主要用到技术：
selenium
百度AI文字识别
滑动验证码的先加速后减速设计
# -*- coding: utf-8 -*-# @AuThor : frank_leefrom selenium import webdriverfrom selenium.webdriver.support.wait import WebDriverWaitimport randomfrom selenium.webdriver import ActionChainsimport timefrom aip import AipOcrimport base64class Pojie: def __init__(self): self.url = 'https://www.52pojie.cn/member.php?mod=logging&action=login' self.driver = webdriver.Chrome() self.wait = WebDriverWait(self.driver, 30) # 设置超时时间 self.zoom = 1 def open(self): self.driver.get(self.url) self.driver.maximize_window() def recognize_image(self): captcha_image = self.driver.find_element_by_xpath('//div[@class="imgCaptcha_img"]/img') captcha = captcha_image.get_attribute('src') # 下载图片 fh = open("captcha.jpg", "wb") fh.write(base64.b64decode(captcha.split(',')[1])) fh.close() APP_ID = '你的 App ID' API_KEY = '你的 Api Key' SECRET_KEY = '你的 Secret Key' with open('./captcha.jpg', 'rb') as bin_data: image_data = bin_data.read() options = {} options["detect_direction"] = "true" options["detect_language"] = "false" options["probability"] = "true" client = AipOcr(APP_ID, API_KEY, SECRET_KEY) result = client.basicAccurate(image_data, options) print(result['words_result'][0]['words']) return result['words_result'][0]['words'].strip() def input_captcha(self, captcha): self.driver.find_element_by_xpath('//input[@id="nc_1_captcha_input"]').send_keys(captcha) time.sleep(5) self.driver.find_element_by_xpath('// *[ @ id = "nc_1_scale_submit"] / span').click() time.sleep(10) if '验证通过' in self.driver.page_source: self.driver.find_element_by_xpath('//*[@name="loginsubmit"]').click() time.sleep(10) print("你已成功登录，可以开始你的骚操作了") if '_errorTEXT' in self.driver.page_source: self.driver.find_element_by_xpath('//*[@id="nc_1__btn_1"]').click() time.sleep(5) secondcap = self.recognize_image() self.input_captcha(secondcap) time.sleep(10) elif '_errorTooMuch' in self.driver.page_source: self.driver.find_element_by_xpath('//*[@id="nc_1__btn_1"]').click() time.sleep(10) secondcap = self.recognize_image() self.input_captcha(secondcap) time.sleep(10) def get_tracks(self, distance): # 初始速度 v = 0 # 0.2秒到0.3s之间随机生成的浮点数来统计轨迹，轨迹即0.2-0.3s内某个时间的位移 t = random.uniform(0.2, 0.3) forward_tracks = [] # 当前位移 current = 0 # 到达mid值开始减速 mid = distance * 3 / 5 while current < distance: if current < mid: # 加速度越小，单位时间的位移越小，模拟的轨迹就越多越详细 a = 2 else: # 先加速后减速，减速时的加速度 a = -3 # 初速度 v0 = v # 0.2-0.3秒时间内某个时间的位移 s = v0 * t 0.5 * a * (t ** 2) # 当前的位置 current = s # 添加到轨迹列表,round()为保留一位小数且该小数要进行四舍五入 forward_tracks.append(round(s)) # 速度已经达到v，该速度作为下次的初速度 v = v0 a * t return {'forward_tracks': forward_tracks} def pojie_slider(self): try: distance = 300 tracks = self.get_tracks(distance) # 对位移的缩放计算 self.open() self.driver.find_element_by_xpath('//input[@name="username"]').send_keys('你的吾爱无限制论坛账号') time.sleep(1) self.driver.find_element_by_xpath('//input[@name="password"]').send_keys('你的论坛密码') time.sleep(2) slider = self.driver.find_element_by_xpath("//*[@id='nc_1_n1z']") time.sleep(1) ActionChains(self.driver).click_and_hold(slider).perform() for track in tracks['forward_tracks']: ActionChains(self.driver).move_by_offset(xoffset=track, yoffset=0).perform() time.sleep(1) except Exception as e: print(e) exit(0)if __name__ == '__main__': p = Pojie() p.pojie_slider() captcha = p.recognize_image() p.input_captcha(captcha)
评论