Python爬虫框架Scrapy实例（二）

目标任务：使用Scrapy框架爬取新浪网导航页所有大类、小类、小类里的子链接、以及子链接页面的新闻内容，最后保存到本地。

大类小类如下图所示：

点击国内这个小类，进入页面后效果如下图（部分截图）：

查看页面元素，得到小类里的子链接如下图所示：

有子链接就可以发送请求来访问对应新闻的内容了。

 

首先创建scrapy项目

# 创建项目
scrapy startproject sinaNews
# 创建爬虫
scrapy genspider sina "sina.com.cn"

 

一、根据要爬取的字段创建item文件：

# -*- coding: utf-8 -*-
import scrapy

import sys

reload(sys)

sys.setdefaultencoding("utf-8")
class SinanewsItem(scrapy.Item):

# 大类的标题和url

parentTitle = scrapy.Field()

parentUrls = scrapy.Field()
</span><span style="color: #008000">#</span><span style="color: #008000"> 小类的标题和子url</span>
subTitle =<span style="color: #000000"> scrapy.Field()
subUrls </span>=<span style="color: #000000"> scrapy.Field()

</span><span style="color: #008000">#</span><span style="color: #008000"> 小类目录存储路径</span>
subFilename =<span style="color: #000000"> scrapy.Field()

</span><span style="color: #008000">#</span><span style="color: #008000"> 小类下的子链接</span>
sonUrls =<span style="color: #000000"> scrapy.Field()

</span><span style="color: #008000">#</span><span style="color: #008000"> 文章标题和内容</span>
head =<span style="color: #000000"> scrapy.Field()
content </span>= scrapy.Field()</pre>

二、编写spiders爬虫文件

# -*- coding: utf-8 -*-
import scrapy

import os

from sinaNews.items import SinanewsItem

import sys

reload(sys)

sys.setdefaultencoding("utf-8")
class SinaSpider(scrapy.Spider):

name = "sina"

allowed_domains = ["sina.com.cn"]

start_urls = ['http://news.sina.com.cn/guide/']
</span><span style="color: #0000ff">def</span><span style="color: #000000"> parse(self, response):
    items</span>=<span style="color: #000000"> []
    </span><span style="color: #008000">#</span><span style="color: #008000"> 所有大类的url 和 标题</span>
    parentUrls = response.xpath(<span style="color: #800000">'</span><span style="color: #800000">//div[@id="tab01"]/div/h3/a/@href</span><span style="color: #800000">'</span><span style="color: #000000">).extract()
    parentTitle </span>= response.xpath(<span style="color: #800000">'</span><span style="color: #800000">//div[@id="tab01"]/div/h3/a/text()</span><span style="color: #800000">'</span><span style="color: #000000">).extract()

    </span><span style="color: #008000">#</span><span style="color: #008000"> 所有小类的ur 和 标题</span>
    subUrls  = response.xpath(<span style="color: #800000">'</span><span style="color: #800000">//div[@id="tab01"]/div/ul/li/a/@href</span><span style="color: #800000">'</span><span style="color: #000000">).extract()
    subTitle </span>= response.xpath(<span style="color: #800000">'</span><span style="color: #800000">//div[@id="tab01"]/div/ul/li/a/text()</span><span style="color: #800000">'</span><span style="color: #000000">).extract()

    </span><span style="color: #008000">#</span><span style="color: #008000">爬取所有大类</span>
    <span style="color: #0000ff">for</span> i <span style="color: #0000ff">in</span><span style="color: #000000"> range(0, len(parentTitle)):
        </span><span style="color: #008000">#</span><span style="color: #008000"> 指定大类目录的路径和目录名</span>
        parentFilename = <span style="color: #800000">"</span><span style="color: #800000">./Data/</span><span style="color: #800000">"</span> +<span style="color: #000000"> parentTitle[i]

        </span><span style="color: #008000">#</span><span style="color: #008000">如果目录不存在，则创建目录</span>
        <span style="color: #0000ff">if</span>(<span style="color: #0000ff">not</span><span style="color: #000000"> os.path.exists(parentFilename)):
            os.makedirs(parentFilename)

        </span><span style="color: #008000">#</span><span style="color: #008000"> 爬取所有小类</span>
        <span style="color: #0000ff">for</span> j <span style="color: #0000ff">in</span><span style="color: #000000"> range(0, len(subUrls)):
            item </span>=<span style="color: #000000"> SinanewsItem()

            </span><span style="color: #008000">#</span><span style="color: #008000"> 保存大类的title和urls</span>
            item[<span style="color: #800000">'</span><span style="color: #800000">parentTitle</span><span style="color: #800000">'</span>] =<span style="color: #000000"> parentTitle[i]
            item[</span><span style="color: #800000">'</span><span style="color: #800000">parentUrls</span><span style="color: #800000">'</span>] =<span style="color: #000000"> parentUrls[i]

