python + selenium + PhantomJS 获取腾讯应用宝APP评论

 1 #/usr/bin/env python
 2 # -*- coding: utf-8 -*-
 3 
 4 from selenium import webdriver
 5 import time
 6 import xlsxwriter
 7 import json
 8 
 9 def main(url):
10 
11     driver = webdriver.PhantomJS()
12     driver.get(url)
13     time.sleep(5)
14 
15     with open("./tencent_yyb.json", "wb") as write_obj:
16         try:
17             # numstr = driver.find_elements_by_xpath('//*[@id="J_CommentCount"]')[0].text
18             # num = int(numstr[1:4])
19             flag = True
20             while flag:  # 循环，无法点击“查看更多评论”报错，重置标志位flag完成获取所有内容
21                 try:
22                     driver.find_element_by_id("J_DetCommentShowMoreBtn").click()  # 点击“查看更多评论”
23                     #driver.implicitly_wait(10)
24                     time.sleep(1)
25                 except Exception as err:
26                     print err
27                     flag = False
28 
29             data = driver.find_elements_by_xpath('//*[@id="J_DetDataContainer"]/div/div[8]/ul/li')
30 
31             repeat = {}  # 缓存数据
32             for item in data:
33                 val_list = item.text.split('\n')
34                 if len(val_list) == 3:
35                     ptime = val_list[1].encode("utf-8")
36                     pcontent = val_list[2].encode("utf-8")
37                     if ptime not in repeat.keys():  # 去重
38                         repeat[ptime] = pcontent
39                         write_obj.write("{\"发布时间\":\"" + ptime + "\", \"发布内容\":\"" + pcontent + "\"}\n" )
40                 else:
41                     ptime = val_list[1].encode("utf-8")
42                     pcontent = ' '
43                     if ptime not in repeat.keys():
44                         repeat[ptime] = pcontent
45                         write_obj.write("{\"发布时间\":\"" + ptime + "\", \"发布内容\":\"" + pcontent + "\"}\n" )
46         except Exception as err:
47             print err
48         finally:
49             driver.quit()
50 
51 def get_num(name):
52     jfilename = "./" + name + ".json"
53     count = len(open(jfilename, "rU").readlines()) - 1
54     return count
55 
56 # xls转换写入
57 def w_xls(url, name):
58     perfix = time.strftime('%Y%m%d')
59     xfilename = "./" + name + '_' + perfix + ".xlsx"
60     jfilename = "./" + name + ".json"
61     workbook = xlsxwriter.Workbook(xfilename)
62     worksheet1 = workbook.add_worksheet()
63     lines = int(get_num(name))
64     worksheet1.set_column = ('A:C', lines+1)
65     worksheet1.write('A1', 'URL')
66     worksheet1.write('B1', u'发布时间')
67     worksheet1.write('C1', u'发布内容')
68 
69     row = 1
70     with open(jfilename, "r") as obj_read:
71         for line in obj_read.readlines():
72             data = json.loads(line)
73             worksheet1.write(row, 0, url)
74             worksheet1.write(row, 1, data[u'发布时间'])
75             worksheet1.write(row, 2, data[u'发布内容'])
76             row += 1
77 
78 
79 if __name__ == '__main__':
80 
81     url = ''
82 
83     main(url)
84     time.sleep(3)
85     w_xls(url, "tencent_yyb")