AQI 爬虫_windows版

系统：win10x64+python3.6
注意事项：
webdriver特别是chromedriver的版本要与自己浏览器的版本一致。
time.sleep()是为了保证爬取成功，不然页面会一闪而过，什么都爬不到。
各个模块的下载安装请自行百度。
原网址已加密，需要使用pandas中的read_html函数读取表格内容。

import time

import pandas as pd
import requests
from bs4 import BeautifulSoup
from urllib import parse
from selenium import webdriver

driver = webdriver.Chrome('D:\ProgramData\Anaconda3\selenium\chromedriver.exe')

base_url = 'https://www.aqistudy.cn/historydata/daydata.php?city='
str_city = '北京'

def get_month_set():
    month_set = list()
    for i in range(1, 10):
        month_set.append(('2014-0%s' % i))
    for i in range(10, 13):
        month_set.append(('2014-%s' % i))
    for i in range(1, 10):
        month_set.append(('2015-0%s' % i))
    for i in range(10, 13):
        month_set.append(('2015-%s' % i))
    for i in range(1, 10):
        month_set.append(('2016-0%s' % i))
    for i in range(10, 13):
        month_set.append(('2016-%s' % i))
    for i in range(1, 10):
        month_set.append(('2017-0%s' % i))
    for i in range(10, 13):
        month_set.append(('2017-%s' % i))
    return month_set

def get_city_set():
    str_file = r'city.txt'
    fp = open(str_file,'rb')
    city_set = list()
    for line in fp.readlines():
        city_set.append(str(line.strip(),encoding='utf-8'))
    return city_set

month_set = get_month_set()
city_set = get_city_set()

for city in city_set:
    file_name = city + '.csv'
    fp = open('aqi/' + file_name, 'w')
    fp.write('%s,%s,%s,%s,%s,%s,%s,%s,%s\n'%('日期','AQI','质量等级','PM25','PM10','SO2','CO','NO2','O3_8h'))#表头
    for i in range(len(month_set)):
        str_month = month_set[i]

        utf8_city = parse.quote(city)
        weburl = ('%s%s&month=%s' % (base_url,utf8_city,str_month))

        driver.get(weburl)
        time.sleep(1)
        dfs = pd.read_html(driver.page_source,header = 0)[0]

        for j in range(0,len(dfs)):
            date = dfs.iloc[j,0]
            aqi = dfs.iloc[j,1]
            grade = dfs.iloc[j,2]
            pm25 = dfs.iloc[j,3]
            pm10 = dfs.iloc[j,4]
            so2 = dfs.iloc[j,5]
            co = dfs.iloc[j,6]
            no2 = dfs.iloc[j,7]
            o3 = dfs.iloc[j,8]
            fp.write(('%s,%s,%s,%s,%s,%s,%s,%s,%s\n' % (date,aqi,grade,pm25,pm10,so2,co,no2,o3)))
        print('%d---%s,%s---DONE' % (city_set.index(city), parse.quote(city), str_month))
    fp.close()
driver.quit()