一奕扣、CrawlSpider模板

• 創(chuàng)建項(xiàng)目

scrapy startproject 項(xiàng)目名稱

• 查看模板

scrapy genspider -l

• 創(chuàng)建crawl模板

scrapy genspider -t crawl 爬蟲名稱 地址

二、Spider爬蟲

# -*- coding: utf-8 -*-
import scrapy
from scrapy.linkextractors import LinkExtractor
# 導(dǎo)入LinkExtractor用于提取鏈接
from scrapy.spiders import CrawlSpider, Rule
# Rule定義一個規(guī)則扁达，然后讓LinkExtractor取根據(jù)這些規(guī)則提取url

from CrawlSpiderDemo.items import CrawlspiderdemoItem

# 在scrapy框架中包了兩個分類的爬蟲分別是：Spider（基本爬蟲）和CrawlSpider（增量模板爬蟲）
# CrawlSpider是Spider的一個派生類正卧，spider類設(shè)計原則只從start_urls列表中提取內(nèi)容，CrawlSpider定義了一些規(guī)則跪解，這些規(guī)則可以跟蹤鏈接炉旷，從而可以使得一個頁面中所有的符合規(guī)則的鏈接都被提取出來放入調(diào)度器中
# 在不斷訪問url的過程中，爬蟲匹配到的url越來越多

class DushuSpider(CrawlSpider):
    name = 'dushu'
    allowed_domains = ['dushu.com']
    start_urls = ['https://www.dushu.com/book/1002.html']

    rules = (
        Rule(LinkExtractor(allow=r'/book/1002_\d+\.html'), callback='parse_item', follow=True),
    )
    # rules 規(guī)則: 包含若干個Rule對象叉讥，每一個Rule對象對我們爬取網(wǎng)站的規(guī)則都做了一些特定的操作窘行，根據(jù)LinkExtractor里面的規(guī)則提取出所有的鏈接，然后把這些鏈接通過引擎壓入調(diào)度器的調(diào)度隊(duì)列中图仓，調(diào)度器進(jìn)而去調(diào)度下載罐盔，然后回調(diào)parse_item  (這里的回調(diào)方法寫成了字符串形式) ，再從二次請求的這些url對應(yīng)的頁面中根據(jù)LinkExtractor的規(guī)則繼續(xù)匹配（如果有重復(fù)救崔，自動剔除）翘骂，依次類推壁熄，直到匹配到所有的頁面

    # LinkExtractor的匹配規(guī)則：
    # 用正則表達(dá)式來匹配：LinkExtractor(allow="某正則") # /book/1002_\d\.html
    # 用xpath匹配：LinkExtractor(restrict_xpath="某xpath路徑")
    # 用css選擇器：LinkExtractor(restrict_css="某css選擇器")

    def parse_item(self, response):
        print(response.url)
        # 解析頁面
        book_list = response.xpath("http://div[@class='bookslist']//li")
        for book in book_list:
            item = CrawlspiderdemoItem()
            
            item["book_name"] = book.xpath(".//h3/a/text()").extract_first()

            # 獲取到二級頁面的url
            next_url = "https://www.dushu.com" + book.xpath(".//h3/a/@href").extract_first()

            yield scrapy.Request(url=next_url,callback=self.parse_next,meta={"item":item})

    def parse_next(self, response):
        item = response.meta["item"]
        item["price"] = response.xpath("http://span[@class='num']/text()").extract_first()
        m = response.xpath("http://div[@class='text txtsummary']")[2]
        item["mulu"] = m.xpath(".//text()").extract()

        yield item

使用xpath或其他規(guī)則匹配下來的所有節(jié)點(diǎn)，返回的類型是列表類型

.extract()方法是提取它的內(nèi)容

.extract_first()方法是提取列表第一個內(nèi)容碳竟，若列表為空返回空草丧，而不會報錯

三、Ip代理設(shè)置

• settings.py 設(shè)置

IPPOOL = [
    {'ip':'113.16.160.101:8118'},
    {'ip':'119.31.210.170:7777'},
    {'ip':'183.129.207.83:10800'},
    # {'ip':''},
    # {'ip':''},
    # {'ip':''},
    # {'ip':''},
    # {'ip':''},
    ]
    
#   下載中間件設(shè)置    
DOWNLOADER_MIDDLEWARES = {
    'scrapy.downloadermiddlewares.httpproxy.HttpProxyMiddleware':123,
   'IpAgent.middlewares.IPPOOLS': 125,
}

定義一個字段莹桅，表示我們收集好的代理

• middlewares.py

# 從settings文件中導(dǎo)入IPPOOL
import random

from .settings import IPPOOL
# 導(dǎo)入官方文檔對應(yīng)的HttpProxyMiddleware
from scrapy.contrib.downloadermiddleware.httpproxy import HttpProxyMiddleware

# 創(chuàng)建一個代理中間件類集成自官方代理中間件
class IPPOOLS(HttpProxyMiddleware):

    # 重寫初始化方法
    def __init__(self,ip=''):
        self.ip = ip

    # 重寫請求處理方法
    def process_request(self, request, spider):
        # 從ip代理池中隨機(jī)挑選一個ip地址
        current_ip = random.choice(IPPOOL)
        print('當(dāng)前ip是：',current_ip['ip'])

