1 # coding = utf-8 
 2 """
 3 从http://lishi.tianqi.com/beijing/下载部分北京天气记录保存到文件中。
 4 利用BeautifulSoup模块
 5 """
 6 __author__ = 'hillfree'
 7 
 8 
 9 
10 from urllib import request
11 from bs4 import BeautifulSoup
12 
13 
14 def downloadPages():
15     # Iterate through year, months and day
16     for year in range(2011, 2014):
17         for month in range(1, 13):
18 
19             # 目前数据只提供从2011年1月至2013年3月的数据
20             if year == 2013 and month == 4:
21                 break
22 
23             stamp = "{0}{1:02d}".format(year, month)
24             url = "http://lishi.tianqi.com/beijing/{0}.html".format(stamp,)
25             filename = "BeijingWeather{0}.html".format(stamp,)
26 
27             # 方法1： 直接使用urlretrieve()。 推荐
28             request.urlretrieve(url, filename)
29             print(stamp + "OK!")
30 
31             # 方法2： 分别打开链接和文件写入
32             # page = request.urlopen(url)
33             # print(url + " Open OK!")
34             # # 注意此处一定用‘wb’打开文件
35             # file = open(filename, 'wb')
36             # file.write(page.read())
37             # file.close()
38             # print(filename + "Save OK!")
39             # page.close()
40 
41 def parsePageToCSV():
42     output = open("BeijingWeather.csv", 'w', encoding="utf-8")
43     for year in range(2011, 2014):
44         for month in range(1, 13):
45 
46             # 目前数据只提供从2011年1月至2013年3月的数据
47             if year == 2013 and month == 4:
48                 break
49 
50             stamp = "{0}{1:02d}".format(year, month)
51             filename = "BeijingWeather{0}.html".format(stamp,)
52 
53             file = open(filename, 'r')
54             soup = BeautifulSoup(file)
55 
56 
57             monthData = soup.find(name="div", attrs={"class":"tqtongji2"});
58             if monthData == None:
59                 break
60             uls = monthData.findAll("ul")[1:]   # 滤去表头汉字
61             for ul in uls:
62                 lis = ul.findAll("li")
63                 # 依次取出日期、最高气温、最低气温、天气、风向、风力
64                 item = "{0}, {1}, {2}, {3}, {4}, {5}".format(lis[0].string, lis[1].string,
65                                                              lis[2].string, lis[3].string, lis[4].string, lis[5].string, )
66                 print(item)
67                 output.write(item + "\n")
68 
69     output.close()
70 
71 
72 if __name__ == "__main__":
73     downloadPages()
74     parsePageToCSV()