            </span><span style="color: #008000">#</span><span style="color: #008000"> 检查小类的url是否以同类别大类url开头，如果是返回True (sports.sina.com.cn 和 sports.sina.com.cn/nba)</span>
            if_belong = subUrls[j].startswith(item[<span style="color: #800000">'</span><span style="color: #800000">parentUrls</span><span style="color: #800000">'</span><span style="color: #000000">])

            </span><span style="color: #008000">#</span><span style="color: #008000"> 如果属于本大类，将存储目录放在本大类目录下</span>
            <span style="color: #0000ff">if</span><span style="color: #000000">(if_belong):
                subFilename </span>=parentFilename + <span style="color: #800000">'</span><span style="color: #800000">/</span><span style="color: #800000">'</span>+<span style="color: #000000"> subTitle[j]
                </span><span style="color: #008000">#</span><span style="color: #008000"> 如果目录不存在，则创建目录</span>
                <span style="color: #0000ff">if</span>(<span style="color: #0000ff">not</span><span style="color: #000000"> os.path.exists(subFilename)):
                    os.makedirs(subFilename)

                </span><span style="color: #008000">#</span><span style="color: #008000"> 存储 小类url、title和filename字段数据</span>
                item[<span style="color: #800000">'</span><span style="color: #800000">subUrls</span><span style="color: #800000">'</span>] =<span style="color: #000000"> subUrls[j]
                item[</span><span style="color: #800000">'</span><span style="color: #800000">subTitle</span><span style="color: #800000">'</span>] =<span style="color: #000000">subTitle[j]
                item[</span><span style="color: #800000">'</span><span style="color: #800000">subFilename</span><span style="color: #800000">'</span>] =<span style="color: #000000"> subFilename

                items.append(item)

    </span><span style="color: #008000">#</span><span style="color: #008000">发送每个小类url的Request请求，得到Response连同包含meta数据 一同交给回调函数 second_parse 方法处理</span>
    <span style="color: #0000ff">for</span> item <span style="color: #0000ff">in</span><span style="color: #000000"> items:
        </span><span style="color: #0000ff">yield</span> scrapy.Request( url = item[<span style="color: #800000">'</span><span style="color: #800000">subUrls</span><span style="color: #800000">'</span>], meta={<span style="color: #800000">'</span><span style="color: #800000">meta_1</span><span style="color: #800000">'</span>: item}, callback=<span style="color: #000000">self.second_parse)

</span><span style="color: #008000">#</span><span style="color: #008000">对于返回的小类的url，再进行递归请求</span>
<span style="color: #0000ff">def</span><span style="color: #000000"> second_parse(self, response):
    </span><span style="color: #008000">#</span><span style="color: #008000"> 提取每次Response的meta数据</span>
    meta_1= response.meta[<span style="color: #800000">'</span><span style="color: #800000">meta_1</span><span style="color: #800000">'</span><span style="color: #000000">]

    </span><span style="color: #008000">#</span><span style="color: #008000"> 取出小类里所有子链接</span>
    sonUrls = response.xpath(<span style="color: #800000">'</span><span style="color: #800000">//a/@href</span><span style="color: #800000">'</span><span style="color: #000000">).extract()

    items</span>=<span style="color: #000000"> []
    </span><span style="color: #0000ff">for</span> i <span style="color: #0000ff">in</span><span style="color: #000000"> range(0, len(sonUrls)):
        </span><span style="color: #008000">#</span><span style="color: #008000"> 检查每个链接是否以大类url开头、以.shtml结尾，如果是返回True</span>
        if_belong = sonUrls[i].endswith(<span style="color: #800000">'</span><span style="color: #800000">.shtml</span><span style="color: #800000">'</span>) <span style="color: #0000ff">and</span> sonUrls[i].startswith(meta_1[<span style="color: #800000">'</span><span style="color: #800000">parentUrls</span><span style="color: #800000">'</span><span style="color: #000000">])

        </span><span style="color: #008000">#</span><span style="color: #008000"> 如果属于本大类，获取字段值放在同一个item下便于传输</span>
        <span style="color: #0000ff">if</span><span style="color: #000000">(if_belong):
            item </span>=<span style="color: #000000"> SinanewsItem()
            item[</span><span style="color: #800000">'</span><span style="color: #800000">parentTitle</span><span style="color: #800000">'</span>] =meta_1[<span style="color: #800000">'</span><span style="color: #800000">parentTitle</span><span style="color: #800000">'</span><span style="color: #000000">]
            item[</span><span style="color: #800000">'</span><span style="color: #800000">parentUrls</span><span style="color: #800000">'</span>] =meta_1[<span style="color: #800000">'</span><span style="color: #800000">parentUrls</span><span style="color: #800000">'</span><span style="color: #000000">]
            item[</span><span style="color: #800000">'</span><span style="color: #800000">subUrls</span><span style="color: #800000">'</span>] = meta_1[<span style="color: #800000">'</span><span style="color: #800000">subUrls</span><span style="color: #800000">'</span><span style="color: #000000">]
            item[</span><span style="color: #800000">'</span><span style="color: #800000">subTitle</span><span style="color: #800000">'</span>] = meta_1[<span style="color: #800000">'</span><span style="color: #800000">subTitle</span><span style="color: #800000">'</span><span style="color: #000000">]
            item[</span><span style="color: #800000">'</span><span style="color: #800000">subFilename</span><span style="color: #800000">'</span>] = meta_1[<span style="color: #800000">'</span><span style="color: #800000">subFilename</span><span style="color: #800000">'</span><span style="color: #000000">]
            item[</span><span style="color: #800000">'</span><span style="color: #800000">sonUrls</span><span style="color: #800000">'</span>] =<span style="color: #000000"> sonUrls[i]
            items.append(item)

