携程滑块,点选文字识别

文章目录

code_datawhale05_xiecheng.py
reslut
code_datawhale06_bzhan_login.py
result
END

code_datawhale05_xiecheng.py

携程滑块,点选文字识别

from selenium import webdriver
from selenium.webdriver import ActionChains
import time
import base64
from PIL import Image
from aip import AipOcr"""
https://blog.csdn.net/weixin_41311624/article/details/86522119
利用Python3安装aip
https://www.jianshu.com/p/d59b47bc4812
利用Python3安装aiphttps://blog.csdn.net/qq_38787214/article/details/87902291
Python3 super().__init__()测试及理解https://www.cnblogs.com/duanwandao/p/9802795.htmlhttps://blog.csdn.net/pythoncsdn111/article/details/96453839
https://blog.csdn.net/qq_42992919/article/details/98483845https://blog.csdn.net/hhy1107786871/article/details/88342976"""# 破解携程反爬验证
class unlockScrapy(object):# super().__init__()的作用也就显而易见了，就是执行父类的构造函数，使得我们能够调用父类的属性。def __init__(self, driver):super(unlockScrapy, self).__init__()# selenium驱动self.driver = driver# self.WAPPID = '百度文字识别appid'# self.WAPPKEY = '百度文字识别appkey'# self.WSECRETKEY = '百度文字识别secretkey'# 百度文字识别sdk客户端# self.WCLIENT = AipOcr(self.WAPPID, self.WAPPKEY, self.WSECRETKEY)self.WAPPID = '17062614'self.WAPPKEY = 'E15mYUgfBRVV3ohVVZZVcCCc'self.WSECRETKEY = 'ClxgLmf2U0DwgX9mSvZG7v4zInrrCT92'# 百度文字识别sdk客户端self.WCLIENT = AipOcr(self.WAPPID, self.WAPPKEY, self.WSECRETKEY)print("5" * 10)## 切换二维码登录,在切换回来,就会滑动出现## 滑动出现后,输错一次密码,再登录,就会出现文字顺序验证码# 破解滑动##  cpt - img - double - right - outerdef unlockScroll(self):try:# 滑块elementprint("1" * 10)scrollElement = self.driver.find_elements_by_class_name('cpt-img-double-right-outer')[0]print("2" * 10)ActionChains(self.driver).click_and_hold(on_element=scrollElement).perform()ActionChains(self.driver).move_to_element_with_offset(to_element=scrollElement, xoffset=30, yoffset=10).perform()ActionChains(self.driver).move_to_element_with_offset(to_element=scrollElement, xoffset=100, yoffset=20).perform()ActionChains(self.driver).move_to_element_with_offset(to_element=scrollElement, xoffset=200, yoffset=50).perform()print("滑块破解成功")except:print("无滑块")# 下载上面的小图和下面的大图def downloadImg(self):# 小图的src"""//*[@id="sliderddnormal-choose"]/div[2]/div[1]/img"""# "/html/body/div[3]/div[1]/img"time.sleep(1)codeSrc = self.driver.find_element_by_xpath("//*[@id='sliderddnormal-choose']/div[2]/div[1]/img").get_attribute("src")print(codeSrc)print("6" * 10)# 大图的src# "/html/body/div[3]/div[3]/img"checkSrc = self.driver.find_element_by_xpath("//*[@id='sliderddnormal-choose']/div[2]/div[3]/img").get_attribute("src")print("7" * 10)print(codeSrc.split(','))"""https://www.cnblogs.com/wswang/p/7717997.htmlPython解码base64遇到Incorrect padding错误"""# 保存下载# 由于其src是base64编码的，因此需要以base64编码形式写入,# 由于标准的Base64编码后可能出现字符+和/，在URL中就不能直接作为参数，所以又有一种"url safe"的base64编码# base64.urlsafe_b64decode(base64_url)# fh.write(base64.b64decode(codeSrc.split(',')[1]))fh = open("code.jpeg", "wb")fh.write(base64.urlsafe_b64decode(codeSrc.split(',')[1]))fh.close()fh = open("checkCode.jpeg", "wb")fh.write(base64.urlsafe_b64decode(checkSrc.split(',')[1]))fh.close()"""https://www.cnblogs.com/kongzhagen/p/6295925.html7. 