scrapy框架爬取国际庄2011-2022的天气情况

目标网站：http://www.tianqihoubao.com/lishi/

一.创建项目+初始化爬虫文件：

scrapy startpoject tianqihoubao
cd tianqihoubao
scrapy genspider weather www.tianqihoubao.com

二.配置settings.py

USER_AGENT = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/99.0.4844.51 Safari/537.36 Edg/99.0.1150.30'

ROBOTSTXT_OBEY = False #君子协议 注释掉或者改为false

ITEM_PIPELINES = {
   'tianqihoubao.pipelines.TianqihoubaoPipeline': 300,
   'tianqihoubao.pipelines.MySQLStoreCnblogsPipeline': 300
}

# 连接数据MySQL
# 数据库地址
MYSQL_HOST = 'localhost'
# 数据库用户名:
MYSQL_USER = 'root'
# 数据库密码
MYSQL_PASSWORD = '123456'
# 数据库端口
MYSQL_PORT = 3306
# 数据库名称
MYSQL_DBNAME = 'data'
# 数据库编码
MYSQL_CHARSET = 'utf8'

三.修改items.py

import scrapy

class TianqihoubaoItem(scrapy.Item):
    # define the fields for your item here like:
    # name = scrapy.Field()
    riqi = scrapy.Field()
    tianqi = scrapy.Field()
    qiwen = scrapy.Field()
    wind = scrapy.Field()

四.编写weather.py

scrapy.Request()教程：Scrapy爬虫入门教程十一 Request和Response（请求和响应）

import scrapy

from tianqihoubao.items import TianqihoubaoItem

class WeatherSpider(scrapy.Spider):
    name = 'weather'
    allowed_domains = ['www.tianqihoubao.com']
    start_urls = ['http://www.tianqihoubao.com/lishi/shijiazhuang.html']

    def parse(self, response):
        #for path in self.start_urls:
        data = response.xpath('//div[@id="content"]/table[@class="b"]/tr')
        next_page = response.xpath('//div[@id="content"]/div[@class="box pcity"]/ul/li/a/@href').extract()
        for i in data[1:]:
            item = TianqihoubaoItem()
            item['riqi'] = i.xpath('./td/a/text()').extract()[0].replace('\r\n', '').replace('\n', '').replace(' ', '').strip()
            item['tianqi'] = i.xpath('./td/text()').extract()[2].replace('\r\n', '').replace('\n', '').replace(' ', '').strip()
            item['qiwen'] = i.xpath('./td/text()').extract()[3].replace('\r\n', '').replace('\n', '').replace(' ', '').strip()
            item['wind'] = i.xpath('./td/text()').extract()[4].replace('\r\n', '').replace('\n', '').replace(' ', '').strip()
            #print(item['tianqi'])

            yield item

        print("下一页", next_page)
        if next_page and len(next_page) > 5:
            # 翻页
            for i in next_page:
                yield scrapy.Request(url='http://www.tianqihoubao.com'+i, callback=self.parse)
        else:
            print("没有下一页了" * 10)
        print(data)

五.配置pipelines.py(一个保存为csv，一个保存到mysql上)

import copy
import csv
import time

from pymysql import cursors
from twisted.enterprise import adbapi

class TianqihoubaoPipeline(object):
    # 保存为csv格式
    def __init__(self):
        # 打开文件，指定方式为写，利用第3个参数把csv写数据时产生的空行消除
        self.f = open("weather.csv", "a", newline="", encoding="utf-8")
        # 设置文件第一行的字段名，注意要跟spider传过来的字典key名称相同
        self.fieldnames = ["riqi", "tianqi", "qiwen", "wind"]
        # 指定文件的写入方式为csv字典写入，参数1为指定具体文件，参数2为指定字段名
        self.writer = csv.DictWriter(self.f, fieldnames=self.fieldnames)
        # 写入第一行字段名，因为只要写入一次，所以文件放在__init__里面
        self.writer.writeheader()

    def process_item(self, item, spider):
        # 写入spider传过来的具体数值
        self.writer.writerow(item)
        # 写入完返回
        return item

    def close(self, spider):
        self.f.close()

class MySQLStoreCnblogsPipeline(object):
    # 初始化函数
    def __init__(self, db_pool):
        self.db_pool = db_pool

    # 从settings配置文件中读取参数
    @classmethod
    def from_settings(cls, settings):
        # 用一个db_params接收连接数据库的参数
        db_params = dict(
            host=settings['MYSQL_HOST'],
            user=settings['MYSQL_USER'],
            password=settings['MYSQL_PASSWORD'],
            port=settings['MYSQL_PORT'],
            database=settings['MYSQL_DBNAME'],
            charset=settings['MYSQL_CHARSET'],
            use_unicode=True,
            # 设置游标类型
            cursorclass=cursors.DictCursor
        )
        # 创建连接池
        db_pool = adbapi.ConnectionPool('pymysql', **db_params)

        # 返回一个pipeline对象
        return cls(db_pool)

    # 处理item函数
    def process_item(self, item, spider):
        # 对象拷贝，深拷贝  --- 这里是解决数据重复问题！！！
        asynItem = copy.deepcopy(item)

        # 把要执行的sql放入连接池
        query = self.db_pool.runInteraction(self.insert_into, asynItem)

        # 如果sql执行发送错误,自动回调addErrBack()函数
        query.addErrback(self.handle_error, item, spider)

        # 返回Item
        return item

    # 处理sql函数
    def insert_into(self, cursor, item):
        # 创建sql语句
        sql = "INSERT INTO weather (riqi,tianqi,qiwen,wind) VALUES ('{}','{}','{}','{}')".format(
            item['riqi'], item['tianqi'], item['qiwen'], item['wind'])
        # 执行sql语句
        cursor.execute(sql)
        # 错误函数

    def handle_error(self, failure, item, spider):
        # #输出错误信息
        print("failure", failure)

六.结果

 这个在爬取的时候没有按时间顺序来，后续在搞一搞