import csv
import json
import time
from datetime import datetime, timedelta
from concurrent.futures import ThreadPoolExecutor
import httpx


# 获取城市信息
def get_cities(cities_json):
    cities_dict = {}
    for city, provinces in cities_json.items():
        city_list = []
        for _, province_mes in provinces.items():
            for province_name in province_mes:
                area_id = cities_json[city][_][province_name]['AREAID']
                city_list.append({province_name: area_id})
        cities_dict[city] = city_list
    return cities_dict


# 生成日期列表
def generate_year_and_month_list(start_year, start_month):
    current_date = datetime.now()
    start_date = datetime(start_year, start_month, 1)
    year_and_month_list = []

    while start_date <= current_date:
        year = start_date.year
        month = start_date.month
        year_and_month_str = f"{year}{month:02d}"
        year_and_month_list.append(year_and_month_str)

        if start_date.month == 12:
            start_date = datetime(start_date.year + 1, 1, 1)
        else:
            start_date += timedelta(days=31)

    return year_and_month_list


# 发送请求
def send_request(url, params, headers):
    while True:
        resp = httpx.get(url, params=params, headers=headers, verify=False)
        content = resp.text
        if "setTimeout" in content:
            # 增加延时
            time.sleep(.05)
            continue
        encode = resp.encoding
        decoded_html_content = content.encode(encode).decode('utf-8')
        return decoded_html_content


# 获取天气数据
def fetch_weather_data_for_city(root_url, city, province, area_id, all_dates_str):
    headers = {
        'Accept': '*/*',
        'Accept-Language': 'zh-CN,zh;q=0.9',
        'Connection': 'keep-alive',
        'Referer': 'http://www.weather.com.cn/',
        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36',
    }

    timestamp = str(int(datetime.timestamp(datetime.now())))
    params = {'_': timestamp}

    with open('weather_data.csv', mode='a', newline='', encoding='utf-8') as csv_file:
        fieldnames = ['province', 'date', 'hgl', 'hmax', 'hmin']
        writer = csv.DictWriter(csv_file, fieldnames=fieldnames)

        for date in all_dates_str:
            uri = f"/{date[:4]}/{area_id}_{date}.html"
            content = send_request(root_url + uri, params, headers)

            json_str = content[len("var fc40 = "):]
            data_dict = json.loads(json_str)

            for item in data_dict:
                row = {
                    'province': province,
                    'date': item["date"],
                    'hgl': item["hgl"],
                    'hmax': item["hmax"],
                    'hmin': item["hmin"]
                }
                writer.writerow(row)


# 主程序入口
if __name__ == "__main__":
    response = httpx.get('https://j.i8tq.com/weather2020/search/city.js')
    cities_str = "\n".join(response.text.split("\n")[1:])
    cities_json = json.loads(cities_str)

    cities_dict = get_cities(cities_json)

    all_dates_str = generate_year_and_month_list(2023, 1)
    root_url = "http://d1.weather.com.cn/calendar_new"

    with open('weather_data.csv', mode='w', newline='', encoding='utf-8') as csv_file:
        fieldnames = ['city', 'date', 'hgl', 'hmax', 'hmin']
        csv_file.write(",".join(fieldnames) + "\n")

    with ThreadPoolExecutor(max_workers=10) as executor:  # 使用最多 10 个线程
        for city, provinces in cities_dict.items():
            if city == "重庆":
                for province in provinces:
                    for area_id in province.values():
                        executor.submit(fetch_weather_data_for_city, root_url, city, list(province.keys())[0], area_id,
                                        all_dates_str)