点操作:im.point(function) #,这个function接受一个参数，且对图片中的每一个点执行这个函数比如：out=im.point(lambdai:i*1.5)#对每个点进行50%的加强"""# 图片二值化，便于识别其中的文字def chageImgLight(self):im = Image.open("code.jpeg")im1 = im.point(lambda p: p * 4)im1.save("code.jpeg")im = Image.open("checkCode.jpeg")im1 = im.point(lambda p: p * 4)im1.save("checkCode.jpeg")# 读取图片文件def getFile(self, filePath):with open(filePath, 'rb') as fp:print("8读取图片" * 2)return fp.read()"""# 请求参数language_type :  识别语言类型，默认为CHN_ENG中英文混合；。可选值包括：detect_direction :是否检测图像朝向，默认不检测, ture 是检测# 返回参数words_result"""# 识别上面小图中的文字def iTow(self):try:print("开始识别小图...")op = {'language_type': 'CHN_ENG', 'detect_direction': 'true'}res = self.WCLIENT.basicAccurate(self.getFile('code.jpeg'), options=op)  # options 可选参数words = ''print("9" * 10)# http://ai.baidu.com/docs#/OCR-Python-SDK/80d64770print(res['words_result'])  # api已经定好的  array    定位和识别结果数组print("10" * 10)for item in res['words_result']:if item['words'].endswith('。'):words = words + item['words'] + '\r\n'else:words = words + item['words']print('小图中的文字: ' + words)print("小图文字识别完成")return wordsexcept:return 'error'"""# 请求参数recognize_granularity:是否定位单字符位置，big：不定位单字符位置，默认值；small：定位单字符位置item['chars'] :+chars  array   单字符结果，recognize_granularity=small时存在"""# 识别下面大图中的文字及坐标def getPos(self, words):try:print("开始识别大图...")op = {'language_type': 'CHN_ENG', 'recognize_granularity': 'small'}res = self.WCLIENT.accurate(self.getFile('checkCode.jpeg'), options=op)# 所有文字的位置信息allPosInfo = []# 需要的文字的位置信息needPosInfo = []print("#1" * 10)# 每日50000次,超时报错{'error_code': 17, 'error_msg': 'Open api daily request limit reached'}print(res)print(res['words_result'])print("#2" * 10)print("11" * 10)for item in res['words_result']:allPosInfo.extend(item['chars'])print(item['chars'])  # 文字及位置信息,见百度apiprint("12" * 10)# 筛选出需要的文字的位置信息for word in words:for item in allPosInfo:if word == item['char']:needPosInfo.append(item)time.sleep(1)print('大图中的文字: ' + item['char'])# 返回出现文字的位置信息print(needPosInfo)print("13" * 10)print("大图识别完成...")return needPosInfoexcept Exception as e:print(e)"""https://blog.csdn.net/huilan_same/article/details/52305176ActionChains: 模拟鼠标操作比如单击、双击、点击鼠标右键、拖拽等等selenium之 玩转鼠标键盘操作（ActionChains）https://blog.csdn.net/ccggaag/article/details/75717186web自动化测试第6步：模拟鼠标操作（ActionChains）"""# 点击大图上的文字def clickWords(self, wordsPosInfo):# 获取到大图的element#  /html/body/div[3]/div[3]/imgimgElement = self.driver.find_element_by_xpath('//*[@id="sliderddnormal-choose"]/div[2]/div[3]/img')# 根据上图文字在下图中的顺序依次点击下图中的文字for info in wordsPosInfo:# move_to_element_with_offset(to_element, xoffset, yoffset) ——移动到距某个元素（左上角坐标）多少距离的位置ActionChains(self.driver).move_to_element_with_offset(to_element=imgElement, xoffset=info['location']['left'] + 20,yoffset=info['location']['top'] + 