selenium 實現點觸驗證碼識別

YY點觸驗證碼識別案例

這個網站在第一次請求網站的時候，用selenium定位不到驗證碼，最后發現是iframe標簽的問題

我們需要再次請求iframe里面的鏈接，打開新的網頁之后就可以定位操作了。

 

# !/usr/bin/env python
# -*- coding:utf-8 -*-

from chaojiying import Chaojiying
import random
import time
from PIL import Image
from io import BytesIO
from selenium import webdriver
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.support import expected_conditions as ec

CHAOJIYING_NAME = '******' # 超級鷹賬號
CHAOJIYING_PWD = '********'  # 超級鷹密碼
CHAOJIYING_ID = 894613  # 軟件ID
CHAOJIYING_KIND = 9004  # 驗證碼類型


class YYVerification(object):
    """
    此類用於YY的驗證碼識別，可以應用到類似的驗證碼識別上，這種驗證碼類型是
    點擊類驗證碼.這里我們是對接超級鷹平台。
    """
    def __init__(self):
        self.url = 'https://aq.yy.com/p/reg/account.do?appid=&url=&fromadv=udbclsd_r'
        self.driver = webdriver.Chrome()
        self.chaojiying = Chaojiying(CHAOJIYING_NAME, CHAOJIYING_PWD, CHAOJIYING_ID)

    def __del__(self):
        self.driver.close()

    def screen_shot(self):
        """
        網頁截屏
        :return: bool值
        """
        # self.driver.maximize_window()
        time.sleep(2)
        # # 用於網頁的向下滾動
        # js = 'var q=document.documentElement.scrollTop=300'
        # self.driver.execute_script(js)
        # time.sleep(1)
        self.driver.save_screenshot('yy.png')
        return True

    def i_url(self):
        """
        實現對iframe標簽的url請求，並定位新網頁
        :return: 定位對象
        """
        i = self.driver.find_element_by_xpath('/html/body/div[2]/div[2]/div/iframe')
        url_1 = i.get_attribute('src')
        print(url_1)
        self.driver.get(url_1)
        wait = WebDriverWait(self.driver, 20)
        element = wait.until(ec.presence_of_element_located((By.XPATH, '//*[@id="mPickWords"]/div[1]')))
        return element

    def shear_location(self):
        """
        計算對截屏剪切的坐標，注意這里不同電腦可能會修改start_x到end_y
        后邊添加的數是根據作者的電腦優化的
        :return: 返回坐標以及網頁剪切定位的對象
        """
        time.sleep(random.random() + 1)
        print('正在獲取div')
        div = self.i_url()
        start_x = div.location['x'] + 25
        start_y = div.location['y'] + 65
        end_x = div.location['x'] + 350 + 25
        end_y = div.location['y'] + 218 + 65
        result = (start_x, start_y, end_x, end_y)
        print(result)
        return result, div

    @staticmethod
    def shear_image(axis):
        im = Image.open('yy.png')
        new_image = im.crop(axis)
        new_image.save('new_img.png')
        return new_image

    def upload_picture(self, img):
        """
        上傳圖片(Byte)，返回點擊的坐標。這里對返回的坐標進行了處理，
        處理過程可以看成 shear_location 函數的坐標數據逆處理過程
        :param img: 上傳的圖片
        :return: 點擊坐標
        """
        image = img
        byte_array = BytesIO()
        image.save(byte_array, format('PNG'))
        # 提交圖片
        result = self.chaojiying.post_pic(byte_array.getvalue(), CHAOJIYING_KIND)
        print(result)
        if '無可用題分' in result['err_str']:
            print('題分已經不足請充值')
            raise ValueError
        pic_str = result['pic_str']
        pic_list = [[i for i in x.split(',')] for x in pic_str.split('|')]
        for i in pic_list:
            i[0] = int(int(i[0])*(282/354))
            i[1] = int(int(i[1])*(219/274))
        print(pic_list)
        return pic_list

    def click_now(self, coordinates, axis, element):
        """
        這里如果驗證失敗會進行回調，實現驗證碼的自動驗證。
        :param coordinates: 點擊坐標
        :param axis: 剪切坐標
        :param element: 剪切位置的定位
        :return: 返回點擊的成功信息
        """
        print('點擊開始')
        print(coordinates)
        for location in coordinates:
            ActionChains(self.driver).move_to_element_with_offset(element, location[0], location[1]).click().perform()
            time.sleep(random.random() + 1.8)
        submission = self.driver.find_element_by_xpath('//*[@id="mPickWords"]/div[2]/span[4]')
        submission.click()
        time.sleep(1)
        if 'pw_submit_disabled' in submission.get_attribute('class'):
            return '點擊成功'
        else:
            time.sleep(2)
            print('正在重新識別新的驗證碼')
            self.revalidation(axis, element)
            return '點擊成功'

    def revalidation(self, axis, element):
        """
        網頁跳轉之后，執行驗證的主程序
        :param axis: 剪切坐標
        :param element: 剪切位置的定位對象
        :return: 驗證成果
        """
        self.screen_shot()
        if not self.screen_shot():
            return '截圖失敗'
        time.sleep(1)
        # 剪切圖片
        new_image = self.shear_image(axis)
        new_image.show()
        # 上傳圖片並返回點擊坐標
        click_coordinates = self.upload_picture(new_image)
        # 點擊驗證碼
        print(self.click_now(click_coordinates, axis, element))
        return '點擊驗證結束'

    def main(self):
        self.driver.get(self.url)
        wait = WebDriverWait(self.driver, 20)
        wait.until(ec.presence_of_element_located((By.XPATH, '/html/body/div[2]/div[2]/div/iframe')))
        axis, element = self.shear_location()
        self.revalidation(axis, element)


if __name__ == '__main__':
    yy = YYVerification()
    yy.main()