# -*- coding: UTF-8 -*-
"""
# @Time: 2022/1/4 11:02
# @Author: 遠方的星
# @CSDN: https://blog.csdn.net/qq_44921056
"""
import chardet
import requests
from lxml import etree
from fake_useragent import UserAgent
import pandas as pd
from matplotlib import pyplot as plt


# 隨機產(chǎn)生請求頭
ua = UserAgent(verify_ssl=False, path='D:/Pycharm/fake_useragent.json')


# 隨機切換請求頭
def random_ua():
    headers = {
        "user-agent": ua.random
    }
    return headers


# 解析頁面
def res_text(url):
    res = requests.get(url=url, headers=random_ua())
    res.encoding = chardet.detect(res.content)['encoding']
    response = res.text
    html = etree.HTML(response)
    return html


# 獲得未來七天及八到十五天的頁面鏈接
def get_url(url):
    html = res_text(url)
    url_7 = 'http://www.weather.com.cn/' + html.xpath('//*[@id="someDayNav"]/li[2]/a/@href')[0]
    url_8_15 = 'http://www.weather.com.cn/' + html.xpath('//*[@id="someDayNav"]/li[3]/a/@href')[0]
    # print(url_7)
    # print(url_8_15)
    return url_7, url_8_15


# 獲取未來七天的天氣情況
def get_data_7(url):
    html = res_text(url)
    list_s = html.xpath('//*[@id="7d"]/ul/li')  # 獲取天氣數(shù)據(jù)列表
    Date, Weather, Low, High = [], [], [], []
    for i in range(len(list_s)):
        list_date = list_s[i].xpath('./h1/text()')[0]  # 獲取日期，如：4日（明天）
        # print(list_data)
        list_weather = list_s[i].xpath('./p[1]/@title')[0]  # 獲取天氣情況，如：小雨轉(zhuǎn)雨夾雪
        # print(list_weather)
        tem_low = list_s[i].xpath('./p[2]/i/text()')  # 獲取最低氣溫
        tem_high = list_s[i].xpath('./p[2]/span/text()')  # 獲取最高氣溫
        if tem_high == []:  # 遇到夜晚情況，篩掉當天的最高氣溫
            tem_high = tem_low  # 無最高氣溫時，使最高氣溫等于最低氣溫
        tem_low = int(tem_low[0].replace('℃', '')) # 將氣溫數(shù)據(jù)處理
        tem_high = int(tem_high[0].replace('℃', ''))
        # print(type(tem_high))
        Date.append(list_date), Weather.append(list_weather), Low.append(tem_low), High.append(tem_high)
    excel = pd.DataFrame()  # 定義一個二維列表
    excel['日期'] = Date
    excel['天氣'] = Weather
    excel['最低氣溫'] = Low
    excel['最高氣溫'] = High
    # print(excel)
    return excel


def get_data_8_15(url):
    html = res_text(url)
    list_s = html.xpath('//*[@id="15d"]/ul/li')
    Date, Weather, Low, High = [], [], [], []
    for i in range(len(list_s)):
        # data_s[0]是日期，如：周二（11日），data_s[1]是天氣情況，如：陰轉(zhuǎn)晴，data_s[2]是最低溫度，如：/-3℃
        data_s = list_s[i].xpath('./span/text()')
        # print(data_s)
        date = modify_str(data_s[0])  # 獲取日期情況
        weather = data_s[1]
        low = int(data_s[2].replace('/', '').replace('℃', ''))
        high = int(list_s[i].xpath('./span/em/text()')[0].replace('℃', ''))
        # print(date, weather, low, high)
        Date.append(date), Weather.append(weather), Low.append(low), High.append(high)
    # print(Date, Weather, Low, High)
    excel = pd.DataFrame()  # 定義一個二維列表
    excel['日期'] = Date
    excel['天氣'] = Weather
    excel['最低氣溫'] = Low
    excel['最高氣溫'] = High
    # print(excel)
    return excel


# 將8-15天日期格式改成與未來7天一致
def modify_str(date):
    date_1 = date.split('（')
    date_2 = date_1[1].replace('）', '')
    date_result = date_2 + '（' + date_1[0] + '）'
    return date_result


# 實現(xiàn)數(shù)據(jù)可視化
def get_image(date, weather, high, low):
    # 用來正常顯示中文標簽
    plt.rcParams['font.sans-serif'] = ['SimHei']
    # 用來正常顯示負號
    plt.rcParams['axes.unicode_minus'] = False
    # 根據(jù)數(shù)據(jù)繪制圖形
    fig = plt.figure(dpi=128, figsize=(10, 6))
    ax = fig.add_subplot(111)
    plt.plot(date, high, c='red', alpha=0.5, marker='*')
    plt.plot(date, low, c='blue', alpha=0.5, marker='o')
    # 給圖表中兩條折線中間的部分上色
    plt.fill_between(date, high, low, facecolor='blue', alpha=0.2)
    # 設(shè)置圖表格式
    plt.title('邳州近15天天氣預(yù)報', fontsize=24)
    plt.xlabel('日期', fontsize=12)
    # 繪制斜的標簽，以免重疊
    fig.autofmt_xdate()
    plt.ylabel('氣溫', fontsize=12)
    # 參數(shù)刻度線設(shè)置
    plt.tick_params(axis='both', which='major', labelsize=10)
    # 修改刻度
    plt.xticks(date[::1])
    # 對點進行標注，在最高氣溫點處標注當天的天氣情況
    for i in range(15):
        ax.annotate(weather[i], xy=(date[i], high[i]))
    # 顯示圖片
    plt.show()


def main():
    base_url = 'http://www.weather.com.cn/weather1d/101190805.shtml'
    url_7, url_8_15 = get_url(base_url)
    data_1 = get_data_7(url_7)
    data_2 = get_data_8_15(url_8_15)
    data = pd.concat([data_1, data_2], axis=0, ignore_index=True)  # ignore_index=True實現(xiàn)兩張表拼接，不保留原索引
    get_image(data['日期'], data['天氣'], data['最高氣溫'], data['最低氣溫'])


if __name__ == '__main__':
    main()