20).click().perform()time.sleep(1)# 若出现点击图片,则破解def pic_main(self):try:##  先下载图片time.sleep(1)self.downloadImg()print("14-0" * 10)## 图片二值化,方便识别self.chageImgLight()## 读取图片(调用百度ocr),识别小图文字text = self.iTow()## 读取图片(调用百度ocr)识别大图文字及位置信息posInfo = self.getPos(text)## 点击提交按钮 ,在点击之前确认一下,大图与小图数字是否完全相等,若不相等,则重新识别print(type(text))print(type(posInfo))print(len(text))print(len(posInfo))print("14" * 10)### 提交之前先判断一下,大小图字数是否一致,若不等,重新生成图片,重新识别while len(text) != len(posInfo) or posInfo is None:## 刷新图片# /html/body/div[3]/div[4]/div/aself.driver.find_elements_by_xpath('//*[@id="sliderddnormal-choose"]/div[2]/div[4]/div/a')[0].click()time.sleep(2)## 下载图片self.downloadImg()print("14-1" * 10)## 图片二值化,方便识别self.chageImgLight()## 识别小图文字text = self.iTow()## 识别大图文字及位置信息posInfo = self.getPos(text)print('匹配成功，开始点击')##  按顺序模拟点击self.clickWords(posInfo)## 点选文字后提交self.driver.find_elements_by_xpath('//*[@id="sliderddnormal-choose"]/div[2]/div[4]/a')[0].click()print("模拟点击完成,已提交...点选图片破解成功...")except:print("无点选文字点击图片")# 破解滑动,点选文字图片
def unlock(name,pwd):# 创建浏览器对象driver = webdriver.Chrome()# 打开Chrome浏览器，需要将Chrome的驱动放在当前文件夹,也可以房子啊google.exe同目录下,需设置到环境变量# login_url = 'https://hotels.ctrip.com/hotel/6278770.html#ctm_ref=hod_hp_hot_dl_n_2_7'#             "https://passport.ctrip.com/user/login?BackUrl=https%3A%2F%2Fhotels.ctrip.com%2Fhotel%2F6278770.html%23ctm_ref%3Dhod_hp_hot_dl_n_2_7 "# 登录页面login_url = "https://passport.ctrip.com/user/login?BackUrl=https%3A%2F%2Fhotels.ctrip.com%2Fhotel%2F6278770.html%23ctm_ref%3Dhod_hp_hot_dl_n_2_7"driver.get(login_url)driver.maximize_window() # 全屏网页窗口time.sleep(3)# 切换账号密码登录表单# js1 = 'document.querySelector("#j_loginTab1").style.display="none";'# browser.execute_script(js1)# time.sleep(1)# js2 = 'document.querySelector("#j_loginTab2").style.display="block";'# browser.execute_script(js2)# driver.find_element_by_id('lbNormal').click()# time.sleep(3)#  输入账号密码input_name = driver.find_element_by_id('nloginname')input_name.clear()input_name.send_keys(name)input_pass = driver.find_element_by_id('npwd')input_pass.clear()input_pass.send_keys(pwd)time.sleep(3)# 此时可能出现有滑动验证码与点选文字##  若出现滑块,则开始破解滑块unlock = unlockScrapy(driver)unlock.unlockScroll()## 若出现点选文字,开始破解点选文字unlock.pic_main()# 点击登录print("3" * 10)"""//*[@id="nsubmit"]"""# browser.find_element_by_xpath('//*[@class="form__button"]/button').click()driver.find_element_by_xpath('//*[@class="form_btn form_btn--block"]').click()time.sleep(19)# 如果破解成功，html的title会变if unlock.driver.title != '携程在手，说走就走':print('破解成功')else:# 再次尝试print('破解失败，再次破解')unlock.pic_main()# 再次点击登录print("3" * 10)"""//*[@id="nsubmit"]"""# browser.find_element_by_xpath('//*[@class="form__button"]/button').click()driver.find_element_by_xpath('//*[@class="form_btn form_btn--block"]').click()time.sleep(19)time.sleep(9)if __name__ == '__main__':name = str(input("请输入账号:"))pwd = str(input('请输入密码:'))for i in range(2):unlock(name,pwd)

