Python3爬虫关于识别点触点选验证码的实例讲解

import requests
from hashlib import md5
class Chaojiying(object):
    def __init__(self, username, password, soft_id):
        self.username = username
        self.password = md5(password.encode('utf-8')).hexdigest()
        self.soft_id = soft_id
        self.base_params = {
            'user': self.username,
            'pass2': self.password,
            'softid': self.soft_id,
        }
        self.headers = {
            'Connection': 'Keep-Alive',
            'User-Agent': 'Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 5.1; Trident/4.0)',
        }
    def post_pic(self, im, codetype):
        """
        im: 图片字节
        codetype: 题目类型 参考 http://www.chaojiying.com/price.html
        """
        params = {
            'codetype': codetype,
        }
        params.update(self.base_params)
        files = {'userfile': ('ccc.jpg', im)}
        r = requests.post('http://upload.chaojiying.net/Upload/Processing.php', data=params, files=files, 
        headers=self.headers)
        return r.json()
    def report_error(self, im_id):
        """
        im_id:报错题目的图片ID
        """
        params = {
            'id': im_id,
        }
        params.update(self.base_params)
        r = requests.post('http://upload.chaojiying.net/Upload/ReportError.php', data=params, headers=self.headers)
        return r.json()

EMAIL = 'cqc@cuiqingcai.com'
PASSWORD = ''
# 超级鹰用户名、密码、软件ID、验证码类型
CHAOJIYING_USERNAME = 'Germey'
CHAOJIYING_PASSWORD = ''
CHAOJIYING_SOFT_ID = 893590
CHAOJIYING_KIND = 9102
class CrackTouClick():
    def __init__(self):
        self.url = 'http://admin.touclick.com/login.html'
        self.browser = webdriver.Chrome()
        self.wait = WebDriverWait(self.browser, 20)
        self.email = EMAIL
        self.password = PASSWORD
        self.chaojiying = Chaojiying(CHAOJIYING_USERNAME, CHAOJIYING_PASSWORD, CHAOJIYING_SOFT_ID)

def open(self):
    """
    打开网页输入用户名密码
    :return: None
    """
    self.browser.get(self.url)
    email = self.wait.until(EC.presence_of_element_located((By.ID, 'email')))
    password = self.wait.until(EC.presence_of_element_located((By.ID, 'password')))
    email.send_keys(self.email)
    password.send_keys(self.password)
def get_touclick_button(self):
    """
    获取初始验证按钮
    :return:
    """
    button = self.wait.until(EC.element_to_be_clickable((By.CLASS_NAME, 'touclick-hod-wrap')))
    return button

def get_touclick_element(self):
    """
    获取验证图片对象
    :return: 图片对象
    """
    element = self.wait.until(EC.presence_of_element_located((By.CLASS_NAME, 'touclick-pub-content')))
    return element
def get_position(self):
    """
    获取验证码位置
    :return: 验证码位置元组
    """
    element = self.get_touclick_element()
    time.sleep(2)
    location = element.location
    size = element.size
    top, bottom, left, right = location['y'], location['y'] + size['height'], location['x'], location['x'] + size[
        'width']
    return (top, bottom, left, right)
def get_screenshot(self):
    """
    获取网页截图
    :return: 截图对象
    """
    screenshot = self.browser.get_screenshot_as_png()
    screenshot = Image.open(BytesIO(screenshot))
    return screenshot
def get_touclick_image(self, name='captcha.png'):
    """
    获取验证码图片
    :return: 图片对象
    """
    top, bottom, left, right = self.get_position()
    print('验证码位置', top, bottom, left, right)
    screenshot = self.get_screenshot()
    captcha = screenshot.crop((left, top, right, bottom))
    return captcha

image = self.get_touclick_image()
bytes_array = BytesIO()
image.save(bytes_array, format='PNG')
# 识别验证码
result = self.chaojiying.post_pic(bytes_array.getvalue(), CHAOJIYING_KIND)
print(result)

{'err_no': 0, 'err_str': 'OK', 'pic_id': '6002001380949200001', 'pic_str': '132,127|56,77', 'md5': '1f8e1d4bef8b
11484cb1f1f34299865b'}

def get_points(self, captcha_result):
    """
    解析识别结果
    :param captcha_result: 识别结果
    :return: 转化后的结果
    """
    groups = captcha_result.get('pic_str').split('|')
    locations = [[int(number) for number in group.split(',')] for group in groups]
    return locations
 
def touch_click_words(self, locations):
    """
    点击验证图片
    :param locations: 点击位置
    :return: None
    """
    for location in locations:
        print(location)
        ActionChains(self.browser).move_to_element_with_offset(self.get_touclick_element(), location[0], 
        location[1]).click().perform()
        time.sleep(1)