1 url ="https://api.waqi.info/mapq/bounds/?bounds=31.064698120353743,121.201171875,31.487235582017444,121.84936523437499&inc=placeholders&k=_2Y2EzVxxIDVsfIydASBRWXmldZA4+LREbFkY3ZQ==&_=1497098578289" #动态网页网址
 2
 3 user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'
 4 headers = { 'User-Agent' : user_agent } #头文件
 5
 6 try:
 7     request = urllib.request.Request(url,headers = headers)
 8     response = urllib.request.urlopen(request) #获取数据
 9 except urllib.error.URLError as e:
10     if hasattr(e,"code"):
11         print(e.code)
12     if hasattr(e,"reason"):
13         print(e.reason)
14
15 aqi_trans=json.load(response) #json数据的解析
16 """
17 aqi_trans[0]数据结构：
18 {'aqi': '68',
19 'city': 'Yangpu Sipiao, Shanghai (上海杨浦四漂)',
20 'idx': 702,
21 'img': '_c_azCs5IzEvPSMzUfzZ3xbOty57Onv1sTxMA',
22 'lat': 31.2659,
23 'lon': 121.536,
24 'pol': 'pm25',
25 'stamp': 1497160800,
26 'tz': '+0800',
27 'utime': '2017-06-11 14:00:00',
28 'x': '482'}
29 共11个数据点
30 """

 1 aqi_table=pd.DataFrame(columns=['Idx','Time','Site','AQI'])
 2
 3 for i in range(len(aqi_trans)):
 4     pos_idx=int(aqi_trans[i]['idx'])
 5     record_time=aqi_trans[i]['utime']
 6     site=re.findall(r'上海\S*(?=\))',aqi_trans[i]['city'])[0] #正则，提取括号里的文字
 7     aqi=int(aqi_trans[i]['aqi'])
 8     aqi_table=aqi_table.append({'Idx':pos_idx,'Time':record_time,'Site':site,'AQI':aqi},ignore_index=True)
 9 aqi_table=aqi_table.sort_values('Idx',ascending=True)
10 aqi_table=aqi_table.set_index(pd.Series(range(0,11)))
11 aqi_table['Level']=pd.cut(aqi_table['AQI'],[0,50,100,150,200,300,10000],labels=[u'优',u'良',u'轻度污染',u'中度污染',u'重度污染',u'严重污染']) #打标签

import urllib #python 3中urllib和urllib2合并了，很多语法不同了
import json
import re
import pandas as pddef Get_AQI(t):url ="https://api.waqi.info/mapq/bounds/?bounds=31.064698120353743,121.201171875,31.487235582017444,121.84936523437499&inc=placeholders&k=_2Y2EzVxxIDVsfIydASBRWXmldZA4+LREbFkY3ZQ==&_=%d"%(t)#这里用的是动态时间戳，但时间戳好像并不影响数据的实时性，那只要隔一段时间提交一次同样的网址就行了
user_agent = 'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'headers = { 'User-Agent' : user_agent }try:request = urllib.request.Request(url,headers = headers)response = urllib.request.urlopen(request)except urllib.error.URLError as e:if hasattr(e,"code"):print(e.code)if hasattr(e,"reason"):print(e.reason)aqi_trans=json.load(response)aqi_table=pd.DataFrame(columns=['Idx','Time','Site','AQI'])for i in range(len(aqi_trans)):pos_idx=int(aqi_trans[i]['idx'])record_time=aqi_trans[i]['utime']site=re.findall(r'上海\S*(?=\))',aqi_trans[i]['city'])[0]aqi=int(aqi_trans[i]['aqi'])aqi_table=aqi_table.append({'Idx':pos_idx,'Time':record_time,'Site':site,'AQI':aqi},ignore_index=True)aqi_table=aqi_table.sort_values('Idx',ascending=True)aqi_table=aqi_table.set_index(pd.Series(range(0,11)))aqi_table['Level']=pd.cut(aqi_table['AQI'],[0,50,100,150,200,300,10000],\labels=[u'优',u'良',u'轻度污染',u'中度污染',u'重度污染',u'严重污染']) #分段打标签return aqi_tableimport timesleep_time=1800 #30分钟运行一次if __name__ == '__main__':while True:now_time=int(time.time()*1000)aqi_database=Get_AQI(now_time)aqi_database.to_csv('D:\\AQI\\aqi_database.csv',mode='a+') #记录csv文件，a+为追加写入       print(time.ctime())print(Get_AQI(now_time))time.sleep(sleep_time)