第十五章 scrapy框架使用

发布时间:2023年12月18日

1. 数据提取

CSS获取数据
xptah和CSS混合提取数据

web.css(".class_name::text").extract()

2. 数据过滤

# 根据元素属性判断
if web.xpath("./@class") == "class_name":
	
#根据元素的值判断
value = web.xpath("./div/text()"):

if not value:
	

3. 使用items格式化数据

格式化数据,便于后期问题定位
定义数据格式

import scrapy

class DirectItem(scrapy.Item):
    # define the fields for your item here like:
    # name = scrapy.Field()
    house_name= scrapy.Field() # 等于字典的key
    huxing = scrapy.Field()
    mianji = scrapy.Field()
    chaoxiang = scrapy.Field()
    louceng = scrapy.Field()
    zhaungxiu = scrapy.Field()
    pass

# 数据的使用
# 导包
from direct.items import DirectItem

class GameSpider(scrapy.Spider):
    name = "game"
    allowed_domains = ["bj.5i5j.com"]
    start_urls = ["https://bj.5i5j.com/zufang/"]

    def parse(self, response):
        div_lst = house_name = response.xpath("//ul[@class='pList rentList']/li/div[2]")
        for i in div_lst:
            house_name = i.xpath("./h3/a/text()").extract()
            house_msg = i.xpath("./div[1]/p[1]/text()").extract()[0].strip().split(" · ")
            val = DirectItem()
            val["house_name"]=house_name,
            val["huxing"]= house_msg[0],
            val["mianji"]=house_msg[1],
            val["chaoxiang"]= house_msg[2],
            val["louceng"]= house_msg[3],
            val["zhaungxiu"]=house_msg[4]
            yield val    

4. 数据存储

数据存储方案

  • 数据存储在csv文件中
  • 数据存储在mysql中
  • 数据存储在MongoDB中
  • 文件的存储

1. 数据存储在csv文件中

# 创建存储类
class CsvSave:
    #打开文件
    def open_spider(self, spider):
        self.f = open("./message.csv",mode="a",encoding="utf-8")
    def close_spider(self, spider):
        if self.f:
            self.f.close()
    def process_item(self,item,spider):
        self.f.write(f"{item}\n")
        return item

# settings中打开管道
ITEM_PIPELINES = {
   "direct.pipelines.DirectPipeline": 300,
   "direct.pipelines.CsvSave":301
}

2. 数据存储到mysql中

数据存储到mysql
连接数据库


class MysqlSave:
    def open_spider(self,spider):
        self.con = pymysql.connect(
            host='localhost',
            port=3306,
            user="root",
            password="root",
            database= "test_day1"
        )
    def close_spider(self, spider):
        if self.con:
            self.con.close()

    def process_item(self,item,spider):
        try:
            cursor = self.con.cursor()
            sql = "INSERT INTO house_msg(house_name,house_huxing,house_mianji,house_chaoxiang,house_louceng,house_zhuangxiu) VALUES (%s,%s,%s,%s,%s,%s);"
            val = (item['house_name'],item['huxing'],item['mianji'],item['chaoxiang'],item['louceng'],item['zhuangxiu'])
            cursor.execute(sql,val)
            self.con.commit()
        except Exception as e:
            print(e)
            self.con.rollback()
        finally:
            if cursor:
                cursor.close()
        return item


数据库定义到settings中

MySql = {
   'host':'localhost',
   'port':3306,
   'user':'root',
   'password':'root',
   'database':'test_day1'
}



class MysqlSave:
    def open_spider(self,spider):
        self.con = pymysql.connect(
            host=MySql['host'],
            port=MySql['port'],
            user=MySql['user'],
            password=MySql['password'],
            database= MySql['database']
        )

3. MongoDB的存储


class MongoDBSave:
    def open_spider(self,spider):
        self.client = pymongo.MongoClient(host="localhost",port=27017)
        db = self.client['house']
        self.collection = db['house_msg']

    def close_spider(self,spider):
        self.client.close()

    def process_item(self, item, spider):
        val = {
            'house_name':item['house_name'],
            'huxing':item['huxing'],
            'mianji':item['mianji'],
            'chaoxiang':item['chaoxiang'],
            'louceng':item['louceng'],
            'zhuangxiu':item['zhuangxiu']
        }
        self.collection.insert_one(val)
        return item

4. 文件的存储

文件会涉及到多个href的请求,可以在spider里,添加多次请求

'''
1.将获取的url与resp中请求的url合并
resp.urljoin(href)

2.callback回调,可以将返回的request返回个callback后面的函数(参数)

'''


class PictureSpider(scrapy.Spider):
    name = "picture"
    allowed_domains = ["75ll.com"]
    start_urls = ["https://www.75ll.com/meinv/"]
    # 获取文件的src,并返回给pipline
    def parse(self, resp,**kwargs):
        src_lst = resp.xpath("//div[@class='channel_list']/ul/div/div/div/div/a/img/@src").extract()
        for i in src_lst:
            src_dic = TupianItem()
            src_dic['src'] = i
            yield src_dic
		# 翻页,返回请求,回调parser函数
        next_href =resp.xpath("//a[@class='next_a']/@href").extract_first()
        if next_href:
            yield scrapy.Request(
                url=next_href,
                method='get',
                callback=self.parse
            )

下载图片的pipline

使用图片下载的pipline时,需要单独配置用来保存图片的文件夹
在这里插入图片描述

class DownLoadPicture(ImagesPipeline):
    # 负责下载,直接返回request请求
    def get_media_requests(self, item, info):
        # 返回请求
        yield scrapy.Request(item['src'])

    # 准备文件路径
    def file_path(self, request, response=None, info=None, *, item=None):
        name = request.url.split('/')[-1]
        # 返回文件路径
        return f"img/{name}"

    # 返回保存的文件的信息
    def item_completed(self, results, item, info):
        print(results)


判断xpath中是否包含xx文字

# 判断标签文本中是否包含下一页
href = resp.xpath("//div[@class='page']/a[contains(text(),'下一页')]")

文章来源:https://blog.csdn.net/weixin_49806206/article/details/134961505
本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。