11.4. 爬蟲專案

weixin_33766168發表於2018-01-11

11.4.1. 建立專案

建立爬蟲專案

scrapy startproject project

在抓取之前，你需要新建一個Scrapy工程

			
neo@MacBook-Pro ~/Documents % scrapy startproject crawler 
New Scrapy project 'crawler', using template directory '/usr/local/lib/python3.6/site-packages/scrapy/templates/project', created in:
    /Users/neo/Documents/crawler

You can start your first spider with:
    cd crawler
    scrapy genspider example example.com

neo@MacBook-Pro ~/Documents % cd crawler 
neo@MacBook-Pro ~/Documents/crawler % find .
.
./crawler
./crawler/__init__.py
./crawler/__pycache__
./crawler/items.py
./crawler/middlewares.py
./crawler/pipelines.py
./crawler/settings.py
./crawler/spiders
./crawler/spiders/__init__.py
./crawler/spiders/__pycache__
./scrapy.cfg

Scrapy 工程目錄主要有以下檔案組成：

scrapy.cfg: 專案配置檔案
middlewares.py : 專案 middlewares 檔案
items.py: 專案items檔案
pipelines.py: 專案管道檔案
settings.py: 專案配置檔案
spiders: 放置spider的目錄

11.4.2. Spider

建立爬蟲，名字是 netkiller, 爬行的地址是 netkiller.cn

			
neo@MacBook-Pro ~/Documents/crawler % scrapy genspider netkiller netkiller.cn
Created spider 'netkiller' using template 'basic' in module:
  crawler.spiders.netkiller
neo@MacBook-Pro ~/Documents/crawler % find .
.
./crawler
./crawler/__init__.py
./crawler/__pycache__
./crawler/__pycache__/__init__.cpython-36.pyc
./crawler/__pycache__/settings.cpython-36.pyc
./crawler/items.py
./crawler/middlewares.py
./crawler/pipelines.py
./crawler/settings.py
./crawler/spiders
./crawler/spiders/__init__.py
./crawler/spiders/__pycache__
./crawler/spiders/__pycache__/__init__.cpython-36.pyc
./crawler/spiders/netkiller.py
./scrapy.cfg

開啟 crawler/spiders/netkiller.py 檔案，修改內容如下

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


class NetkillerSpider(scrapy.Spider):
    name = 'netkiller'
    allowed_domains = ['netkiller.cn']
    start_urls = ['http://www.netkiller.cn/']

    def parse(self, response):
        for link in response.xpath('//div[@class="blockquote"]')[1].css('a.ulink'):
            # self.log('This url is %s' % link)
            yield {
                'name': link.css('a::text').extract(),
                'url': link.css('a.ulink::attr(href)').extract()
                }
            
