day6 

昨日作业：自动登陆抽屉新热榜
 1 from selenium import webdriver
 2 import time
 3 
 4 driver = webdriver.Chrome(r'D:\BaiduNetdiskDownload\chromedriver_win32\chromedriver.exe')
 5 
 6 # 把窗口转成全屏
 7 driver.maximize_window()
 8 
 9 try:
10     driver.get('https://dig.chouti.com/')
11     driver.implicitly_wait(10)
12     time.sleep(5)
13 
14     # 1、点击登录
15     login_btn = driver.find_element_by_id('login_btn')
16     login_btn.click()
17     time.sleep(2)
18 
19     # 2、输入用户名
20     phone = driver.find_element_by_class_name('login-phone')
21     phone.send_keys('15622792660')
22 
23     # 3、输入密码
24     pwd = driver.find_element_by_class_name('pwd-password-input')
25     pwd.send_keys('kermit46709394')
26 
27     # 4、确认登录
28     login_submit = driver.find_element_by_class_name('btn-large')
29     login_submit.click()
30 
31     time.sleep(20)
32 
33 # 捕获异常并打印
34 except Exception as e:
35     print(e)
36 
37 finally:
38     driver.close()

今日内容:

注意: selenium驱动的浏览器是干净的，没有任何缓存。

    1、selenium剩余用法
    2、selenium万能登录破解
    3、selenium爬取京东商品信息
    4、破解极验滑动验证码

Xpath语法:

今日作业:
    1、总结课堂知识点，写博客
    2、爬取京东商品信息
    3、滑动验证（提高题）

1。selenium选择器之Xpath:
 1 from selenium import webdriver
 2 
 3 driver = webdriver.Chrome(r'D:\BaiduNetdiskDownload\chromedriver_win32\chromedriver.exe')
 4 
 5 
 6 try:
 7     # 隐式等待: 写在get请求前
 8     driver.implicitly_wait(5)
 9 
10     driver.get('https://doc.scrapy.org/en/latest/_static/selectors-sample1.html')
11 
12     # 显式等待: 写在get请求后
13     # wait.until(...)
14 
15     '''
16     
17 <html>
18  <head>
19   <base href='http://example.com/' />
20   <title>Example website</title>
21  </head>
22  <body>
23   <div id='images'>
24    <a href='image1.html'>Name: My image 1 <br /><img src='image1_thumb.jpg' /></a>
25    <a href='image2.html'>Name: My image 2 <br /><img src='image2_thumb.jpg' /></a>
26    <a href='image3.html'>Name: My image 3 <br /><img src='image3_thumb.jpg' /></a>
27    <a href='image4.html'>Name: My image 4 <br /><img src='image4_thumb.jpg' /></a>
28    <a href='image5.html'>Name: My image 5 <br /><img src='image5_thumb.jpg' /></a>
29   </div>
30  </body>
31 </html>
32     '''
33     # 根据xpath语法查找元素
34     # / 从根节点开始找第一个
35     html = driver.find_element_by_xpath('/html')
36     # html = driver.find_element_by_xpath('/head')  # 报错
37     print(html.tag_name)
38 
39     # // 从根节点开始找任意一个节点
40     div = driver.find_element_by_xpath('//div')
41     print(div.tag_name)
42 
43     # @
44     # 查找id为images的div节点
45     div = driver.find_element_by_xpath('//div[@id="images"]')
46     print(div.tag_name)
47     print(div.text)
48 
49     # 找到第一个a节点
50     a = driver.find_element_by_xpath('//a')
51     print(a.tag_name)
52 
53     # 找到所有a节点
54     a_s = driver.find_elements_by_xpath('//a')
55     print(a_s)
56 
57     # 找到第一个a节点的href属性
58     # get_attribute:获取节点中某个属性
59     a = driver.find_element_by_xpath('//a').get_attribute('href')
60     print(a)
61 
62 finally:
63     driver.close()

 

