天气网天气数据爬取

# -*- coding: utf-8 -*-
# 引入必要库
import pyecharts
from pyecharts.charts import Radar
from pyecharts import options as opts
import requests
from bs4 import BeautifulSoup
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import matplotlib
import numpy as np

plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False  # 用来正常显示负号(X轴)


class item:
    def __init__(self):
        self.date = list()  # 日期
        self.max_temp = list()  # 最高温
        self.min_temp = list()  # 最低温
        self.weather = list()  # 天气
        self.wind_direction = list()  # 风向


Data_Box = item()  # 数据盒子


# 函数默认返回北京市2018年1月到12月的url
def get_url(city='beijing'):
    '''
    city为城市拼写的字符串，year为年份+月份
    '''
    for time in range(201801, 201813):
        url = "http://lishi.tianqi.com/{}/{}.html".format(city, time)
        yield url


def get_datas():
    urls = get_url()
    cookie = {
   
        "cityPy": "UM_distinctid=171f2280ef23fb-02a4939f3c1bd4-335e4e71-144000-171f2280ef3dab; Hm_lvt_ab6a683aa97a52202eab5b3a9042a8d2=1588905651; CNZZDATA1275796416=871124600-1588903268-%7C1588990372; Hm_lpvt_ab6a683aa97a52202eab5b3a9042a8d2=1588994046"}
    header = {
   
        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko