python爬虫爬取一年天气数据

以下是使用Python爬虫爬取一年天气数据的示例代码：

import requests
from bs4 import BeautifulSoup
import pandas as pd

def get_weather_data(year):
    # 创建一个空的DataFrame用于保存数据
    weather_data = pd.DataFrame(columns=['日期', '天气', '最高温度', '最低温度'])

    # 循环遍历每个月份
    for month in range(1, 13):
        # 构造URL
        url = f'http://www.weather.com.cn/weather/{year}{month:02d}.shtml'

        # 发送HTTP请求获取页面内容
        response = requests.get(url)
        response.encoding = 'utf-8'
        html = response.text

        # 使用BeautifulSoup解析页面内容
        soup = BeautifulSoup(html, 'html.parser')

        # 获取每天的天气数据
        days = soup.select('.t li')

        # 循环遍历每天的数据
        for day in days:
            date = day.select('.time')[0].text.strip()
            weather = day.select('.wea')[0].text.strip()
            high_temp = day.select('.tem span')[0].text.strip()
            low_temp = day.select('.tem i')[0].text.strip()

            # 将数据添加到DataFrame中
            weather_data = weather_data.append({'日期': date, '天气': weather, '最高温度': high_temp, '最低温度': low_temp}, ignore_index=True)

    return weather_data

# 调用函数获取2021年的天气数据
weather_data_2021 = get_weather_data(2021)

# 打印结果
print(weather_data_2021)