2.selenium剩余操作：

  1 ''''''
  2 '''
  3 点击、清除操作
  4 '''
  5 # from selenium import webdriver
  6 # from selenium.webdriver.common.keys import Keys
  7 # import time
  8 #
  9 # driver = webdriver.Chrome(r'D:\BaiduNetdiskDownload\chromedriver_win32\chromedriver.exe')
 10 #
 11 # try:
 12 #     driver.implicitly_wait(10)
 13 #     # 1、往jd发送请求
 14 #     driver.get('https://www.jd.com/')
 15 #     # 找到输入框输入围城
 16 #     input_tag = driver.find_element_by_id('key')
 17 #     input_tag.send_keys('围城')
 18 #     # 键盘回车
 19 #     input_tag.send_keys(Keys.ENTER)
 20 #     time.sleep(2)
 21 #     # 找到输入框输入墨菲定律
 22 #     input_tag = driver.find_element_by_id('key')
 23 #     input_tag.clear()
 24 #     input_tag.send_keys('墨菲定律')
 25 #     # 找到搜索按钮点击搜索
 26 #     button = driver.find_element_by_class_name('button')
 27 #     button.click()
 28 #     time.sleep(10)
 29 #
 30 # finally:
 31 #     driver.close()
 32 
 33 
 34 '''
 35 获取cookies  （了解）
 36 '''
 37 # from selenium import webdriver
 38 # import time
 39 #
 40 # driver = webdriver.Chrome(r'D:\BaiduNetdiskDownload\chromedriver_win32\chromedriver.exe')
 41 #
 42 # try:
 43 #     driver.implicitly_wait(10)
 44 #     driver.get('https://www.zhihu.com/explore')
 45 #     print(driver.get_cookies())
 46 #
 47 #     time.sleep(10)
 48 # finally:
 49 #     driver.close()
 50 
 51 '''
 52 选项卡
 53 '''
 54 #选项卡管理：切换选项卡，有js的方式windows.open,有windows快捷键：
 55 # ctrl+t等，最通用的就是js的方式
 56 # import time
 57 # from selenium import webdriver
 58 #
 59 # browser = webdriver.Chrome()
 60 # try:
 61 #     browser.get('https://www.baidu.com')
 62 #
 63 #     # execute_script: 执行javascrpit代码
 64 #     # 弹窗操作
 65 #     # browser.execute_script('alert("tank")')
 66 #     # 新建浏览器窗口
 67 #     browser.execute_script(
 68 #         '''
 69 #         window.open();
 70 #         '''
 71 #     )
 72 #     time.sleep(1)
 73 #     print(browser.window_handles)  # 获取所有的选项卡
 74 #     # 切换到第二个窗口
 75 #     # 新:
 76 #     browser.switch_to.window(browser.window_handles[1])
 77 #     # 旧:
 78 #     # browser.switch_to_window(browser.window_handles[1])
 79 #
 80 #     # 第二个窗口往淘宝发送请求
 81 #     browser.get('https://www.taobao.com')
 82 #     time.sleep(5)
 83 #
 84 #     # 切换到第一个窗口
 85 #     browser.switch_to_window(browser.window_handles[0])
 86 #     browser.get('https://www.sina.com.cn')
 87 #
 88 #     time.sleep(10)
 89 # finally:
 90 #     browser.close()
 91 
 92 
 93 '''
 94 ActionChangs动作链
 95 '''
 96 # from selenium import webdriver
 97 # from selenium.webdriver import ActionChains
 98 # import time
 99 #
100 # driver = webdriver.Chrome()
101 # driver.implicitly_wait(10)
102 # driver.get('http://www.runoob.com/try/try.php?filename=jqueryui-api-droppable')
103 #
104 # try:
105 #
106 #     # driver.switch_to_frame('iframeResult')
107 #     # 切换到id为iframeResult的窗口内
108 #     driver.switch_to.frame('iframeResult')
109 #
110 #     # 源位置
111 #     draggable = driver.find_element_by_id('draggable')
112 #
113 #     # 目标位置
114 #     droppable = driver.find_element_by_id('droppable')
115 #
116 #     # 调用ActionChains，必须把驱动对象传进去
117 #     # 得到一个动作链对象，复制给一个变量
118 #     actions = ActionChains(driver)
119 #
120 #     # 方式一: 机器人
121 #     # 瞬间把源图片位置秒移到目标图片位置
122 #     # actions.drag_and_drop(draggable, droppable)  # 编写一个行为
123 #     # actions.perform()  # 执行编写好的行为
124 #
125 #
126 #     # 方式二: 模拟人的行为
127 #     source = draggable.location['x']
128 #     target = droppable.location['x']
129 #     print(source, target)
130 #
131 #     distance = target - source
132 #     print(distance)
133 #
134 #     # perform：每个动作都要调用perform执行
135 #
136 #     # 点击并摁住源图片
137 #     ActionChains(driver).click_and_hold(draggable).perform()
138 #
139 #     s = 0
140 #     while s < distance:
141 #         # 执行位移操作
142 #         ActionChains(driver).move_by_offset(xoffset=2, yoffset=0).perform()
143 #         s += 2
144 #
145 #     # 释放动作链
146 #     ActionChains(driver).release().perform()
147 #
148 #     time.sleep(10)
149 #
150 #
151 # finally:
152 #     driver.close()
153 
154 
155 '''
156 前进、后退
157 '''
158 # from selenium import webdriver
159 # import time
160 #
161 # driver = webdriver.Chrome()
162 #
163 # try:
164 #     driver.implicitly_wait(10)
165 #     driver.get('https://www.jd.com/')
166 #     driver.get('https://www.baidu.com/')
167 #     driver.get('https://www.cnblogs.com/')
168 #
169 #     time.sleep(2)
170 #
171 #     # 回退操作
172 #     driver.back()
173 #     time.sleep(1)
174 #     # 前进操作
175 #     driver.forward()
176 #     time.sleep(1)
177 #     driver.back()
178 #     time.sleep(10)
179 #
180 # finally:
181 #     driver.close()