        pass

執行爬蟲

			
neo@MacBook-Pro ~/Documents/crawler % scrapy crawl netkiller -o output.json
2017-09-08 11:42:30 [scrapy.utils.log] INFO: Scrapy 1.4.0 started (bot: crawler)
2017-09-08 11:42:30 [scrapy.utils.log] INFO: Overridden settings: {'BOT_NAME': 'crawler', 'FEED_FORMAT': 'json', 'FEED_URI': 'output.json', 'NEWSPIDER_MODULE': 'crawler.spiders', 'ROBOTSTXT_OBEY': True, 'SPIDER_MODULES': ['crawler.spiders']}
2017-09-08 11:42:30 [scrapy.middleware] INFO: Enabled extensions:
['scrapy.extensions.corestats.CoreStats',
 'scrapy.extensions.telnet.TelnetConsole',
 'scrapy.extensions.memusage.MemoryUsage',
 'scrapy.extensions.feedexport.FeedExporter',
 'scrapy.extensions.logstats.LogStats']
2017-09-08 11:42:30 [scrapy.middleware] INFO: Enabled downloader middlewares:
['scrapy.downloadermiddlewares.robotstxt.RobotsTxtMiddleware',
 'scrapy.downloadermiddlewares.httpauth.HttpAuthMiddleware',
 'scrapy.downloadermiddlewares.downloadtimeout.DownloadTimeoutMiddleware',
 'scrapy.downloadermiddlewares.defaultheaders.DefaultHeadersMiddleware',
 'scrapy.downloadermiddlewares.useragent.UserAgentMiddleware',
 'scrapy.downloadermiddlewares.retry.RetryMiddleware',
 'scrapy.downloadermiddlewares.redirect.MetaRefreshMiddleware',
 'scrapy.downloadermiddlewares.httpcompression.HttpCompressionMiddleware',
 'scrapy.downloadermiddlewares.redirect.RedirectMiddleware',
 'scrapy.downloadermiddlewares.cookies.CookiesMiddleware',
 'scrapy.downloadermiddlewares.httpproxy.HttpProxyMiddleware',
 'scrapy.downloadermiddlewares.stats.DownloaderStats']
2017-09-08 11:42:30 [scrapy.middleware] INFO: Enabled spider middlewares:
['scrapy.spidermiddlewares.httperror.HttpErrorMiddleware',
 'scrapy.spidermiddlewares.offsite.OffsiteMiddleware',
 'scrapy.spidermiddlewares.referer.RefererMiddleware',
 'scrapy.spidermiddlewares.urllength.UrlLengthMiddleware',
 'scrapy.spidermiddlewares.depth.DepthMiddleware']
2017-09-08 11:42:30 [scrapy.middleware] INFO: Enabled item pipelines:
[]
2017-09-08 11:42:30 [scrapy.core.engine] INFO: Spider opened
2017-09-08 11:42:30 [scrapy.extensions.logstats] INFO: Crawled 0 pages (at 0 pages/min), scraped 0 items (at 0 items/min)
2017-09-08 11:42:30 [scrapy.extensions.telnet] DEBUG: Telnet console listening on 127.0.0.1:6023
2017-09-08 11:42:30 [scrapy.core.engine] DEBUG: Crawled (200) <GET http://www.netkiller.cn/robots.txt> (referer: None)
2017-09-08 11:42:31 [scrapy.core.engine] DEBUG: Crawled (200) <GET http://www.netkiller.cn/> (referer: None)
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Architect 手札'], 'url': ['../architect/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Developer 手札'], 'url': ['../developer/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller PHP 手札'], 'url': ['../php/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Python 手札'], 'url': ['../python/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Testing 手札'], 'url': ['../testing/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Java 手札'], 'url': ['../java/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Cryptography 手札'], 'url': ['../cryptography/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Linux 手札'], 'url': ['../linux/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller FreeBSD 手札'], 'url': ['../freebsd/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Shell 手札'], 'url': ['../shell/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Security 手札'], 'url': ['../security/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Web 手札'], 'url': ['../www/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Monitoring 手札'], 'url': ['../monitoring/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Storage 手札'], 'url': ['../storage/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Mail 手札'], 'url': ['../mail/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Docbook 手札'], 'url': ['../docbook/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Project 手札'], 'url': ['../project/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Database 手札'], 'url': ['../database/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller PostgreSQL 手札'], 'url': ['../postgresql/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller MySQL 手札'], 'url': ['../mysql/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller NoSQL 手札'], 'url': ['../nosql/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller LDAP 手札'], 'url': ['../ldap/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Network 手札'], 'url': ['../network/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Cisco IOS 手札'], 'url': ['../cisco/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller H3C 手札'], 'url': ['../h3c/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Multimedia 手札'], 'url': ['../multimedia/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Perl 手札'], 'url': ['../perl/index.html']}
2017-09-08 11:42:31 [scrapy.core.scraper] DEBUG: Scraped from <200 http://www.netkiller.cn/>
{'name': ['Netkiller Amateur Radio 手札'], 'url': ['../radio/index.html']}
2017-09-08 11:42:31 [scrapy.core.engine] INFO: Closing spider (finished)
2017-09-08 11:42:31 [scrapy.extensions.feedexport] INFO: Stored json feed (28 items) in: output.json
2017-09-08 11:42:31 [scrapy.statscollectors] INFO: Dumping Scrapy stats:
{'downloader/request_bytes': 438,
 'downloader/request_count': 2,
 'downloader/request_method_count/GET': 2,
 'downloader/response_bytes': 6075,
 'downloader/response_count': 2,
 'downloader/response_status_count/200': 2,
 'finish_reason': 'finished',
 'finish_time': datetime.datetime(2017, 9, 8, 3, 42, 31, 157395),
 'item_scraped_count': 28,
 'log_count/DEBUG': 31,
 'log_count/INFO': 8,
 'memusage/max': 49434624,
 'memusage/startup': 49434624,
 'response_received_count': 2,
 'scheduler/dequeued': 1,
 'scheduler/dequeued/memory': 1,
 'scheduler/enqueued': 1,
 'scheduler/enqueued/memory': 1,
 'start_time': datetime.datetime(2017, 9, 8, 3, 42, 30, 931267)}
2017-09-08 11:42:31 [scrapy.core.engine] INFO: Spider closed (finished)

