出现在 iframe 中的页面上。您应该废弃其元素，单击图像并在单击确定后检查错误消息。

"""

from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.common.exceptions import *
import re
import os
import sys
import time
import random
import requests
from PIL import Image

class capcha_resolver:
    def __init__(self, captchakey, proxy = None):
        """
       It is assumed that you have phantomjs installed into /bin folder on your linux system.
       """

        self.TWOCAPTCHA_API_KEY = captchakey
        phantom_args = []
        if proxy:
            self.PROXY = proxy
            phantom_args = ['--proxy='+self.PROXY, '--proxy-type=http', '--proxy-type=https']
        self.driver = webdriver.PhantomJS('/bin/phantomjs',service_args=phantom_args)
        self.driver.set_page_load_timeout(20)

    def fail(self, msg):
        print "[!] Error: " + msg
        self.driver.save_screenshot('error.png')

    def get_page(self):
        self.driver.get('https://www.google.com/recaptcha/api2/demo')
        self.driver.save_screenshot('page.png')
        return 0

    def send_capcha(self, filename):
        numbers = []
        captchafile = {'file': open(filename, 'rb')}
        data = {'key': self.TWOCAPTCHA_API_KEY, 'method': 'post'}
        r = requests.post('http://2captcha.com/in.php', files=captchafile, data=data)
        if r.ok and r.text.find('OK') > -1:
            reqid = r.text[r.text.find('|')+1:]
            print "[+] Capcha id: "+reqid
            for timeout in range(40):
                r = requests.get('http://2captcha.com/res.php?key={0}&action=get&id={1}'.format(self.TWOCAPTCHA_API_KEY, reqid))
                if r.text.find('CAPCHA_NOT_READY') > -1:
                    print r.text
                    time.sleep(3)
                if r.text.find('ERROR') > -1:
                    return []
                if r.text.find('OK') > -1:
                    return list(r.text[r.text.find('|')+1:])
        return []

    def bypass_captcha(self):
        """
       Google recaptcha could be found by id. Frame with checkbox has id which starts with I0, recapcha frame has id with I1
       """

        capthcaboxframe = self.driver.find_element_by_xpath('//iframe[starts-with(@id, "I0")]')
        self.driver.switch_to.frame(capthcaboxframe)
        time.sleep(1)
        checkbox = self.driver.find_element_by_id('recaptcha-anchor')
        checkbox.click()
        print "[*] Clicked on checkbox"
        time.sleep(2)
        self.driver.switch_to.default_content()

        capcthaframe = self.driver.find_element_by_xpath('//iframe[starts-with(@id, "I1")]')

        bounding_box = (
            capcthaframe.location['x'], # left
            capcthaframe.location['y'], # upper
            (capcthaframe.location['x'] + capcthaframe.size['width']), # right
            (capcthaframe.location['y'] + capcthaframe.size['height'])) # bottom
        imgname = 'capcha.jpeg' #use jpeg because png images can exceed 2capcha file size limit
        time.sleep(2)
        self.driver.save_screenshot(imgname)
        base_image = Image.open(imgname)
        cropped_image = base_image.crop(bounding_box)
        base_image = base_image.resize(cropped_image.size)
        base_image.paste(cropped_image, (0, 0))
        base_image.save(imgname)

        numbers = self.send_capcha(imgname)
        if numbers == []:
            return -1

        self.driver.switch_to.frame(capcthaframe)
        picturetable = self.driver.find_element_by_css_selector('.rc-imageselect-table-3')
        images = []
        for row in picturetable.find_elements_by_tag_name('tr'):
            for col in row.find_elements_by_tag_name('td'):
                images.append(col.find_element_by_tag_name('img'))
        if images == []:
            self.fail("Found no captcha images")
            return -1

        print "[*] Got answer : " + str(numbers)
        for number in numbers:
            index = int(number) - 1
            images[index].click()
            print '[+] clicked on image '+str(index)
        self.driver.save_screenshot('res.png')
        verifybutton = self.driver.find_element_by_id('recaptcha-verify-button')
        verifybutton.click()
        print "[*] Clicked verify button"
        time.sleep(2)
        if self.driver.find_element_by_css_selector('.rc-imageselect-incorrect-response').is_displayed() or \
                        self.driver.find_element_by_css_selector('.rc-imageselect-error-select-one').is_displayed() or \
                        self.driver.find_element_by_css_selector('.rc-imageselect-error-select-more').is_displayed():
            self.fail("Incorrect answer from 2captcha")
            return -1
        self.driver.switch_to.default_content()

        self.driver.switch_to.frame(capthcaboxframe)
        if self.driver.find_element_by_css_selector('.recaptcha-checkbox').get_attribute('aria-checked') == 'false':
            self.fail("Capctha not passed")
            return -1
        self.driver.switch_to.default_content()
        self.driver.save_screenshot('passed.png')
        return 0

proxy = None
if len(sys.argv) < 2:
    print "Usage: python resolver.py 2CAPCHA_API_KEY [PROXY]"
if len(sys.argv) > 2:
    proxy = sys.argv[2]
resolver = capcha_resolver(sys.argv[1], proxy)

if resolver.get_page() == -1:
    print "[!] Error while getting page"
else:
    print "[+] Opened URL"

if resolver.bypass_captcha() == -1:
    print "[!] Error on captcha resolving"
else:
    print "[+] Resolved captcha"

祝你好运！

于 2015-10-28T19:34:38.583 回答
1

查看请求模块

url = 'http://2captcha.com/in.php'
files = {'file': open('image.png', 'rb')}
data = {'key': 'key', 'method': 'post'}
r = requests.post(url, files=files, data=data)
if r.ok:
    # do something with the response data
于 2015-04-30T01:05:12.030 回答