zoukankan      html  css  js  c++  java
  • python Scrapy爬虫框架

    简介

    Scrapy是一个为了爬取网站数据,提取结构性数据而编写的应用框架。 其可以应用在数据挖掘,信息处理或存储历史数据等一系列的程序中。

    其最初是为了页面抓取 (更确切来说, 网络抓取 )所设计的, 也可以应用在获取API所返回的数据(例如 Amazon Associates Web Services ) 或者通用的网络爬虫。Scrapy用途广泛,可以用于数据挖掘、监测和自动化测试。

    Scrapy 使用了 Twisted异步网络库来处理网络通讯。整体架构大致如下:

    Scrapy组件

    l  引擎(Scrapy)

    用来处理整个系统的数据流处理, 触发事务(框架核心)

    l  调度器(Scheduler)

    用来接受引擎发过来的请求, 压入队列中, 并在引擎再次请求的时候返回. 可以想像成一个URL(抓取网页的网址或者说是链接)的优先队列, <br data-filtered="filtered">由它来决定下一个要抓取的网址是什么, 同时去除重复的网址

    l  下载器(Downloader)

    用于下载网页内容, 并将网页内容返回给蜘蛛(Scrapy下载器是建立在twisted这个高效的异步模型上的)

    l  爬虫(Spiders)

    爬虫是主要干活的, 用于从特定的网页中提取自己需要的信息, 即所谓的实体(Item)。用户也可以从中提取出链接,让Scrapy继续抓取下一个页面

    l  项目管道(Pipeline)

    负责处理爬虫从网页中抽取的实体,主要的功能是持久化实体、验证实体的有效性、清除不需要的信息。当页面被爬虫解析后,<br data-filtered="filtered">将被发送到项目管道,并经过几个特定的次序处理数据。

    l  下载器中间件(Downloader Middlewares)

    位于Scrapy引擎和下载器之间的框架,主要是处理Scrapy引擎与下载器之间的请求及响应。

    l  爬虫中间件(Spider Middlewares)

    介于Scrapy引擎和爬虫之间的框架,主要工作是处理蜘蛛的响应输入和请求输出。

    l  调度中间件(Scheduler Middewares)

    介于Scrapy引擎和调度之间的中间件,从Scrapy引擎发送到调度的请求和响应

    Scrapy运行流程

    1)        引擎从调度器中取出一个链接(URL)用于接下来的抓取

    2)        引擎把URL封装成一个请求(Request)传给下载器

    3)        下载器把资源下载下来,并封装成应答包(Response)

    4)        爬虫解析Response

    5)        解析出实体(Item),则交给实体管道进行进一步的处理

    6)     解析出的是链接(URL),则把URL交给调度器等待抓取

    安装Scrapy

    (1)依赖包

    xml, an efficient XML and HTML parser

    parsel, an HTML/XML data extraction library written on top of lxml,

    w3lib, a multi-purpose helper for dealing with URLs and web page encodings

    twisted, an asynchronous networking framework

    cryptography and pyOpenSSL, to deal with various network-level security needs

       (2)安装

                    1、安装wheel

            pip install wheel

                2、安装lxml

            https://pypi.python.org/pypi/lxml/4.1.0

                3、安装pyopenssl

            https://pypi.python.org/pypi/pyOpenSSL/17.5.0

                4、安装Twisted

            https://www.lfd.uci.edu/~gohlke/pythonlibs/

                5、安装pywin32

            https://sourceforge.net/projects/pywin32/files/

                6、安装scrapy

            pip install scrapy

    查看安装是否成功,输入命令

    scrapy version

     