你會看到返回結果

			
{'name': ['Netkiller Architect 手札'], 'url': ['../architect/index.html']}

11.4.2.1. 翻頁操作

下面我們演示爬蟲翻頁，例如我們需要遍歷這部電子書《Netkiller Linux 手札》 https://netkiller.cn/linux/index.html，首先建立一個爬蟲任務

				
neo@MacBook-Pro ~/Documents/crawler % scrapy genspider book netkiller.cn
Created spider 'book' using template 'basic' in module:
  crawler.spiders.book

編輯爬蟲任務

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


class BookSpider(scrapy.Spider):
    name = 'book'
    allowed_domains = ['netkiller.cn']
    start_urls = ['https://netkiller.cn/linux/index.html']

    def parse(self, response):
        yield {'title': response.css('title::text').extract()}
        # 這裡取出下一頁連線地址
        next_page = response.xpath('//a[@accesskey="n"]/@href').extract_first() 
        self.log('Next page: %s' % next_page)
        # 如果頁面不為空交給 response.follow 來爬取這個頁面
        if next_page is not None:
            yield response.follow(next_page, callback=self.parse)    

        pass

11.4.2.2. 採集內容儲存到檔案

下面的例子是將 response.body 返回採集內容儲存到檔案中

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


class BookSpider(scrapy.Spider):
    name = 'book'
    allowed_domains = ['netkiller.cn']
    start_urls = ['https://netkiller.cn/linux/index.html']

    def parse(self, response):
        yield {'title': response.css('title::text').extract()}

        filename = '/tmp/%s' % response.url.split("/")[-1]

        with open(filename, 'wb') as f:
            f.write(response.body)
        self.log('Saved file %s' % filename)

        next_page = response.xpath('//a[@accesskey="n"]/@href').extract_first()
        self.log('Next page: %s' % next_page)
        if next_page is not None:
            yield response.follow(next_page, callback=self.parse)    