    </span><span style="color: #008000">#</span><span style="color: #008000">发送每个小类下子链接url的Request请求，得到Response后连同包含meta数据 一同交给回调函数 detail_parse 方法处理</span>
    <span style="color: #0000ff">for</span> item <span style="color: #0000ff">in</span><span style="color: #000000"> items:
            </span><span style="color: #0000ff">yield</span> scrapy.Request(url=item[<span style="color: #800000">'</span><span style="color: #800000">sonUrls</span><span style="color: #800000">'</span>], meta={<span style="color: #800000">'</span><span style="color: #800000">meta_2</span><span style="color: #800000">'</span>:item}, callback =<span style="color: #000000"> self.detail_parse)

</span><span style="color: #008000">#</span><span style="color: #008000"> 数据解析方法，获取文章标题和内容</span>
<span style="color: #0000ff">def</span><span style="color: #000000"> detail_parse(self, response):
    item </span>= response.meta[<span style="color: #800000">'</span><span style="color: #800000">meta_2</span><span style="color: #800000">'</span><span style="color: #000000">]
    content </span>= <span style="color: #800000">""</span><span style="color: #000000">
    head </span>= response.xpath(<span style="color: #800000">'</span><span style="color: #800000">//h1[@id="main_title"]/text()</span><span style="color: #800000">'</span><span style="color: #000000">)
    content_list </span>= response.xpath(<span style="color: #800000">'</span><span style="color: #800000">//div[@id="artibody"]/p/text()</span><span style="color: #800000">'</span><span style="color: #000000">).extract()

    </span><span style="color: #008000">#</span><span style="color: #008000"> 将p标签里的文本内容合并到一起</span>
    <span style="color: #0000ff">for</span> content_one <span style="color: #0000ff">in</span><span style="color: #000000"> content_list:
        content </span>+=<span style="color: #000000"> content_one

    item[</span><span style="color: #800000">'</span><span style="color: #800000">head</span><span style="color: #800000">'</span>]=<span style="color: #000000"> head
    item[</span><span style="color: #800000">'</span><span style="color: #800000">content</span><span style="color: #800000">'</span>]=<span style="color: #000000"> content

    </span><span style="color: #0000ff">yield</span> item</pre>

 

三、编写pipelines文件

# -*- coding: utf-8 -*-
from scrapy import signals

import sys

reload(sys)

sys.setdefaultencoding("utf-8")
class SinanewsPipeline(object):

def process_item(self, item, spider):

sonUrls = item['sonUrls']
    </span><span style="color: #008000">#</span><span style="color: #008000"> 文件名为子链接url中间部分，并将 / 替换为 _，保存为 .txt格式</span>
    filename = sonUrls[7:-6].replace(<span style="color: #800000">'</span><span style="color: #800000">/</span><span style="color: #800000">'</span>,<span style="color: #800000">'</span><span style="color: #800000">_</span><span style="color: #800000">'</span><span style="color: #000000">)
    filename </span>+= <span style="color: #800000">"</span><span style="color: #800000">.txt</span><span style="color: #800000">"</span><span style="color: #000000">

    fp </span>= open(item[<span style="color: #800000">'</span><span style="color: #800000">subFilename</span><span style="color: #800000">'</span>]+<span style="color: #800000">'</span><span style="color: #800000">/</span><span style="color: #800000">'</span>+filename, <span style="color: #800000">'</span><span style="color: #800000">w</span><span style="color: #800000">'</span><span style="color: #000000">)
    fp.write(item[</span><span style="color: #800000">'</span><span style="color: #800000">content</span><span style="color: #800000">'</span><span style="color: #000000">])
    fp.close()

    </span><span style="color: #0000ff">return</span> item</pre>

四、settings文件的设置

# 设置管道文件
ITEM_PIPELINES = {
   'sinaNews.pipelines.SinanewsPipeline': 300,
}

 执行命令

scrapy crwal sina

效果如下图所示：

打开工作目录下的Data目录，显示大类文件夹

 大开一个大类文件夹，显示小类文件夹：

打开一个小类文件夹，显示文章：