【python】【scrapy】使用方法概要（四）

【请初学者作为参考，不建议高手看这个浪费时间】

上一篇文章，我们抓取到了一大批代理ip，本篇文章介绍如何实现downloaderMiddleware，达到随即使用代理ip对目标网站进行抓取的。

抓取的目标网站是现在炙手可热的旅游网站 www.qunar.com, 目标信息是qunar的所有seo页面，及页面的seo相关信息。

qunar并没有一般网站具有的 robots.txt文件，所以无法利用列表进行抓取，但是，可以发现，qunar的seo页面主要部署在

http://www.qunar.com/routes/  下，这个页面为入口文件，由此页面及此页面上所有带有routes的链接开始递归的抓取所有带有routes/字段的链接即可。

开始吧

目标信息为目标站点的seo信息，所以为head中的meta和description字段。

# Define here the models for your scraped items
#
# See documentation in:
# http://doc.scrapy.org/topics/items.html

from scrapy.item import Item, Field

class SitemapItem(Item):
    # define the fields for your item here like:
    # name = Field()
    url = Field()
    keywords = Field()
    description = Field()

因为要使用代理ip，所以需要实现自己的downloadermiddlerware，主要功能是从代理ip文件中随即选取一个ip端口作为代理服务，代码如下

import random

class ProxyMiddleware(object):
    def process_request(self, request, spider):
        fd = open('/home/xxx/services_runenv/crawlers/sitemap/sitemap/data/proxy_list.txt','r')
        data = fd.readlines()
        fd.close()
        length = len(data)
        index  = random.randint(0, length -1)
        item   = data[index]
        arr    = item.split(',')
        request.meta['proxy'] = 'http://%s:%s' % (arr[0],arr[1])

最重要的还是爬虫，主要功能是提取页面所有的链接，把满足条件的url实例成Request对象并yield， 同时提取页面的keywords，description信息，以item的形式yield，代码如下:

from scrapy.selector import HtmlXPathSelector
from sitemap.items import SitemapItem

import urllib
import simplejson
import exceptions
import pickle

class SitemapSpider(CrawlSpider):
    name = 'sitemap_spider'
    allowed_domains = ['qunar.com']
    start_urls = ['http://www.qunar.com/routes/']

    rules = (
        #Rule(SgmlLinkExtractor(allow=(r'http://www.qunar.com/routes/.*')), callback='parse'),
        #Rule(SgmlLinkExtractor(allow=('http:.*/routes/.*')), callback='parse'),
    )

    def parse(self, response):
        item = SitemapItem()
        x         = HtmlXPathSelector(response)
        raw_urls  = x.select("//a/@href").extract()
        urls      = []
        for url in raw_urls:
            if 'routes' in url:
                if 'http' not in url:
                    url = 'http://www.qunar.com' + url
                urls.append(url)

        for url in urls:
            yield Request(url)

        item['url']         = response.url.encode('UTF-8')
        arr_keywords        = x.select("//meta[@name='keywords']/@content").extract()
        item['keywords']    = arr_keywords[0].encode('UTF-8')
        arr_description     = x.select("//meta[@name='description']/@content").extract()
        item['description'] = arr_description[0].encode('UTF-8')

        yield item

pipe文件比较简单，只是把抓取到的数据存储起来，代码如下

# Define your item pipelines here
#
# Don't forget to add your pipeline to the ITEM_PIPELINES setting
# See: http://doc.scrapy.org/topics/item-pipeline.html

class SitemapPipeline(object):
    def process_item(self, item, spider):
        data_path = '/home/xxx/services_runenv/crawlers/sitemap/sitemap/data/output/sitemap_data.txt'
        fd = open(data_path, 'a')
        line = str(item['url']) + '#$#' + str(item['keywords']) + '#$#' + str(item['description']) + '\n'
        fd.write(line)
        fd.close
        return item

最后附上的是setting.py文件

# Scrapy settings for sitemap project
#
# For simplicity, this file contains only the most important settings by
# default. All the other settings are documented here:
#
#     http://doc.scrapy.org/topics/settings.html
#

BOT_NAME = 'sitemap hello,world~!'
BOT_VERSION = '1.0'

SPIDER_MODULES = ['sitemap.spiders']
NEWSPIDER_MODULE = 'sitemap.spiders'
USER_AGENT = '%s/%s' % (BOT_NAME, BOT_VERSION)

DOWNLOAD_DELAY = 0

ITEM_PIPELINES = [
    'sitemap.pipelines.SitemapPipeline'
]

DOWNLOADER_MIDDLEWARES = {
    'scrapy.contrib.downloadermiddleware.httpproxy.HttpProxyMiddleware': 110,
    'sitemap.middlewares.ProxyMiddleware': 100,
    }

CONCURRENT_ITEMS = 128
CONCURRENT_REQUEST = 64
CONCURRENT_REQUEST_PER_DOMAIN = 64


LOG_ENABLED = True
LOG_ENCODING = 'utf-8'
LOG_FILE = '/home/xxx/services_runenv/crawlers/sitemap/sitemap/log/sitemap.log'
LOG_LEVEL = 'DEBUG'
LOG_STDOUT = False

对scrapy的介绍将告一段落，更复杂的应用还没有接触过，想等看完redis的源码，再来研究下scrapy的源码~~ 希望通过分享能给正在入门scrapy的童鞋带来帮助~