Python 分析天气，告诉你中秋应该去哪里

中秋佳节将近，不知道各位小伙伴儿有没有想好去哪里玩呢。不过说实在的，每到节假日，到处都是人山人海，那句“我动也不能动”，还不时的出现在我的耳畔呢。

但是又说回来，假期出游，除了人的因素外，天气的因素是不是也要考虑下呢，今天，我们就带大家来看看，中秋小长假，哪些地方适宜出游。

数据的获取，就从中国天气网站上直接抓取，网络上的一些 API，有的信息不是很全，只能获取最近3天的数据，有的又需要付费，还不如自己抓来的痛快。

http://www.weather.com.cn/weather15d/10124020102A.shtml

网站也没有做什么限制，我们抓数据的时候，只需要控制好访问频率，不要影响人家的正常运行就可以。

同时还需要准备四个数据文件

• 省会城市列表，provincial_capital
• 全国城市 id 信息表，china-city-list.csv
• 著名景点名称列表，attractions
• 全国景点 id 信息表，china-scenic-list.txt

抓取的过程不再详细说明了，直接给出完整代码

# coding = utf-8
"""
@author: zhou
@time:2019/9/5 14:36
@File: main.py
"""

import requests
from bs4 import BeautifulSoup
import time
import os


def get_data(name, city, code):
    print("正在下载城市%s的数据" % city)
    url = 'http://www.weather.com.cn/weather15d/%s.shtml' % code[2:]
    res = requests.get(url).content.decode()
    content = BeautifulSoup(res, "html.parser")
    weather_list = content.find('ul', attrs={'class': 't clearfix'}).find_all('li')
    items = map(parse_item, weather_list)
    save_to_csv(name, city, items)
    time.sleep(1)


def parse_item(item):
    time = item.find('span', attrs={'class': 'time'}).text
    wea = item.find('span', attrs={'class': 'wea'}).text
    tem = item.find('span', attrs={'class': 'tem'}).text
    wind = item.find('span', attrs={'class': 'wind'}).text
    wind_level = item.find('span', attrs={'class': 'wind1'}).text
    result = {
 "time": time,
 "wea": wea,
 "tem": tem,
 "wind": wind,
 "wind_level": wind_level
    }
    return result


def save_to_csv(name, city, data):
    if not os.path.exists('%s_data.csv' % name):
 with open('%s_data.csv' % name, 'a+', encoding='utf-8') as f:
     f.write('city,time,wea,tem,wind,wind_leveln')
     for d in data:
  try:
      row = '{},{},{},{},{},{}'.format(city,
    d['time'],
    d['wea'],
    d['tem'],
    d['wind'],
    d['wind_level'])
      f.write(row)
      f.write('n')
  except:
      continue
    else:
 with open('%s_data.csv' % name, 'a+', encoding='utf-8') as f:
     for d in data:
  try:
      row = '{},{},{},{},{},{}'.format(city,
    d['time'],
    d['wea'],
    d['tem'],
    d['wind'],
    d['wind_level'])
      f.write(row)
      f.write('n')
  except:
      continue


if __name__ == '__main__':
    import pandas as pd
    provincial = pd.read_csv('provincial_capital')
    china_city_code = pd.read_csv('china-city-list.csv')
    china_scenic_code = pd.read_csv('china-scenic-list.txt', sep='t')
    china_scenic_code.columns = ['ID', 'name', 'area', 'provincial']
    attraction = pd.read_csv('attractions')
    provincial_data = pd.Dataframe()
    attraction_data = pd.Dataframe()

    # 省会抓取
    for i in provincial['city'].values.tolist():
 for j in china_city_code['City_CN'].values.tolist():
     if j == i:
  provincial_data = pd.concat([china_city_code[china_city_code['City_CN'] == j], provincial_data])

    for city in provincial_data['City_CN'].values.tolist():
 city_id = provincial_data[provincial_data['City_CN'] == city]['City_ID'].values.tolist()[0]
 get_data('weather', city, city_id)

    # 景点抓取
    for a in attraction['attractions'].values.tolist():
 for c in china_scenic_code['name'].values.tolist():
     if c == a:
  attraction_data = pd.concat([china_scenic_code[china_scenic_code['name'] == c], attraction_data])

    for attrac in attraction_data['name'].values.tolist():
 city_id = attraction_data[attraction_data['name'] == attrac]['ID'].values.tolist()[0]
 get_data('attraction', attrac, city_id)

weather_dict = {
    "snow": 100,
    "rain": 80,
    "cloud": 50,
    "overcast": 60,
    "sun": 20
}