reslut

D:\ProgramData\Anaconda3\python.exe D:/py1805/badou/share_folder/datawhale_spider/datawhale05_xiecheng.py
请输入账号:66***
请输入密码:88***
5555555555
1111111111
2222222222
滑块破解成功
data:image/jpg;base64,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
6666666666
7777777777
['data:image/jpg;base64', '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']
14-014-014-014-014-014-014-014-014-014-0
开始识别小图...
8读取图片8读取图片
9999999999
[{'words': '青城山'}]
10101010101010101010
小图中的文字: 青城山
小图文字识别完成
开始识别大图...
8读取图片8读取图片
#1#1#1#1#1#1#1#1#1#1
{'log_id': 2128196181742769400, 'words_result_num': 2, 'words_result': [{'chars': [{'char': '城', 'location': {'width': 18, 'top': 134, 'left': 62, 'height': 30}}], 'location': {'width': 28, 'top': 134, 'left': 62, 'height': 30}, 'words': '城'}, {'chars': [{'char': '“', 'location': {'width': 47, 'top': 72, 'left': 138, 'height': 42}}], 'location': {'width': 152, 'top': 0, 'left': 82, 'height': 184}, 'words': '“'}]}
[{'chars': [{'char': '城', 'location': {'width': 18, 'top': 134, 'left': 62, 'height': 30}}], 'location': {'width': 28, 'top': 134, 'left': 62, 'height': 30}, 'words': '城'}, {'chars': [{'char': '“', 'location': {'width': 47, 'top': 72, 'left': 138, 'height': 42}}], 'location': {'width': 152, 'top': 0, 'left': 82, 'height': 184}, 'words': '“'}]
#2#2#2#2#2#2#2#2#2#2
11111111111111111111
[{'char': '城', 'location': {'width': 18, 'top': 134, 'left': 62, 'height': 30}}]
12121212121212121212
[{'char': '“', 'location': {'width': 47, 'top': 72, 'left': 138, 'height': 42}}]
12121212121212121212
大图中的文字: 城
[{'char': '城', 'location': {'width': 18, 'top': 134, 'left': 62, 'height': 30}}]
13131313131313131313
大图识别完成...
<class 'str'>
<class 'list'>
3
1
14141414141414141414
data:image/jpg;base64,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
6666666666
7777777777
['data:image/jpg;base64', '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']
14-114-114-114-114-114-114-114-114-114-1
开始识别小图...
8读取图片8读取图片
9999999999
[{'words': '六和塔'}]
10101010101010101010
小图中的文字: 六和塔
小图文字识别完成
开始识别大图...
8读取图片8读取图片
#1#1#1#1#1#1#1#1#1#1
{'log_id': 3303545943648964856, 'words_result_num': 4, 'words_result': [{'chars': [{'char': '水', 'location': {'width': 28, 'top': 14, 'left': 134, 'height': 17}}, {'char': '六', 'location': {'width': 28, 'top': 53, 'left': 134, 'height': 17}}], 'location': {'width': 28, 'top': 14, 'left': 134, 'height': 69}, 'words': '水六'}, {'chars': [{'char': '手', 'location': {'width': 21, 'top': 18, 'left': 199, 'height': 37}}], 'location': {'width': 32, 'top': 18, 'left': 188, 'height': 37}, 'words': '手'}, {'chars': [{'char': '塔', 'location': {'width': 38, 'top': 69, 'left': 17, 'height': 20}}, {'char': '和', 'location': {'width': 38, 'top': 115, 'left': 17, 'height': 20}}], 'location': {'width': 38, 'top': 69, 'left': 17, 'height': 82}, 'words': '塔和'}, {'chars': [{'char': '追', 'location': {'width': 20, 'top': 114, 'left': 239, 'height': 33}}], 'location': {'width': 29, 'top': 114, 'left': 239, 'height': 33}, 'words': '追'}]}
[{'chars': [{'char': '水', 'location': {'width': 28, 'top': 14, 'left': 134, 'height': 17}}, {'char': '六', 'location': {'width': 28, 'top': 53, 'left': 134, 'height': 17}}], 'location': {'width': 28, 'top': 14, 'left': 134, 'height': 69}, 'words': '水六'}, {'chars': [{'char': '手', 'location': {'width': 21, 'top': 18, 'left': 199, 'height': 37}}], 'location': {'width': 32, 'top': 18, 'left': 188, 'height': 37}, 'words': '手'}, {'chars': [{'char': '塔', 'location': {'width': 38, 'top': 69, 'left': 17, 'height': 20}}, {'char': '和', 'location': {'width': 38, 'top': 115, 'left': 17, 'height': 20}}], 'location': {'width': 38, 'top': 69, 'left': 17, 'height': 82}, 'words': '塔和'}, {'chars': [{'char': '追', 'location': {'width': 20, 'top': 114, 'left': 239, 'height': 33}}], 'location': {'width': 29, 'top': 114, 'left': 239, 'height': 33}, 'words': '追'}]
#2#2#2#2#2#2#2#2#2#2
11111111111111111111
[{'char': '水', 'location': {'width': 28, 'top': 14, 'left': 134, 'height': 17}}, {'char': '六', 'location': {'width': 28, 'top': 53, 'left': 134, 'height': 17}}]
12121212121212121212
[{'char': '手', 'location': {'width': 21, 'top': 18, 'left': 199, 'height': 37}}]
12121212121212121212
[{'char': '塔', 'location': {'width': 38, 'top': 69, 'left': 17, 'height': 20}}, {'char': '和', 'location': {'width': 38, 'top': 115, 'left': 17, 'height': 20}}]
12121212121212121212
[{'char': '追', 'location': {'width': 20, 'top': 114, 'left': 239, 'height': 33}}]
12121212121212121212
大图中的文字: 六
大图中的文字: 和
大图中的文字: 塔
[{'char': '六', 'location': {'width': 28, 'top': 53, 'left': 134, 'height': 17}}, {'char': '和', 'location': {'width': 38, 'top': 115, 'left': 17, 'height': 20}}, {'char': '塔', 'location': {'width': 38, 'top': 69, 'left': 17, 'height': 20}}]
13131313131313131313
大图识别完成...
匹配成功，开始点击
模拟点击完成,已提交...点选图片破解成功...
3333333333