3.破解登陆：

 1 from selenium import webdriver
 2 from selenium.webdriver import ChromeOptions
 3 import time
 4 r'''
 5 步骤:
 6     1、打开文件的查看，显示隐藏文件
 7     2、找到C:\Users\administortra\AppData\Local\Google\Chrome\User Data
 8         删除Default文件
 9     3、重新打开浏览器，并登陆百度账号
10         - 此时会创建一个新的Default缓存文件
11     4、添加cookies
12     5、关闭谷歌浏览器后执行程序
13 '''
14 # 获取options对象，参数对象
15 options = ChromeOptions()
16 
17 # 获取cookies保存路径
18 # 'C:\Users\administortra\AppData\Local\Google\Chrome\User Data'
19 profile_directory = r'--user-data-dir=C:\Users\administortra\AppData\Local\Google\Chrome\User Data'
20 
21 # 添加用户信息目录
22 options.add_argument(profile_directory)
23 
24 # 把参数加载到当前驱动中  chrome_options默认参数，用来接收options对象
25 driver = webdriver.Chrome(chrome_options=options)
26 
27 try:
28     driver.implicitly_wait(10)
29     driver.get('https://www.baidu.com/')
30     '''
31     BDUSS:*****
32     '''
33     # 添加用户cookies信息
34     # name、value必须小写
35     driver.add_cookie({"name": "BDUSS", "value": "用户session字符串"})
36 
37     # 刷新操作
38     driver.refresh()
39 
40     time.sleep(10)
41 
42 finally:
43     driver.close()

