requessts库,fake_useragent库，lxml库爬虫中国天气网最低气温，全国气温

1：html5lib，lxml , html.parser的比较使用

2：pyecharts图形制作库,制作可视化图形数据

3：map函数的使用

4：lambda的使用

上述小点要学习。。。。。。

# -*- coding: utf-8 -*-
# @Time    : 2019/1/26 0026 17:08
# @Author  : huangtao
# @Site    : 
# @File    : 3.中国天气网.py
# @Software: PyCharm
# @Blog    :https://blog.csdn.net/Programmer_huangtao
import requests
from bs4 import BeautifulSoup
from fake_useragent import UserAgent
from pyecharts import Bar

ALL_DATA = []

def parse_page(url):
    HEADERS = {
        'User-Agent': UserAgent().chrome}
    res = requests.get(url,headers=HEADERS)
    # print(res.content.decode('utf-8'))
    text = res.content
    soup = BeautifulSoup(text,'html5lib')
    conMidtab = soup.find('div',attrs={'class':'conMidtab'})
    tables = conMidtab.find_all('table')
    for table in tables:
        trs = table.find_all('tr')[2:]
        for index,tr in enumerate(trs):
            tds = tr.find_all('td')
            city_td = tds[0]
            if index == 0:
                city_td = tds[1]
            city = list(city_td.stripped_strings)[0]
            temp_td = tds[-2]
            min_temp = list(temp_td.stripped_strings)[0]
            ALL_DATA.append({"city":city,"min_temp":int(min_temp)})
            # print({"city":city,"min_temp":int(min_temp)})
def main():
    urls = [
        'http://www.weather.com.cn/textFC/hb.shtml',
         'http://www.weather.com.cn/textFC/db.shtml',
        'http://www.weather.com.cn/textFC/hd.shtml',
        'http://www.weather.com.cn/textFC/hz.shtml',
        'http://www.weather.com.cn/textFC/hn.shtml',
        'http://www.weather.com.cn/textFC/xb.shtml',
        'http://www.weather.com.cn/textFC/xn.shtml',
        'http://www.weather.com.cn/textFC/gat.shtml'
        ]
    for url in urls:
        parse_page(url)
    ALL_DATA.sort(key=lambda data:data['min_temp'])
    data = ALL_DATA[0:10]
    cities = list(map(lambda x:x['city'],data))
    temps = list(map(lambda x:x['min_temp'],data))
    chart = Bar('中国天气最低气温')
    chart.add('',cities,temps)
    chart.render('wendu.html')
    print(ALL_DATA)

if __name__ == '__main__':
    main()