Python Scrapy爬虫框架爬取51job职位信息并保存至数据库

Python Scrapy爬虫框架爬取51job职位信息并保存至数据库

————————————————————————————————
版权声明：本文为CSDN博主「杠精运动员」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/CN_Nidus/article/details/109695639

项目要求

• 使用Scrapy爬虫框架练习爬取51Job网站招聘职位信息并保存至数据库（已sqlite3为例）。

工具软件

python == 3.7
pycharm == 20.2.3
sqlite == 3.33.0
scrapy == 2.4.0

爬取字段

• 职位名称、薪资、发布日期、工作地点、任职要求、公司名称、公司规模、福利待遇、详情链接。

思路分析

• 网页连接：https://search.51job.com/list/010000,000000,0000,00,9,99,python%25E7%2588%25AC%25E8%2599%25AB,2,1.html?lang=c&postchannel=0000&workyear=99&cotype=99&degreefrom=99&jobterm=99&companysize=99&ord_field=0&dibiaoid=0&line=&welfare=
• 分析：只要更改链接中的两个关键参数即可实现自定义爬取内容和翻页。
• 查看网页源代码

• 分析：页面展示的每一条职位信息都隐藏在script标签内以json格式保存。

完整代码展示

• 定义Items

import scrapy

class Job51Item(scrapy.Item):
    jobname = scrapy.Field()
    salary = scrapy.Field()
    pubdate = scrapy.Field()
    job_adress = scrapy.Field()
    requirement = scrapy.Field()
    company_name = scrapy.Field()
    company_size = scrapy.Field()
    welfare = scrapy.Field()
    url = scrapy.Field()

• 定义Pipelines

import sqlite3

class Job51Pipeline:
    def open_spider(self, spider):
        # 连接数据库
        self.conn = sqlite3.connect('D:\\OfficeLib\\JetBrains\\PyCharmFiles\\PycharmProjects\\MyScrapy\\db.sqlite3')
        self.cur = self.conn.cursor()
        spider.index = 1

    def close_spider(self, spider):
    	# 关闭数据库
        self.cur.close()
        self.conn.close()

    def process_item(self, item, spider):
        # 构造sql语句
        insert_sql = 'insert into job51 (jobname, salary, pubdate, job_adress, requirement, company_name, company_size, welfare, url)' \
                     'values ("{}", "{}", "{}", "{}", "{}", "{}", "{}", "{}", "{}")'.format(
                        item["jobname"],
                        item["salary"],
                        item["pubdate"],
                        item["job_adress"],
                        item["requirement"],
                        item["company_name"],
                        item["company_size"],
                        item["welfare"],
                        item["url"])
        self.cur.execute(insert_sql)
        self.conn.commit()
        print("第{}条保存成功".format(spider.index))
        spider.index += 1
        return item

• 爬虫代码

# -*- coding: utf-8 -*-

import re
import scrapy
import json
from MyScrapy.items import Job51Item
from urllib import parse


class Job51Spider(scrapy.Spider):
    name = 'Job51'
    allowed_domains = ['51job.com']

    # 指定管道
    custom_settings = {
        'ITEM_PIPELINES': {'MyScrapy.pipelines.Job51Pipeline': 299}
    }

    # 自定义爬取内容
    infokey = str(input("请输入要爬取的职位："))
    print("开始爬取,请稍等……")
    infokey_encode = parse.quote(infokey)   # 进行url编码
    base_url = 'https://search.51job.com/list/010000,000000,0000,00,9,99,{},2,1.html?' \
               'lang=c&postchannel=0000&workyear=99&cotype=99&degreefrom=99&jobterm=99&companysize=99&ord_field=0&dibiaoid=0&line=&welfare='

    start_urls = [base_url.format(infokey_encode)]

    def parse(self, response, **kwargs):
        item = Job51Item()
        # 获取响应数据并格式化
        body = response.body.decode('gbk')
        data_json = re.findall(r'window.__SEARCH_RESULT__ =(.+)}</script>', body)[0] + "}"
        data_py = json.loads(data_json)

        for data in data_py['engine_search_result']:
            print("正在爬取第{}条……".format(self.index))
            item["jobname"] = data["job_title"]
            if len(data["providesalary_text"]) <= 0:
                item["salary"] = ""
            else:
                item["salary"] = data["providesalary_text"]
            item["pubdate"] = data["updatedate"]
            if len(data["workarea_text"]) <= 0:
                item["job_adress"] = ""
            else:
                item["job_adress"] = data["workarea_text"]
            item["requirement"] = "、".join(data["attribute_text"])
            item["company_name"] = data["company_name"]
            item["company_size"] = data["companysize_text"]
            if len(data["jobwelf_list"]) <= 0:
                item["welfare"] = ""
            else:
                item["welfare"] = "、".join(data["jobwelf_list"])
            item["url"] = data["job_href"]
            yield item

        # 构造下一页url，当前设置只爬取前5页数据
        for i in range(2, 5):
            next_url = "https://search.51job.com/list/010000,000000,0000,00,9,99,{},2,{}.html?" \
                       "lang=c&postchannel=0000&workyear=99&cotype=99&degreefrom=99&jobterm=99" \
                       "&companysize=99&ord_field=0&dibiaoid=0&line=&welfare=".format(self.infokey_encode, i)
            yield scrapy.Request(url=next_url, callback=self.parse)

• Settings配置

# 设置请求延时
DOWNLOAD_DELAY = 3
RANDOMIZE_DOWNLOAD_DELAY = True

# 设置user-agent，可以设置多个
USER_AGENT = [
    "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/535.1 (KHTML, like Gecko) Chrome/14.0.835.163 Safari/535.1",
]

# 关闭robot协议
ROBOTSTXT_OBEY = False

# 设置下载中间件
DOWNLOADER_MIDDLEWARES = {
    'MyScrapy.middlewares.RandomUserAgentMiddleware': 542,
}

# 启用管道
ITEM_PIPELINES = {
    'MyScrapy.pipelines.Job51Pipeline': 299,
}

• 下载中间件使用随机user-agent

from scrapy import signals
import random

class RandomUserAgentMiddleware:
    def process_request(self, request, spider):
        useragent = random.choice(spider.settings.get("USER_AGENT"))
        request.headers["User-Agent"] = useragent

运行结果

结束语
爬虫有风险！抓数需谨慎！！！
————————————————
版权声明：本文为CSDN博主「杠精运动员」的原创文章，遵循CC 4.0 BY-SA版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/CN_Nidus/article/details/109695639

本文章仅供学习交流使用，切勿他用！