4.selenium爬取京东商品信息：

  1 # ''''''
  2 # '''
  3 # 爬取京东商品信息:
  4 #     请求url:
  5 #         https://www.jd.com/
  6 #     提取商品信息:
  7 #         1.商品详情页
  8 #         2.商品名称
  9 #         3.商品价格
 10 #         4.评价人数
 11 #         5.商品商家
 12 # '''
 13 # from selenium import webdriver
 14 # from selenium.webdriver.common.keys import Keys
 15 # import time
 16 #
 17 # driver = webdriver.Chrome()
 18 #
 19 # try:
 20 #     driver.implicitly_wait(10)
 21 #     # 1、往京东主页发送请求
 22 #     driver.get('https://www.jd.com/')
 23 #
 24 #     # 2、输入商品名称，并回车搜索
 25 #     input_tag = driver.find_element_by_id('key')
 26 #     input_tag.send_keys('macbook')
 27 #     input_tag.send_keys(Keys.ENTER)
 28 #     time.sleep(2)
 29 #
 30 #     # 通过JS控制滚轮滑动获取所有商品信息
 31 #     js_code = '''
 32 #         window.scrollTo(0,5000);
 33 #     '''
 34 #     driver.execute_script(js_code)  # 执行js代码
 35 #
 36 #     # 等待数据加载
 37 #     time.sleep(2)
 38 #
 39 #     # 3、查找所有商品div
 40 #     # good_div = driver.find_element_by_id('J_goodsList')
 41 #     good_list = driver.find_elements_by_class_name('gl-item')
 42 #     n = 1
 43 #     for good in good_list:
 44 #         # 根据属性选择器查找
 45 #         # 商品链接
 46 #         good_url = good.find_element_by_css_selector(
 47 #             '.p-img a').get_attribute('href')
 48 #
 49 #         # 商品名称
 50 #         good_name = good.find_element_by_css_selector(
 51 #             '.p-name em').text.replace("\n", "--")
 52 #
 53 #         # 商品价格
 54 #         good_price = good.find_element_by_class_name(
 55 #             'p-price').text.replace("\n", ":")
 56 #
 57 #         # 评价人数
 58 #         good_commit = good.find_element_by_class_name(
 59 #             'p-commit').text.replace("\n", " ")
 60 #
 61 #         # 商品商家
 62 #         good_from = good.find_element_by_class_name(
 63 #             'J_im_icon').text.replace("\n", " ")
 64 #
 65 #         good_content = f'''
 66 #                     商品链接: {good_url}
 67 #                     商品名称: {good_name}
 68 #                     商品价格: {good_price}
 69 #                     评价人数: {good_commit}
 70 #                     商品商家: {good_from}
 71 #                     \n
 72 #                     '''
 73 #         print(good_content)
 74 #         with open('jd.txt', 'a', encoding='utf-8') as f:
 75 #             f.write(good_content)
 76 #
 77 #     next_tag = driver.find_element_by_link_text('下一页')
 78 #
 79 #     next_tag.click()
 80 #
 81 #     time.sleep(10)
 82 #
 83 #
 84 # finally:
 85 #     driver.close()
 86 
 87 
 88 
 89 ''''''
 90 '''
 91 爬取京东商品信息:
 92     请求url:
 93         https://www.jd.com/
 94     提取商品信息:
 95         1.商品详情页
 96         2.商品名称
 97         3.商品价格
 98         4.评价人数
 99         5.商品商家
100 '''
101 from selenium import webdriver
102 from selenium.webdriver.common.keys import Keys
103 import time
104 
105 
106 def get_good(driver):
107     try:
108 
109         # 通过JS控制滚轮滑动获取所有商品信息
110         js_code = '''
111             window.scrollTo(0,5000);
112         '''
113         driver.execute_script(js_code)  # 执行js代码
114 
115         # 等待数据加载
116         time.sleep(2)
117 
118         # 3、查找所有商品div
119         # good_div = driver.find_element_by_id('J_goodsList')
120         good_list = driver.find_elements_by_class_name('gl-item')
121         n = 1
122         for good in good_list:
123             # 根据属性选择器查找
124             # 商品链接
125             good_url = good.find_element_by_css_selector(
126                 '.p-img a').get_attribute('href')
127 
128             # 商品名称
129             good_name = good.find_element_by_css_selector(
130                 '.p-name em').text.replace("\n", "--")
131 
132             # 商品价格
133             good_price = good.find_element_by_class_name(
134                 'p-price').text.replace("\n", ":")
135 
136             # 评价人数
137             good_commit = good.find_element_by_class_name(
138                 'p-commit').text.replace("\n", " ")
139 
140             good_content = f'''
141                         商品链接: {good_url}
142                         商品名称: {good_name}
143                         商品价格: {good_price}
144                         评价人数: {good_commit}
145                         \n
146                         '''
147             print(good_content)
148             with open('jd.txt', 'a', encoding='utf-8') as f:
149                 f.write(good_content)
150 
151         next_tag = driver.find_element_by_class_name('pn-next')
152         next_tag.click()
153 
154         time.sleep(2)
155 
156         # 递归调用函数
157         get_good(driver)
158 
159         time.sleep(10)
160 
161     finally:
162         driver.close()
163 
164 
165 if __name__ == '__main__':
166 
167     good_name = input('请输入爬取商品信息:').strip()
168 
169     driver = webdriver.Chrome()
170     driver.implicitly_wait(10)
171     # 1、往京东主页发送请求
172     driver.get('https://www.jd.com/')
173 
174     # 2、输入商品名称，并回车搜索
175     input_tag = driver.find_element_by_id('key')
176     input_tag.send_keys(good_name)
177     input_tag.send_keys(Keys.ENTER)
178     time.sleep(2)
179 
180     get_good(driver)

