AQI 爬虫程序,原网页进行了加密处理,不能直接进行爬去了,需要进行改造。
系统:Ubuntu16.04+python3.6.3
关键:使用pandas的read_html函数读取表格内容。使用selenium的webdriver获取页面。
参考:https://zhuanlan.zhihu.com/p/31099968
import time
import pandas as pd
import requests
from bs4 import BeautifulSoup
from django.utils.http import urlquote
from selenium import webdriver
driver = webdriver.Firefox(executable_path ='/usr/local/bin/geckodriver')
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(7, 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))
month_set.append(('2016-%s' % 10))
month_set.append(('2016-%s' % 11))
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'
time.sleep(10)
for i in range(len(month_set)):
str_month = month_set[i]
utf8_city = urlquote(city)
weburl = ('%s%s&month=%s' % (base_url,utf8_city,str_month))
driver.get(weburl)
print (driver)
dfs = pd.read_html(driver.page_source,header = 0)[0]
driver.quit()
dfs.to_csv(file_name)
print('%d---%s,%s---DONE' % (city_set.index(city), city, str_month))
time.sleep(30)