    新建第一个项目

    新建tutorial工程

    scrapy startproject scrapytest

    scrapytest/

          scrapy.cfg          # deploy configuration file

          spiders/           # project's Python module, you'll import your code from here

            __init__.py

            items.py         # project items definition file

            middlewares.py   # project middlewares file

            pipelines.py     # project pipelines file

            settings.py      # project settings file

            spiders/         # a directory where you'll later put your spiders

             __init__.py

    进入scrapytest目录,使用命令创建一个基础爬虫类

    cd scrapytest

    scrapy genspider Mycninfo cninfo.com.cn

    编辑文件

    【scrapytest/scrapytest/spiders/Mycninfo.py】

     1 # -*- coding: utf-8 -*-
     2 import scrapy
     3 import json
     4 import os
     5 import time
     6 
     7 # scrapy crawl Mycninfo
     8 class MycninfoSpider(scrapy.Spider):
     9     name = 'Mycninfo'
    10     allowed_domains = ['cninfo.com.cn']
    11     start_urls = ['http://www.cninfo.com.cn/new/index']
    12 
    13     def parse(self, response):
    14         keyWord='002304'
    15         url='http://www.cninfo.com.cn/new/information/topSearch/query'
    16         yield scrapy.FormRequest(
    17             url = url,
    18             formdata = {"keyWord" : keyWord, "maxNum" : "10"},
    19             callback = self.parse_url_info
    20         )
    21         
    22     def parse_url_info(self,response):
    23         try:
    24             #print(response.text)
    25             data = eval(response.text)
    26             orgId = data[0].get("orgId")
    27             code = data[0].get("code")
    28             url = "http://www.cninfo.com.cn/new/disclosure/stock?stockCode={}&orgId={}".format(code,orgId)
    29             yield scrapy.Request(url=url,meta={"code":code,"orgId":orgId},callback=self.parse_page,dont_filter=True)
    30         except Exception as e:
    31             print(e)
    32             
    33     def parse_page(self,response):
    34         try:
    35             code = response.meta.get("code")
    36             orgId = response.meta.get("orgId")
    37             url="http://www.cninfo.com.cn/new/hisAnnouncement/query"
    38             plate = "sh" if code.startswith("60") else "sz"
    39             parm = {"sh":{"plate":"sh","column":"sse"},"sz":{"plate":"sz","column":"szse"}}
    40             query={
    41                 'stock': '{},{}'.format(code,orgId),
    42                 'tabName': 'fulltext',
    43                 'pageSize': '30',
    44                 'pageNum': '1',
    45                 'column': parm[plate]["column"],  
    46                 'category': 'category_ndbg_szsh',
    47                 'plate': parm[plate]["plate"],
    48                 'seDate': '',
    49                 'searchkey': '',
    50                 'secid': '',
    51                 'sortName': '',
    52                 'sortType': '',
    53                 'isHLtitle': 'true'
    54             }
    55             #print(query)
    56             yield scrapy.FormRequest(
    57                 url = url,
    58                 formdata = query,
    59                 callback = self.parse_ndbg
    60             )
    61         except Exception as e:
    62             print("e",e)
    63         
    64     def parse_ndbg(self,response):
    65         try:
    66             data = response.text
    67             data_json = json.loads(data)
    68             for data in data_json.get('announcements',[]):
    69                 name = data.get("secName")
    70                 announcementId = data.get("announcementId")
    71                 announcementTitle = data.get("announcementTitle")
    72                 announcementTime = data.get("announcementTime")
    73                 t = time.localtime(int(announcementTime)//1000)
    74                 date = time.strftime("%Y-%m-%d",t)
    75                 url = "http://www.cninfo.com.cn/new/announcement/download?bulletinId={}&announceTime={}".format(announcementId,date)
    76                 #print("{}_{}".format(name,announcementTitle),url)
    77                 data = {"name":name,"announcementTitle":announcementTitle,"url":url}
    78                 yield data
    79         except Exception as e:
    80             print(e)
    81         
    82         
             

    scrapytest/scrapytest/pipelines.py

     1 #!/usr/bin/env python3
     2 # -*- coding: utf-8 -*-
     3 import json
     4 from scrapy.pipelines.files import FilesPipeline
     5 from scrapy.exceptions import DropItem
     6 
     7 from crawler.items import *
     8 from crawler.DB import MysqlDB
     9 
    10 class DownloadPipeline(FilesPipeline):
    11     #下载文件
    12     def get_media_requests(self, item, info):       
    13         # meta里面的数据是从spider获取,然后通过meta传递给下面方法:file_path
    14         yield scrapy.Request(url=item['url'],meta={'text':item['name'],'name':item['announcementTitle']})
    15 
    16     def item_completed(self, results, item, info):
    17         # 是一个元组,第一个元素是布尔值表示是否成功
    18         if not results[0][0]:
    19             raise DropItem('下载失败')
    20         return item
    21     
    22     def file_path(self, request, response=None, info=None):
    23         # 重命名,若不重写这函数,图片名为哈希
    24         # 接收上面meta传递过来的图片名称
    25         name = request.meta['name']
    26         text = request.meta['text']
    27         # 提取url前面名称作为图片名
    28         # 分文件夹存储的关键:{0}对应着name;{1}对应着image_guid
    29         filename = u'{0}/{1}.pdf'.format(text, name)
    30         print(filename)
    31         return filename

    【scrapytest/scrapytest/settings.py】

     1 # -*- coding: utf-8 -*-
     2 
     3 BOT_NAME = 'crawler'
     4 
     5 SPIDER_MODULES = ['crawler.spiders']
     6 NEWSPIDER_MODULE = 'crawler.spiders'
     7 LOG_LEVEL= 'ERROR'
     8 
     9 # Crawl responsibly by identifying yourself (and your website) on the user-agent
    10 #USER_AGENT = 'crawler (+http://www.yourdomain.com)'
    11 
    12 # Obey robots.txt rules
    13 
    14 
    15 DOWNLOAD_DELAY = 1
    16 ROBOTSTXT_OBEY = False
    17 DEFAULT_REQUEST_HEADERS = {
    18    #'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
    19    'Accept':'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8',
    20    'Connection':'keep-alive',
    21    'User-Agent':'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36',
    22    'Accept-Language':'zh-CN,zh;q=0.9',
    23    
    24 }
    25 #下载文件存放的主目录
    26 FILES_STORE = 'data' 
    27 ITEM_PIPELINES = {
    28     'crawler.pipelines.DownloadPipeline': 300,
    29 }

     启动

     cd scrapy

    scrapy crawl Mycninfo

  • 相关阅读:
    SQLite无法使用drop column删除表字段解决办法
    Selenium之自动发送163邮件(转自https://www.cnblogs.com/zhang-da/p/12230415.html)
    frp 内网穿透
    smart.supor.com
    python colorama模块
    nohup和&后台运行,进程查看及终止
    Beef搭建并通过Nginx绑定域名
    How to Install Ruby on CentOS/RHEL 7/6
    Linux环境下安装python3
    启用IIS Express SSL(Https)的注意事项
  • 原文地址:https://www.cnblogs.com/boye169/p/13904822.html
Copyright © 2011-2022 走看看