5.破解极验滑动验证：

  1 ''''''
  2 '''
  3 破解极验滑动验证
  4 博客园登录url:
  5     https://account.cnblogs.com/signin?returnUrl=https%3A%2F%2Fwww.cnblogs.com%2F
  6 1、输入用户名与密码，并点击登录
  7 2、弹出滑动验证，获取有缺口与完整的图片
  8 3、通过像素点进行比对，获取滑动位移距离
  9 4、模拟人的行为轨迹
 10 5、开始滑动
 11 '''
 12 from selenium import webdriver  # 用来驱动浏览器的
 13 from selenium.webdriver import ActionChains  # 破解滑动验证码的时候用的 可以拖动图片
 14 import time
 15 from PIL import Image  # pip3 install pillow
 16 import random
 17 
 18 option = webdriver.ChromeOptions()
 19 option.add_argument('disable-infobars')
 20 
 21 driver = webdriver.Chrome(chrome_options=option)
 22 
 23 
 24 def get_snap(driver):
 25     # selenium自带的截图网页全屏图片
 26     driver.save_screenshot('snap.png')
 27 
 28     img = driver.find_element_by_class_name('geetest_canvas_img')
 29 
 30     left = img.location['x']
 31 
 32     upper = img.location['y']
 33 
 34     right = left + img.size['width']
 35     lower = upper + img.size['height']
 36 
 37     # print(left, upper, right, lower)
 38     img_obj = Image.open('snap.png')
 39 
 40     # 对屏幕进行截取,获取滑动验证图片
 41     image = img_obj.crop((left, upper, right, lower))
 42 
 43     return image
 44 
 45 
 46 def get_image1(driver):
 47     time.sleep(0.2)
 48     js_code = '''
 49     var x = document.getElementsByClassName('geetest_canvas_fullbg')[0].style.display="block";
 50     console.log(x)
 51     '''
 52 
 53     time.sleep(1)
 54     driver.execute_script(js_code)
 55 
 56     # 截取图片
 57     img_obj = get_snap(driver)
 58 
 59     return img_obj
 60 
 61 
 62 def get_image2(driver):
 63     time.sleep(0.2)
 64 
 65     js_code = '''
 66     var x = document.getElementsByClassName('geetest_canvas_fullbg')[0].style.display="none";
 67     console.log(x)
 68     '''
 69 
 70     driver.execute_script(js_code)
 71 
 72     time.sleep(1)
 73 
 74     # 截取图片
 75     img_obj = get_snap(driver)
 76 
 77     return img_obj
 78 
 79 
 80 def get_distance(image1, image2):
 81     # 初始值
 82     start = 60
 83 
 84     # 滑块色差
 85     color_num = 60
 86 
 87     for x in range(start, image1.size[0]):
 88         for y in range(image1.size[1]):
 89 
 90             rgb1 = image1.load()[x, y]
 91 
 92             rgb2 = image2.load()[x, y]
 93 
 94             r = abs(rgb1[0] - rgb2[0])
 95             g = abs(rgb1[1] - rgb2[1])
 96             b = abs(rgb1[2] - rgb2[2])
 97 
 98             if not (r < color_num and g < color_num and b < color_num):
 99                 return x - 7
100 
101 
102 def get_stacks(distance):
103     distance += 20
104 
105     '''
106     匀加速\减速运行
107         v = v0 + a * t
108 
109     位移:
110     s = v * t + 0.5 * a * (t**2)
111     '''
112 
113     # 初速度
114     v0 = 0
115 
116     # 加减速度列表
117     a_list = [3, 4, 5]
118 
119     # 时间
120     t = 0.2
121 
122     # 初始位置
123     s = 0
124 
125     # 向前滑动轨迹
126     forward_stacks = []
127 
128     mid = distance * 3 / 5
129 