code_datawhale06_bzhan_login.py

# encoding:utf-8
import randomfrom selenium import webdriver
from selenium.webdriver import ActionChains
from selenium.webdriver.common.by import By
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.support import expected_conditions as EC
from selenium.webdriver.support.wait import WebDriverWait
from PIL import Image
import time"""
https://www.cnblogs.com/xumBlog/p/10992721.html
Selenium常见元素定位方法和操作的学习介绍https://zhidao.baidu.com/question/1952564842626224588.html
python+selenium使用location定位元素坐标偏差https://www.cppentry.com/bencandy.php?fid=77&id=195925
selenium登录 京东滑动验证码(一)"""# 获得大,小缺口图片
def get_image(driver, n, canvas):# canvas = driver.find_element_by_xpath('/html/body/div[3]/div[2]/div[2]/div[1]/div[1]/div/a/div[1]/div/canvas[2]')#  /html/body/div[2]/div[2]/div[6]/div/div[1]/div[1]/div/a/div[1]/div/canvas[2]# /html/body/div[2]/div[2]/div[6]/div/div[1]/div[1]/div/a/div[1]/canvas# <canvas class="geetest_canvas_fullbg geetest_fade geetest_absolute" height="160" width="260" style="opacity: 1; display: bolck;"></canvas># 获取缺口图片元素print(canvas)print("2" * 10)# 缺口图片元素的左上角位置信息,location定位元素坐标# #图片坐标。（注：图片坐标为该标签相对整个页面的坐标位置）即截图的起始点left = canvas.location['x']top = canvas.location['y']print(left)print (top)print(type(canvas))print("3" * 10)# 缺口图片,x,y 的右上角位置信息# canvas.size 标签图片的大小# 计算出需要截图的长宽、高度等elementWidth = canvas.location['x']  + canvas.size['width']elementHeight = canvas.location['y']  + canvas.size['height']# 保存缺口图片driver.save_screenshot(n + '.png')picture = Image.open(n + '.png')# 截图crop(区域由一个4元组定义，表示为坐标是 (left, upper, right, lower)。# Python Imaging Library 使用左上角为 (0, 0)的坐标系统。# https://zhidao.baidu.com/question/1689969007353725628.html# 因为自己的显示器分辨率较高,故放大了1.25倍,同理截图的元素也需要放大1.25倍picture = picture.crop((left * 1.25, top * 1.25, elementWidth * 1.25, elementHeight * 1.25))picture.save('photo' + n + '.png')# 打开图片,发现图片截图位置不对,因为电脑系统显示缩放了,return picture"""
https://blog.csdn.net/tashanhongye/article/details/49869487
imagesize()函数获取图片信息
"""
# 比较两张图片的异同处,返回差异出x方向的坐标
def get_space(picture1, picture2):print("# 比较两站图片的异同...")start = 60threhold = 60 # 临界值for i in range(start, picture1.size[0]): # 索引 0 给出的是图像宽度的像素值for j in range(picture1.size[1]):    # 索引 1 给出的是图像高度的像素值rgb1 = picture1.load()[i, j]      # 加载二值化后的图片数据为rgbrgb2 = picture2.load()[i, j]res1 = abs(rgb1[0] - rgb2[0])     # abs取绝对值,连个像素值差异值res2 = abs(rgb1[1] - rgb2[1])res3 = abs(rgb1[2] - rgb2[2])# 若差异值大于临界值,就是缺口值,rentun i  ,,只返回i 即可,因为只在x 方向滑动if not (res1 < threhold and res2 < threhold and res3 < threhold):return i# 小滑块实际未到边,故减去6,才是实际距离,6为用可以显示图片的软件测量得到(beyound compare)return i - 6# 模拟人工滑动