        # 設(shè)置請求對象的代理服務(wù)器是當(dāng)前ip
        request.meta['proxy'] = 'https://' + current_ip['ip']
        # 此時就可以把代理ip植入到下載器中

四昌执、動態(tài)頁面請求之selenium

• settings.py設(shè)置

# 下載中間件設(shè)置
DOWNLOADER_MIDDLEWARES = {
   'Toutiao.middlewares.ToutiaoDownloaderMiddleware': 543,
    'scrapy.downloadermiddlewares.useragent.UserAgentMiddleware':None,
}

• middlewares.py設(shè)置

from scrapy import signals
from selenium import webdriver
from time import sleep

from scrapy.http import HtmlResponse

class ToutiaoDownloaderMiddleware(object):
    # Not all methods need to be defined. If a method is not defined,
    # scrapy acts as if the downloader middleware does not modify the
    # passed objects.

    @classmethod
    def from_crawler(cls, crawler):
        # This method is used by Scrapy to create your spiders.
        s = cls()
        crawler.signals.connect(s.spider_opened, signal=signals.spider_opened)
        return s

    def process_request(self, request, spider):
        

        # 創(chuàng)建一個webdriver對象
        opt = webdriver.ChromeOptions()
        opt.add_argument("--headless")
        driver = webdriver.Chrome(options=opt)
        driver.get(request.url)
        sleep(3)
        # 讓頁面滾動
        js = "var q = document.documentElement.scrollTop=%d"
        distance = 100
        for i in range(100):
            driver.execute_script(js%distance)
            distance += 100
            sleep(0.5)
        body = driver.page_source
        print("正在使用中間件下載...")
        print("當(dāng)前瀏覽器正在訪問的網(wǎng)址是：",driver.current_url)
        # 響應(yīng)體需要重新定義
        res = HtmlResponse(url=driver.current_url,body=body,encoding='utf-8',request=request)

        return res

    def process_response(self, request, response, spider):
        # Called with the response returned from the downloader.

        # Must either;
        # - return a Response object
        # - return a Request object
        # - or raise IgnoreRequest
        return response

    def process_exception(self, request, exception, spider):
        # Called when a download handler or a process_request()
        # (from other downloader middleware) raises an exception.

        # Must either:
        # - return None: continue processing this exception
        # - return a Response object: stops process_exception() chain
        # - return a Request object: stops process_exception() chain
        pass

    def spider_opened(self, spider):
        spider.logger.info('Spider opened: %s' % spider.name)

五、基本分布式爬蟲部署redis儲存

scrapy_redis組件

pip install scrapy_redis

1诈泼、scrapy和scrapy_redis的區(qū)別

    scrapy是一個通用的爬蟲框架懂拾，不支持分布式

    scrapy_redis就是為實(shí)現(xiàn)scrapy的分布式而誕生的，它里面提功了redis的組件铐达，通過這些redis組件岖赋，就可以實(shí)現(xiàn)分布式
2、部署分布式

服務(wù)器端（master端）：

可以用某一臺主機(jī)作為redis服務(wù)器的運(yùn)行方（即服務(wù)端）瓮孙，也稱為master

客戶端（slaver端）：

1）把普通爬蟲修改成分布式唐断，去掉start_urls(不讓slaver隨意的執(zhí)行)，替換成redis_key（為了讓master能夠控制slaver的爬去）

• settings.py 設(shè)置

配置管道中間件

ITEM_PIPELINES = {
    # 分布式的爬蟲的數(shù)據(jù)可以不通過本地的管道（數(shù)據(jù)不需要往本地存）杭抠，數(shù)據(jù)需要存在redis數(shù)據(jù)庫中脸甘，在這里需要加入一個redis數(shù)據(jù)庫的管道組件
    "scrapy_redis.pipelines.RedisPipeline": 400
}

# 指定Redis數(shù)據(jù)庫相關(guān)配置
# Redis的主機(jī)地址
REDIS_HOST = "134.175.114.102"
# 端口號
REDIS_PORT = 6379
# 密碼
# REDIS_PARAMS = {"password":'xxxx'}

# 1、調(diào)度器需要切換成Scrapy_Redis的調(diào)度器（這個調(diào)度器是Scrapy_Redis組件對scrapy原生調(diào)度器的重寫偏灿，加入一些分布式調(diào)度的算法）
SCHEDULER = "scrapy_redis.scheduler.Scheduler"
# 2丹诀、加入scrapy_redis的去重組件
DUPEFILTER_CLASS = "scrapy_redis.dupefilter.RFPDupeFilter"
# 3、爬取過程中是否允許暫停
SCHEDULER_PERSIST = True

• spider設(shè)置

from scrapy_redis.spiders import RedisCrawlSpider


class ReadbookSpider(RedisCrawlSpider): # 注意繼承RedisCrawlSpider
    name = 'Readbook'
    allowed_domains = ['www.dushu.com']
    # start_urls = ['http://www.dushu.com/book/1002.html']
    # start_urls = ['https://www.dushu.com/book/1002.html'] # 分布式的爬蟲所有的url都是從redis數(shù)據(jù)庫的相關(guān)鍵下面提取

    # redis_key這個屬性指定了分布式爬蟲在獲取url的時候從哪些鍵中獲取的
    redis_key = "dushu:start_urls"

    rules = (
        Rule(LinkExtractor(allow=r'/book/1002_?\d*\.html'), callback='parse_item', follow=True),
    )