python爬虫项目(scrapy-redis分布式爬取房天下租房信息)

python爬虫scrapy项目（二）

　　爬取目标：房天下全国租房信息网站（起始url：http://zu.fang.com/cities.aspx）

　　爬取内容：城市；名字；出租方式；价格；户型；面积；地址；交通

　　反反爬措施：设置随机user-agent、设置请求延时操作、

1、开始创建项目

1 scrapy startproject fang

2、进入fang文件夹，执行启动spider爬虫文件代码，编写爬虫文件。

1 scrapy genspider zufang "zu.fang.com"

　　命令执行完，用Python最好的IDE---pycharm打开该文件目录

3、编写该目录下的items.py文件，设置你需要爬取的字段。

 import scrapy

 class HomeproItem(scrapy.Item):
     # define the fields for your item here like:
     # name = scrapy.Field()

     city = scrapy.Field()  #城市
     title = scrapy.Field()  # 名字
     rentway = scrapy.Field()  # 出租方式
     price = scrapy.Field()    #价格
     housetype = scrapy.Field()  # 户型
     area = scrapy.Field()  # 面积
     address = scrapy.Field()  # 地址
     traffic = scrapy.Field()  # 交通

4、进入spiders文件夹，打开hr.py文件,开始编写爬虫文件

 # -*- coding: utf-8 -*-
 import scrapy
 from homepro.items import HomeproItem
 from scrapy_redis.spiders import RedisCrawlSpider
 # scrapy.Spider
 class HomeSpider(RedisCrawlSpider):
     name = 'home'
     allowed_domains = ['zu.fang.com']
     # start_urls = ['http://zu.fang.com/cities.aspx']

     redis_key = 'homespider:start_urls'
     def parse(self, response):
         hrefs = response.xpath('//div[@class="onCont"]/ul/li/a/@href').extract()
         for href in hrefs:
             href = 'http:'+ href
             yield scrapy.Request(url=href,callback=self.parse_city,dont_filter=True)

     def parse_city(self, response):
         page_num = response.xpath('//div[@id="rentid_D10_01"]/span[@class="txt"]/text()').extract()[0].strip('共页')
         # print('*' * 100)
         # print(page_num)
         # print(response.url)

         for page in range(1, int(page_num)):
             if page == 1:
                 url = response.url
             else:
                 url = response.url + 'house/i%d' % (page + 30)
             print('*' * 100)
             print(url)
             yield scrapy.Request(url=url, callback=self.parse_houseinfo, dont_filter=True)

     def parse_houseinfo(self, response):
         divs = response.xpath('//dd[@class="info rel"]')
         for info in divs:
             city = info.xpath('//div[@class="guide rel"]/a[2]/text()').extract()[0].rstrip("租房")
             title = info.xpath('.//p[@class="title"]/a/text()').extract()[0]
             rentway = info.xpath('.//p[@class="font15 mt12 bold"]/text()')[0].extract().replace(" ", '').lstrip('\r\n')
             housetype = info.xpath('.//p[@class="font15 mt12 bold"]/text()')[1].extract().replace(" ", '')
             area = info.xpath('.//p[@class="font15 mt12 bold"]/text()')[2].extract().replace(" ", '')
             addresses = info.xpath('.//p[@class ="gray6 mt12"]//span/text()').extract()
             address = '-'.join(i for i in addresses)
             try:
                 des = info.xpath('.//p[@class ="mt12"]//span/text()').extract()
                 traffic = '-'.join(i for i in des)
             except Exception as e:
                 traffic = "暂无详细信息"

             p_name = info.xpath('.//div[@class ="moreInfo"]/p/text()').extract()[0]
             p_price = info.xpath('.//div[@class ="moreInfo"]/p/span/text()').extract()[0]
             price = p_price + p_name

             item = HomeproItem()
             item['city'] = city
             item['title'] = title
             item['rentway'] = rentway
             item['price'] = price
             item['housetype'] = housetype
             item['area'] = area
             item['address'] = address
             item['traffic'] = traffic
             yield item

5、设置setting.py文件，配置scrapy运行的相关内容

 # 指定使用scrapy-redis的调度器
 SCHEDULER = "scrapy_redis.scheduler.Scheduler"

 # 指定使用scrapy-redis的去重
 DUPEFILTER_CLASS = 'scrapy_redis.dupefilter.RFPDupeFilter'

 # 指定排序爬取地址时使用的队列，
 # 默认的 按优先级排序(Scrapy默认)，由sorted set实现的一种非FIFO、LIFO方式。
 SCHEDULER_QUEUE_CLASS = 'scrapy_redis.queue.SpiderPriorityQueue'

 REDIS_HOST = '10.8.153.73'
 REDIS_PORT = 6379
 # 是否在关闭时候保留原来的调度器和去重记录，True=保留，False=清空
 SCHEDULER_PERSIST = True                                            

6、然后把代码发给其他附属机器,分别启动.子程序redis链接主服务器redis。

 redis-cli   -h  主服务器ip

7、主服务器先启动redis-server，再启动redis-cli

 lpush homespider:start_urls   起始的url