scrapy案例教程

1 scrapy简介

• scrapy常用命令
  |命令 | 格式 |说明|
  |–|–|–|
  |startproject |scrapy startproject <项目名> |创建一个新项目|
  |genspider| scrapy genspider <爬虫文件名> <域名> |新建爬虫文件。
  |runspider| scrapy runspider <爬虫文件> |运行一个爬虫文件，不需要创建项目。
  |crawl| scrapy crawl |运行一个爬虫项目，必须要创建项目。
  |list |scrapy list |列出项目中所有爬虫文件。
  |view| scrapy view | 从浏览器中打开 url 地址。
  |shell| csrapy shell |命令行交互模式。
  |settings |scrapy settings |查看当前项目的配置信息。
• 项目的目录树结构
  
• Scrapy 五大组件

名称	作用说明
Engine(引擎)	整个 Scrapy 框架的核心，主要负责数据和信号在不同模块间传递。
Scheduler(调度器)	用来维护引擎发送过来的 request 请求队列。
Downloader(下载器)	接收引擎发送过来的 request 请求，并生成请求的响应对象，将响应结果返回给引擎。
Spider(爬虫程序)	处理引擎发送过来的 response， 主要用来解析、提取数据和获取需要跟进的二级URL，然后将这些数据交回给引擎。
Pipeline(项目管道)	用实现数据存储，对引擎发送过来的数据进一步处理，比如存 MySQL 数据库等。

• 两大中间件
• 下载器中间件，位于引擎和下载器之间，主要用来包装 request 请求头，比如 UersAgent、Cookies 和代理 IP 等
• 蜘蛛中间件，位于引擎与爬虫文件之间，它主要用来修改响应对象的属性。
• 工作流程图
  

2 创建项目

# 创建项目
scrapy startproject Medical
# 进入项目
cd Medical
# 创建爬虫文件
scrapy genspider medical www.baidu.com

3 自定义初始化请求url

import scrapy
import json
from scrapy.http import Response
from Medical.items import MedicalItem
from tqdm import tqdm

'''
具体的爬虫程序
'''


class MedicalSpider(scrapy.Spider):
    name = "medical"
    allowed_domains = ["beian.cfdi.org.cn"]

    # start_urls = ["https://beian.cfdi.org.cn/CTMDS/pub/PUB010100.do?method=handle05&_dt=20231101162330"]

    # 重写第一次请求处理函数
    def start_requests(self):
        start_url = 'https://www.baidu.com/CTMDS/pub/PUB010100.do?method=handle05&_dt=20231101162330'
        # 发送post请求
        data = {
            'pageSize': '1353',
            'curPage': '1',
        }
        yield scrapy.FormRequest(url=start_url, formdata=data, callback=self.parse)

    def parse(self, response):
        # 转换为json
        jsonRes = json.loads(response.body)
        # 查看响应状态码
        status = jsonRes['success']
        # 如果状态为True
        if status:
            # 获取数据
            dataList = jsonRes['data']
            # 调用详细方法，发起请求（循环发起）
            for row in tqdm(dataList,desc='爬取进度'):
                # 请求详情页url
                urlDetail = f"https://www.baidu.com/CTMDS/pub/PUB010100.do?method=handle04&compId={row['companyId']}"
                # 发起请求
                yield scrapy.Request(url=urlDetail, callback=self.parseDetail, meta={'row': row})

    def parseDetail(self, response: Response):
        # new 一个MedicalItem实例
        item = MedicalItem()

        # 获取上次请求的数据源
        row = response.meta['row']
        item['companyId'] = row['companyId']
        item['linkTel'] = row['linkTel']
        item['recordNo'] = row['recordNo']
        item['areaName'] = row['areaName']
        item['linkMan'] = row['linkMan']
        item['address'] = row['address']
        item['compName'] = row['compName']
        item['recordStatus'] = row['recordStatus']
        item['cancelRecordTime'] = row.get('cancelRecordTime', '')