def get_tracks(space):print("#3" * 10)print(space)print("#4" * 10)# 模拟人工滑动，避免被识别为机器 # 匀加速运动公式s = Vo X t + 1/2 X a X t^2# space += 7  # 先滑过一点，最后再反着滑动回来v = 0t = 0.2current = 0s = 0forward_tracks = []  # 前进的轨道mid = space * 4 / 5  # 前3/5加速,后面减速while abs(current - space) > 1:if current < mid:  # 加速a = 2else:a = -3      # 减速v = v + a * t  # 当前实时速度s = v * t + 0.5 * a * (t ** 2) # 此时实时路径current += s   # 当前滑动的距离forward_tracks.append(s)print("#8" * 10)print(current)print(s)print("#9" * 10)# 每次滑动的距离添加到forward_tracks列表中,小数太多,可四舍五入# [0.04000000000000001, 0.12000000000000002, 0.20000000000000004, 0.28, 0.3600000000000001, 0.44000000000000006# 经分析可知保留三位小数即可# forward_tracks.append(round(s, 4))  # 每次滑动的距离添加到forward_tracks列表中,并四舍五入# 反着滑动到准确位置,分10次回退多余的20 个距离# 共回退24,因为实际运行时未完全回退回去,估计是上面四舍五入,导致前进时超出20# 前9个数加起来为20,后面一个数考虑到上面四舍五入,根据各个电脑情况,可稍微修正# back_tracks = [-3, -3, -2, -2, -2, -2, -2, -1, -3]# back_tracks = [-3, -3, -2, -2, -2, -2, -2, -1, -3, -0.000006]forward_tracks = [space/3, space/3, space/3]back_tracks = [-3, -3, -1]return {'forward_tracks': forward_tracks, 'back_tracks': back_tracks}def main(name,pwd):# 创建浏览器对象driver = webdriver.Chrome()login_url = 'https://passport.bilibili.com/login'driver.get(login_url)driver.maximize_window()  # 打开网页窗口time.sleep(1)# 输入账号密码input_name = driver.find_element_by_id('login-username')input_name.clear()input_name.send_keys(name)input_pass = driver.find_element_by_id('login-passwd')input_pass.clear()input_pass.send_keys(pwd)# 点击登录# print("1"* 10)time.sleep(1)driver.find_element_by_xpath('//*[@id="geetest-wrap"]/ul/li[5]/a[1]').click()# 移动滑块,闭合图形slider_button(driver)# 最后判断下是否成功,若成功:# 如果破解成功，html的title会变# <a data-v-4d9bc88b="" href="//message.bilibili.com" target="_blank" title="消息" class="t"><!----> <!---->消息</a>while driver.title != '消息':print('破解失败，再次破解')# 刷新图片# //*[@id="sliderddnormal-choose"]/div[2]/div[4]/div/a  # /html/body/div[2]/div[2]/div[6]/div/div[2]/div/a[2]# <a class="geetest_refresh_1" href="javascript:;"><div class="geetest_refresh_icon"></div><div class="geetest_refresh_tip">刷新验证</div></a># href="javascript:;是假链接,点击后没反应,待解决driver.find_elements_by_xpath('//*[@class="geetest_refresh_1"]')[0].click()print("刷新图片")time.sleep(2)# 移动滑块,闭合图形slider_button(driver)print("3" * 10)print('破解成功')time.sleep(19)# 释放浏览器ActionChains(driver).release().perform()time.sleep(9)# 点击滑块使图形闭合
