zoukankan      html  css  js  c++  java
  • scrapy-splash抓取动态数据例子五

      一、介绍

        本例子用scrapy-splash抓取智能电视网网站给定关键字抓取咨询信息。

        给定关键字:打通;融合;电视

        抓取信息内如下:

          1、资讯标题

          2、资讯链接

          3、资讯时间

          4、资讯来源

      二、网站信息

        

        

        

      三、数据抓取

        针对上面的网站信息,来进行抓取

        1、首先抓取信息列表

          抓取代码:sels = site.xpath('//div[@class="listl list2"]/ul/li')

        2、抓取标题

          首先列表页面,根据标题和日期来判断是否自己需要的资讯,如果是,就今日到资讯对应的链接,来抓取来源,如果不是,就不用抓取了

          抓取代码:titles = sel.xpath('.//h3/a/text()')

        3、抓取链接

          抓取代码:url = 'http://news.znds.com' + str(sel.xpath('.//h3/a/@href')[0].extract())

        4、抓取日期

          抓取代码:strdate = str(sel.xpath('.//span/text()').extract())

        5、抓取来源

          抓取代码:sources = site.xpath('//span[@class="spanimg2"]/text()')

       

      四、完整代码

    # -*- coding: utf-8 -*-
    import scrapy
    from scrapy import Request
    from scrapy.spiders import Spider
    from scrapy_splash import SplashRequest
    from scrapy_splash import SplashMiddleware
    from scrapy.http import Request, HtmlResponse
    from scrapy.selector import Selector
    from scrapy_splash import SplashRequest
    from splash_test.items import SplashTestItem
    import IniFile
    import sys
    import os
    import re
    import time
    
    reload(sys)
    sys.setdefaultencoding('utf-8')
    
    # sys.stdout = open('output.txt', 'w')
    
    class zndsSpider(Spider):
        name = 'znds'
    
        configfile = os.path.join(os.getcwd(), 'splash_testspiderssetting.conf')
    
        cf = IniFile.ConfigFile(configfile)
        information_wordlist = cf.GetValue("section", "information_keywords").split(';')
        websearch_urls = cf.GetValue("znds", "websearchurl").split(';')
        start_urls = []
        for word in websearch_urls:
            start_urls.append(word)
    
        # request需要封装成SplashRequest
        def start_requests(self):
            for url in self.start_urls:
                yield SplashRequest(url
                                    , self.parse
                                    , args={'wait': '2'}
                                    )
    
        def Comapre_to_days(self,leftdate, rightdate):
            '''
            比较连个字符串日期,左边日期大于右边日期多少天
            :param leftdate: 格式:2017-04-15
            :param rightdate: 格式:2017-04-15
            :return: 天数
            '''
            l_time = time.mktime(time.strptime(leftdate, '%Y-%m-%d'))
            r_time = time.mktime(time.strptime(rightdate, '%Y-%m-%d'))
            result = int(l_time - r_time) / 86400
            return result
    
        def date_isValid(self, strDateText):
            '''
            判断日期时间字符串是否合法:如果给定时间大于当前时间是合法,或者说当前时间给定的范围内
            :param strDateText: 四种格式 '2小时前'; '2天前' ; '昨天' ;'2017.2.12 '
            :return: True:合法;False:不合法
            '''
            currentDate = time.strftime('%Y-%m-%d')
            datePattern = re.compile(r'd{4}-d{1,2}-d{1,2}')
            strDate = re.findall(datePattern, strDateText)
            if len(strDate) == 1:
                if self.Comapre_to_days(currentDate,strDate[0])==0:
                    return True,currentDate
            return False, ''
        def parse(self, response):
    
            site = Selector(response)
            # keyword = response.meta['keyword']
            sels = site.xpath('//div[@class="listl list2"]/ul/li')
            for sel in sels:
                titles = sel.xpath('.//h3/a/text()')
                if len(titles)>0:
                    title = str(titles[0].extract())
                    strdate = str(sel.xpath('.//span/text()').extract())
                    flag,date =self.date_isValid(strdate)
                    if flag:
                        url = 'http://news.znds.com' + str(sel.xpath('.//h3/a/@href')[0].extract())
                        for keyword in self.information_wordlist:
                            if title.find(keyword)>-1:yield SplashRequest(url
                                            , self.parse_item
                                            , args={'wait': '1'},
                                            meta={'date': date, 'url': url,
                                                      'keyword': keyword,'title':title}
                                                )
    
    
        def parse_item(self, response):
            site = Selector(response)
            it = SplashTestItem()
            it['title'] = response.meta['title']
            it['url'] = response.meta['url']
            it['date'] = response.meta['date']
            it['keyword'] = response.meta['keyword']
            sources = site.xpath('//span[@class="spanimg2"]/text()')
            if len(sources)>0:
                source =str(sources[0].extract()).replace(u'来源: ','')
                it['source'] = source
            return it
  • 相关阅读:
    Sicily 1153. 马的周游问题 解题报告
    回溯法与八皇后问题
    Sicily 1151. 魔板 解题报告
    Sicily 1176. Two Ends 解题报告
    Sicily 1046. Plane Spotting 解题报告
    Java多线程11:ReentrantLock的使用和Condition
    Java多线程10:ThreadLocal的作用及使用
    Java多线程9:ThreadLocal源码剖析
    Java多线程8:wait()和notify()/notifyAll()
    Java多线程7:死锁
  • 原文地址:https://www.cnblogs.com/shaosks/p/6963615.html
Copyright © 2011-2022 走看看