        # 获取备案信息
        divTextList = response.xpath("//div[@class='col-md-8 textlabel']/text()").extract()
        # 去空白
        divtextList = [text.strip() for text in divTextList]
        compLevel = ''
        if len(divtextList) > 2:
            compLevel = divtextList[2]
        recordTime = ''
        if len(divtextList) > 5:
            recordTime = divtextList[6]
        item['compLevel'] = compLevel
        item['recordTime'] = recordTime

        # 获取其他机构地址
        divListOther = response.xpath("//div[@class='col-sm-8 textlabel']/text()").extract()
        # 去空白
        divtextListOther = [text.strip() for text in divListOther]
        otherOrgAdd = ','.join(divtextListOther)
        item['otherOrgAdd'] = otherOrgAdd

        # 获取备案专业和主要研究者信息
        trList = response.xpath("//table[@class='table table-striped']/tbody/tr")
        tdTextList = [tr.xpath("./td/text()").extract() for tr in trList]
        item['tdTextList'] = tdTextList

        # 返回item
        yield item

4 定义item

# Define here the models for your scraped items
#
# See documentation in:
# https://docs.scrapy.org/en/latest/topics/items.html
import scrapy


class MedicalItem(scrapy.Item):
    # define the fields for your item here like:
    # 省份/地区
    areaName = scrapy.Field()
    # 公司id
    companyId = scrapy.Field()
    # 公司名称
    compName = scrapy.Field()
    # 公司等级
    compLevel = scrapy.Field()
    # 联系人
    linkMan = scrapy.Field()
    # 联系电话
    linkTel = scrapy.Field()
    # 备案号
    recordNo = scrapy.Field()
    # 地址
    address = scrapy.Field()
    # 备案状态
    recordStatus = scrapy.Field()
    # 取消备案时间
    cancelRecordTime = scrapy.Field()
    # 备案时间
    recordTime = scrapy.Field()

    # 其他机构地址
    otherOrgAdd = scrapy.Field()

    # 子表详情(矩阵)
    tdTextList = scrapy.Field()

5 定义管道

# Define your item pipelines here
#
# Don't forget to add your pipeline to the ITEM_PIPELINES setting
# See: https://docs.scrapy.org/en/latest/topics/item-pipeline.html


# useful for handling different item types with a single interface
from itemadapter import ItemAdapter
import pymysql

from Medical.items import MedicalItem


class MedicalPipeline:

    # 开始
    def open_spider(self, spider):
        # 初始化数据库
        self.db = pymysql.connect(
            host='localhost',
            port=3306,
            user='root',
            password='logicfeng',
            database='test2'
        )
        # 创建游标对象
        self.cursor = self.db.cursor()

    def process_item(self, item, spider):
        companyId = item['companyId']
        linkTel = item['linkTel']
        recordNo = item['recordNo']
        areaName = item['areaName']
        linkMan = item['linkMan']
        address = item['address']
        compName = item['compName']
        recordStatus = item['recordStatus']
        cancelRecordTime = item.get('cancelRecordTime', '')
        compLevel = item.get('compLevel', '')
        recordTime = item.get('recordTime', '')
        otherOrgAdd = item.get('otherOrgAdd', '')

        tdTextList = item['tdTextList']

        sql1 = "insert INTO medical_register(company_id,area_name,record_no,comp_name,address,link_man,link_tel,record_status,comp_level,record_time,cancel_record_time,other_org_add) "
        sql2 = f"values('{companyId}','{areaName}','{recordNo}','{compName}','{address}','{linkMan}','{linkTel}','{recordStatus}','{compLevel}','{recordTime}','{cancelRecordTime}','{otherOrgAdd}')"
        sql3 = sql1 + sql2
        # 执行sql
        self.cursor.execute(sql3)
        # 提交
        self.db.commit()

        for tdText in tdTextList:
            tdText.insert(0,companyId)
            # 插入数据库
            sql4 = "insert into medical_register_sub (company_id,professional_name,principal_investigator,job_title) values(%s,%s,%s,%s)"
            self.cursor.execute(sql4, tdText)
            # 提交到数据库
            self.db.commit()

        return item

    def close_spider(self, spider):
        self.cursor.close()
        self.db.close()
        print("关闭数据库！")