130     while s < distance:
131         if s < mid:
132             a = a_list[random.randint(0, 2)]
133 
134         else:
135             a = -a_list[random.randint(0, 2)]
136 
137         v = v0
138 
139         stack = v * t + 0.5 * a * (t ** 2)
140 
141         # 每次拿到的位移
142         stack = round(stack)
143 
144         s += stack
145 
146         v0 = v + a * t
147 
148         forward_stacks.append(stack)
149 
150     back_stacks = [-1, -1, -2, -3, -2, -3, -2, -2, -3, -1]
151 
152     return {'forward_stacks': forward_stacks, 'back_stacks': back_stacks}
153 
154 
155 def main():
156     try:
157 
158         driver.get('https://passport.cnblogs.com/user/signin')
159         driver.implicitly_wait(5)
160 
161         # 1.输入用户名与密码,点击登录
162         username = driver.find_element_by_id('LoginName')
163         password = driver.find_element_by_id('Password')
164         login_button = driver.find_element_by_class_name('ladda-label')
165         time.sleep(1)
166         username.send_keys('_tank_')
167         time.sleep(1)
168         password.send_keys('k46709394.')
169 
170         # 这里需要等待账号密码输入完毕后再点击登录按钮，否则的不弹框
171         time.sleep(1)
172         login_button.click()
173         # time.sleep(3)
174 
175         # 2.点击滑动验证按钮,获取图片
176         geetest_button = driver.find_element_by_class_name('geetest_slider_button')
177         geetest_button.click()
178 
179         time.sleep(0.2)
180 
181         # 3.针对完整的图片进行截取
182         image1 = get_image1(driver)
183 
184         # 4.针对有缺口的图片进行截取
185         image2 = get_image2(driver)
186 
187         # 5.对比两张图片,获取滑动距离
188         distance = get_distance(image1, image2)
189 
190         # 6.模拟人为滑动轨迹
191         stacks = get_stacks(distance)
192 
193         # 7.根据滑动轨迹进行滑动
194         forward_stacks = stacks['forward_stacks']
195         back_stacks = stacks['back_stacks']
196 
197         slider_button = driver.find_element_by_class_name('geetest_slider_button')
198         time.sleep(0.2)
199 
200         ActionChains(driver).click_and_hold(slider_button).perform()
201 
202         time.sleep(0.2)
203         for forward_stack in forward_stacks:
204             ActionChains(driver).move_by_offset(xoffset=forward_stack, yoffset=0).perform()
205             time.sleep(0.1)
206         for back_stack in back_stacks:
207             ActionChains(driver).move_by_offset(xoffset=back_stack, yoffset=0).perform()
208             time.sleep(0.1)
209 
210         time.sleep(0.2)
211 
212         ActionChains(driver).move_by_offset(xoffset=5, yoffset=0).perform()
213         ActionChains(driver).move_by_offset(xoffset=-5, yoffset=0).perform()
214 
215         ActionChains(driver).release().perform()
216 
217         time.sleep(50)
218 
219 
220     finally:
221         driver.close()
222 
223 
224 if __name__ == '__main__':
225     main()

 

posted @ 2019-06-18 21:11  亓呆子  阅读(260)  评论(0)  编辑  收藏  举报