def slider_button(driver):"""docstring for 点击滑块使缺口闭合验证码"""time.sleep(1)# 1、出现滑块验证，获取 有缺口的图片# driver.find_element_by_xpath('//*[@id="app"]/section/div/ul/li[2]/h2').click()# <canvas class="geetest_canvas_slice geetest_absolute" width="260" height="160" style="opacity: 1; display: block;"></canvas># geetest_canvas_slice geetest_absolute# document.querySelector("h2, h3").style.backgroundColor = "red";# driver.execute_script('document.querySelectorAll(".geetest_canvas_slice geetest_absolute")[1].style="opacity: 0; display: block;')##  js 使小图隐藏,以便截图driver.execute_script('document.querySelectorAll("canvas")[2].style="opacity: 0; display: block;"')# driver.execute_script('document.querySelectorAll("canvas")[3].style=""')# /html/body/div[2]/div[2]/div[6]/div/div[1]/div[1]/div/a/div[1]/div/canvas[1]# <canvas class="geetest_canvas_bg geetest_absolute" height="160" width="260" style="display: none"></canvas>## 大图且缺口, 截图canvas_1 = driver.find_element_by_xpath('//canvas[@class="geetest_canvas_bg geetest_absolute"]')picture1 = get_image(driver, '1',canvas_1)print("4" * 10)print(picture1)print("5" * 10)# 2、执行js改变css样式，显示背景图！！！！！重点是这一步！# <canvas class="geetest_canvas_fullbg geetest_fade geetest_absolute" height="160" width="260" style="display: bolck; opacity: 1;"></canvas># driver.execute_script('document.querySelectorAll("canvas")[2].style=""')# driver.execute_script('document.querySelectorAll("canvas")[4].style="display: bolck; opacity: 1"')## 大图且完图,css让其显示driver.execute_script('document.querySelectorAll("canvas")[3].style=""')print("6" * 10)time.sleep(1)# 3、大图且没有缺口  截图canvas_2 = driver.find_element_by_xpath('//canvas[@class="geetest_canvas_fullbg geetest_fade geetest_absolute"]')picture2 = get_image(driver, '2',canvas_2)# 恢复小图css 为显示状态driver.execute_script('document.querySelectorAll("canvas")[2].style="opacity: 1; display: block;"')# 4、对比两种图片的像素点，找出位移space = get_space(picture1, picture2)print(space)print("#7" * 10)# 5. 模拟人工滑动,调用get_tracks()返回  {'forward_tracks': forward_tracks, 'back_tracks': back_tracks}tracks = get_tracks(space)# 6.点击button = driver.find_element_by_class_name('geetest_slider_button')ActionChains(driver).click_and_hold(button).perform()# 前进轨迹print("#1" * 10)print(tracks['forward_tracks'])print(tracks['back_tracks'])print("#2" * 10)for track in tracks['forward_tracks']:ActionChains(driver).move_by_offset(xoffset=track, yoffset=0).perform()time.sleep(0.5)# 上面for 循环后,滑块,应该缺口右边多出20个距离处# 回退轨迹,只要回退20个距离就好for back_track in tracks['back_tracks']:# 实际回退时,总时查了一点,估计分辨率的问题,现乘以1.25 后okActionChains(driver).move_by_offset(xoffset=back_track * 1.25, yoffset=0).perform()# 来回震荡下 随机数r_n# r_n = random.randint(1,3)# ActionChains(driver).move_by_offset(xoffset=-r_n, yoffset=0).perform()# ActionChains(driver).move_by_offset(xoffset=r_n, yoffset=0).perform()time.sleep(1)# driver.close()# driver.quit()if __name__ == '__main__':name = str(input('请输入账号:'))pwd = str(input('请输入密码:'))main(name, pwd )

