scrapy 爬取智联招聘

 

准备工作

　　1. scrapy startproject Jobs
　　2. cd Jobs
　　3. scrapy genspider ZhaopinSpider www.zhaopin.com
　　4. scrapy crawl ZhaopinSpider
　　5. pip install diskcache
　　6. pip install tinydb
　　7. scrapy crawl ZhaopinSpider -o chongqing.json

 

ZhaopinSpider

　

# -*- coding: utf-8 -*-
import os
import json

from tinydb import TinyDB, Query
from furl import furl
import scrapy


class ZhaopinspiderSpider(scrapy.Spider):
    name = 'ZhaopinSpider'
    allowed_domains = ['www.zhaopin.com', 'sou.zhaopin.com', 'fe-api.zhaopin.com']
    start_urls = ['https://www.zhaopin.com/citymap']
    cache_db = TinyDB('ZhaopinSpider-cache.json')  # 缓存数据库
    allowed_cities = ['重庆', ]# '成都', '上海', '深圳', '昆明', '杭州', '贵阳', '宁波']  ## 允许的城市
    F = furl('https://fe-api.zhaopin.com/c/i/sou?pageSize=90&kt=3')  # URL母版
    PAGE_SIZE = 90  # 分页大小

    def get_city_code(self, city_name):
        '''(根据城市名)获取城市代码'''
        Q = Query()
        city = self.cache_db.get(Q.name.search(city_name))
        if isinstance(city, dict):
            return city['code']
        else:
            print('@' * 100)
            print(type(city))

    def init_city_info(self, response):
        '''初始化城市信息'''
        # 取源码
        script_text = response.xpath('//script[text()[contains(., "__INITIAL_STATE__")]]/text()').extract_first()
        # 去收尾空格
        script_text = script_text.strip()
        # 预处理为符合json规范的数据
        script_json = script_text[script_text.index('=') + 1:]
        # 将json字符串转为字典
        script_dict = json.loads(script_json)
        '''
        # 存储取得的json, 便于调试查看
        with open('text.json', 'wt', encoding='utf-8') as f:
            json.dump(script_dict, f, indent=4, ensure_ascii=False)
        '''
        '''
        city_list = []  # 存储城市列表
        # 将字典中的城市提取到列表中，便于查找
        for ch in script_dict['cityList']['cityMapList']:
            city_list.extend(script_dict['cityList']['cityMapList'][ch])
        # 筛选出重庆，并获取城市码
        city_code = (list(filter(lambda city: city['name'] == '重庆', city_list)) or [{'code': None}])[0]['code']
        '''
        for ch in script_dict['cityList']['cityMapList']:
            for city in script_dict['cityList']['cityMapList'][ch]:
                self.cache_db.insert(city)

    def parse(self, response):
        # if not os.path.exists('ZhaopinSpider-cache.json'):
        if not bool(self.eache_db.all()):
            self.init_city_info(response)
        # 迭代每一个要爬取的城市
        for city_name in self.allowed_cities:
            # 启动 爬取某个城市 第一个请求
            # import ipdb; ipdb.set_trace()
            yield self.request_city(city_name)

    def request_city(self, city_name, page_start=0):
        '''构造 爬取某个具体的城市 的请求对象'''
        city_code = self.get_city_code(city_name)
        url_data = {
            'cityId': city_code,
            'kw': 'python',
            'start': page_start
        }
        # 要爬取的页面的URL
        url = self.F.copy().add(url_data).url
        # import ipdb; ipdb.set_trace()
        req = scrapy.Request(url, callback=self.parse_city, dont_filter=False)
        # 使用 meta 传递附加数据，在 callback 中可以通过 respo.meta 取得
        req.meta['city_name'] = city_name
        req.meta['page_start'] = page_start
        return req

    def parse_city(self, response):
        '''解析具体的页面'''
        # 解析json格式的响应结果
        resp_dict = json.loads(response.body_as_unicode())
        # 总共所能爬取的条数
        num_found = resp_dict['data']['numFound']
        # 获取当前请求的 page_start
        page_start = response.meta['page_start']
        # 下一次请求，需要的 start 参数
        next_start = page_start + self.PAGE_SIZE
        # import ipdb; ipdb.set_trace()
        # 判断是否有下一页
        if next_start < num_found:
            # 获取当前请求的 城市名
            city_name = response.meta['city_name']
            # 发送下一页请求
            yield self.request_city(city_name, page_start=next_start)
        # 解析数据
        for item in resp_dict['data']['results']:
            # TODO: 解析数据，只取我们需要的信息
            item['spiderName'] = self.name
            # 返回每一条数据
            yield item

 

转载于:https://www.cnblogs.com/sxqfuture/p/10256437.html