        pass

任務運維結束後檢視採集出來的檔案

				
neo@MacBook-Pro ~/Documents/crawler % ls /tmp/*.html
/tmp/apt-get.html            /tmp/disc.html               /tmp/infomation.html         /tmp/lspci.html              /tmp/svgatextmode.html
/tmp/aptitude.html           /tmp/dmidecode.html          /tmp/install.html            /tmp/lsscsi.html             /tmp/swap.html
/tmp/author.html             /tmp/do-release-upgrade.html /tmp/install.partition.html  /tmp/lsusb.html              /tmp/sys.html
/tmp/avlinux.html            /tmp/dpkg.html               /tmp/introduction.html       /tmp/package.html            /tmp/sysctl.html
/tmp/centos.html             /tmp/du.max-depth.html       /tmp/kernel.html             /tmp/pr01s02.html            /tmp/system.infomation.html
/tmp/cfdisk.html             /tmp/ethtool.html            /tmp/kernel.modules.html     /tmp/pr01s03.html            /tmp/system.profile.html
/tmp/console.html            /tmp/framebuffer.html        /tmp/kudzu.html              /tmp/pr01s05.html            /tmp/system.shutdown.html
/tmp/console.timeout.html    /tmp/gpt.html                /tmp/linux.html              /tmp/preface.html            /tmp/tune2fs.html
/tmp/dd.clone.html           /tmp/hdd.label.html          /tmp/locale.html             /tmp/proc.html               /tmp/udev.html
/tmp/deb.html                /tmp/hdd.partition.html      /tmp/loop.html               /tmp/rpm.html                /tmp/upgrades.html
/tmp/device.cpu.html         /tmp/hwinfo.html             /tmp/lsblk.html              /tmp/rpmbuild.html           /tmp/yum.html
/tmp/device.hba.html         /tmp/index.html              /tmp/lshw.html               /tmp/smartctl.html

這裡只是做演示，生產環境請不要在 parse(self, response) 中處理，後面會講到 Pipeline。

11.4.3. settings.py 爬蟲配置檔案

11.4.3.1. 忽略 robots.txt 規則

				
# Obey robots.txt rules
ROBOTSTXT_OBEY = False

11.4.4. Item

Item 在 scrapy 中的類似“實體”或者“POJO”的概念，是一個資料結構類。爬蟲通過ItemLoader將資料放到Item中

下面是 items.py 檔案

			
# -*- coding: utf-8 -*-

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

import scrapy


class CrawlerItem(scrapy.Item):
    # define the fields for your item here like:
    title = scrapy.Field()
    author = scrapy.Field()
    content = scrapy.Field()
    ctime = scrapy.Field()
    
    pass

下面是爬蟲檔案

			
# -*- coding: utf-8 -*-
import scrapy
from scrapy.loader import ItemLoader
from crawler.items import CrawlerItem 
import time

class ExampleSpider(scrapy.Spider):
    name = 'example'
    allowed_domains = ['netkiller.cn']
    start_urls = ['https://netkiller.cn/java/index.html']
    def parse(self, response):

        item_selector = response.xpath('//a/@href')
        for url in item_selector.extract():
            if 'html' in url.split('.'):
                url = response.urljoin(url)
                yield response.follow( url, callback=self.parse_item)

        next_page = response.xpath('//a[@accesskey="n"]/@href').extract_first()
        self.log('Next page: %s' % next_page)
        if next_page is not None:
            yield response.follow(next_page, callback=self.parse)   
        
    def parse_item(self, response):
        l = ItemLoader(item=CrawlerItem(), response=response)
        l.add_css('title', 'title::text')
        l.add_value('ctime', time.strftime( '%Y-%m-%d %X', time.localtime() ))
        l.add_value('content', response.body)
        return l.load_item()

yield response.follow( url, callback=self.parse_item) 會回撥 parse_item(self, response) 將爬到的資料放置到 Item 中

11.4.5. Pipeline

Pipeline 管道線，主要的功能是對 Item 的資料處理，例如計算、合併等等。通常我們在這裡做資料儲存。下面的例子是將爬到的資料儲存到 json 檔案中。

預設情況 Pipeline 是禁用的，首先我們需要開啟 Pipeline 支援，修改 settings.py 檔案，找到下面配置項，去掉註釋。

			
# Configure item pipelines
# See http://scrapy.readthedocs.org/en/latest/topics/item-pipeline.html
ITEM_PIPELINES = {
    'crawler.pipelines.CrawlerPipeline': 300,
}

修改 pipelines.py 檔案。

			
# -*- coding: utf-8 -*-

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

import json

class CrawlerPipeline(object):
    def open_spider(self, spider):
        self.file = open('items.json', 'w')

    def close_spider(self, spider):
        self.file.close()
    def process_item(self, item, spider):
        # self.log("PIPE: %s" % item)
        line = json.dumps(dict(item)) + "\n"
        self.file.write(line)   
        return item