经过三天的“摸爬滚打”，终于搞定了一个简单的爬虫项目，因为个人初学爬虫，没有一个系统的框架很难完整爬一个项目，所以参照诸多教程与博客，终于拿下一个简单的爬取“伯乐在线”所有文章的爬虫。

1、准备工作——安装scrapy框架

Command "python setup.py egg_info" failed with error code 1 in

2、开始爬取——新建scrapy项目

1. scrapy startproject article 
   文件架构
   
   
2. 开始编写爬虫代码
   

   # -*- coding: utf-8 -*-
   import scrapy
   from scrapy import Spider
   from ..items import ArticleItem
   
   class ArticleSpider(scrapy.Spider):
       name = 'article'
       start_urls = ['http://python.jobbole.com/all-posts/']
       def parse(self, response):
           item = ArticleItem()
           posts = response.xpath('//div[@class="post floated-thumb"]')
           # print(posts)
           for post in posts:
               item['title'] = post.xpath('.//a[@class="archive-title"]/text()').extract()[0]
               # print(item['title'])
               item['date'] = post.xpath('.//div[@class="post-meta"]/p/text()').re('d+/+d+/+d+')[0]
               # print(item['date'])
               item['short'] = post.xpath('.//span[@class="excerpt"]/p/text()').extract()[0]
               # print(item['short'])
               item['link'] = post.xpath('.//span[@class="read-more"]/a/@href').extract()[0]
               print(item['title']+item['date']+item['short']+item['link'])
               yield item
           urls = response.xpath('//a[@class="next page-numbers"]/@href').extract()[0]
           if urls:
               yield scrapy.Request(urls, callback=self.parse)

3. 存入excel
   piprlines.py
   

   from openpyxl import Workbook
   
   class TuniuPipeline(object):  # 设置工序一
       wb = Workbook()
       ws = wb.active
       ws.append(['标题', '链接', '发布时间', '简介'])  # 设置表头
   
   
       def process_item(self, item, spider):  # 工序具体内容
           line = [item['title'], item['link'], item['date'], item['short']]  # 把数据中每一项整理出来
           self.ws.append(line)  # 将数据以行的形式添加到xlsx中
           self.wb.save('article.xlsx')  # 保存xlsx文件
           return item

   settings.py
   

   ITEM_PIPELINES = {
       'article.pipelines.TuniuPipeline': 200,  # 200是为了设置工序顺序
   }

4. 启动爬虫 scrapy crawl article

5. 结果