result

    raise exception_class(message, screen, stacktrace)
selenium.common.exceptions.JavascriptException: Message: javascript error: Cannot set property 'style' of undefined(Session info: chrome=76.0.3809.100)不知道为何,滑块移动的距离总是超出缺口,

END

https://blog.csdn.net/chenxiao17301/article/details/82911155

https://blog.csdn.net/u012914436/article/details/100031810

https://blog.csdn.net/qq_38612964/article/details/79756492

Selenium + PhantomJS + python图片全屏截取+定位坐标+抠图+图片识别

https://blog.csdn.net/qq_38949193/article/details/81395787

https://blog.csdn.net/qq_31032181/article/details/78799433

https://blog.csdn.net/qq_43614355/article/details/90372449

携程滑块,点选文字识别相关推荐

selenium 超级鹰通过携程滑块和汉字验证码
大概流程 1.进入页面点击同意按钮. 2.得到滑块和滑块区域的xpath 3.定位滑块,设置按住事件,进行拖动 4.xpath定位汉字图片,获取屏幕截图,保存本地 5.超级鹰识别图片返回汉字对应的坐标 ...
Python 识别携程中文验证码（95%正确率）并自动登陆携程+图灵图像验证码识别平台
这两天有一个业务需求,需要登陆不同的携程账号获取订单信息,但是由于携程有验证码检测机制,而且是个中文验证码比较难识别,试了几家人工打码平台,要么贵,要么延时高,要么没办法24小时运行.最后总算让我找到 ...
破解携程中文验证码爬取机票价格数据
国内机票预定APP携程处于垄断地位,但是携程有反爬虫策略,对于密集的查询请求会要求验证,验证操作有两次,一次是拖动验证,一次是点选中文,selenium+webdriver可以轻松绕过这一反爬虫设置. ...
UED专栏 | 带你认识携程插画系统1.0
本文作者为携程平台UED团队,同时感谢机票.度假.酒店UED团队协同搭建插画系统. 对于每一个设计师来说,插画总是让人喜爱又烦恼.喜爱是因为插画具有特殊的表现力.丰富的图形语言.鲜明的个性特征,运用在 ...
Spring Boot + 百度 OCR 图片文字识别功能
一.知识点简介 OCR(optical character recognition)文字识别是指电子设备(例如扫描仪或数码相机)检查纸上打印的字符,然后用字符识别方法将形状翻译成计算机文字的过程. - ...
利用baidu的api和python的GUI实现通用文字识别功能
文章目录 1. 准备工作 1.1. 注册和申请 1.2. python 的库 2. 小工具的开发 2.1. 截图功能 2.2. 保存到剪贴板 2.3. 连接到文本识别 2.4. 简单的界面 2.5. ...
本地通用验证码识别DLL/通用滑块识别/文字点选/图标点选/OCR文字识别/本地识别DLL
背景验证码识别一直都是一个重要的话题,近日有一些公司询问本地DLL验证码识别定制的事,可以联系QQ:[167231471]定制本地离线DLL验证码识别.另外给大家普及一下通用验证码识别和滑块缺口检 ...
本地DLL通用验证码识别/滑块识别/文字点选/图标点选/推理/OCR文字识别
背景验证码识别一直都是一个重要的话题,近日有一些公司询问本地DLL验证码识别定制的事,可以联系QQ:[167231471]定制本地离线DLL验证码识别.另外给大家普及一下通用验证码识别和滑块缺口检 ...
本地腾讯滑块识别DLL/本地通用验证码识别DLL/文字点选/图标点选/本地识别DLL
背景验证码识别一直都是一个重要的话题,近日有一些公司询问本地DLL验证码识别定制的事,可以联系QQ:[167231471]定制本地离线DLL验证码识别.另外给大家普及一下通用验证码识别和滑块缺口检 ...

携程滑块,点选文字识别

文章目录

code_datawhale05_xiecheng.py

reslut

code_datawhale06_bzhan_login.py

result

END

携程滑块,点选文字识别相关推荐

最新文章

热门文章