前言

本篇文章实现python的scrapy框架爬取全书网小说，scrapy框架的安装我在这里就不在赘述了，建议window用户使用anaconda安装，这里比较省心一些。运行环境python3（实际python2运行页没有太大问题）

一，项目的创建

在想要创建项目的目录内，打开cmd命令行，输出代码

scrapy startproject Fiction

然后进入项目内

cd Fiction

创建一个爬虫项目

scrapy genspider novel quanshuwang.com

完成后其项目目录是这样的

二，具体代码部分（关键部分已经给了注释）

1.items.py文件

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

class FictionItem(scrapy.Item): name = scrapy.Field() #小说名字 chapter_name = scrapy.Field() #小说章节名字 chapter_content = scrapy.Field() #小说章节内容

2.novel.py文件

# -*- coding: utf-8 -*- import scrapy import re from Fiction.items import FictionItem from scrapy.http import Request

class NovelSpider(scrapy.Spider): name = 'novel' allowed_domains = ['quanshuwang.com'] start_urls = [ 'http://www.quanshuwang.com/list/1_2.html', 'http://www.quanshuwang.com/list/1_3.html', 'http://www.quanshuwang.com/list/1_4.html', 'http://www.quanshuwang.com/list/1_5.html', 'http://www.quanshuwang.com/list/1_6.html', 'http://www.quanshuwang.com/list/1_7.html', 'http://www.quanshuwang.com/list/1_8.html', 'http://www.quanshuwang.com/list/1_9.html', 'http://www.quanshuwang.com/list/1_10.html', ] #全书网玄幻魔法类前10页 #获取每一本书的URL def parse(self, response): book_urls = response.xpath('//li/a[@class="l mr10"]/@href').extract() for book_url in book_urls: yield Request(book_url, callback=self.parse_read)

#获取马上阅读按钮的URL，进入章节目录 def parse_read(self, response): read_url = response.xpath('//a[@class="reader"]/@href').extract()[0] yield Request(read_url, callback=self.parse_chapter)

#获取小说章节的URL def parse_chapter(self, response): chapter_urls = response.xpath('//div[@class="clearfix dirconone"]/li/a/@href').extract() for chapter_url in chapter_urls: yield Request(chapter_url, callback=self.parse_content)

#获取小说名字,章节的名字和内容 def parse_content(self, response): #小说名字 name = response.xpath('//div[@class="main-index"]/a[@class="article_title"]/text()').extract_first()

result = response.text #小说章节名字 chapter_name = response.xpath('//strong[@class="l jieqi_title"]/text()').extract_first() #小说章节内容 chapter_content_reg = r'style5\(\);</script>(.*?)<script type="text/javascript">' chapter_content_2 = re.findall(chapter_content_reg, result, re.S)[0] chapter_content_1 = chapter_content_2.replace('    ', '') chapter_content = chapter_content_1.replace('<br />', '')

item = FictionItem() item['name'] = name item['chapter_name'] = chapter_name item['chapter_content'] = chapter_content yield item

3.pipelines.py文件

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

class FictionPipeline(object):

def process_item(self, item, spider): curPath = 'F:\全书网小说' tempPath = str(item['name']) targetPath = curPath + os.path.sep + tempPath if not os.path.exists(targetPath): os.makedirs(targetPath)

filename_path = 'F:\全书网小说'+ os.path.sep + str(item['name'])+ os.path.sep + str(item['chapter_name']) + '.txt' with open(filename_path, 'w', encoding='utf-8') as f: f.write(item['chapter_content'] + "\n") return item

4.settings.py文件

BOT_NAME = 'Fiction'

SPIDER_MODULES = ['Fiction.spiders'] NEWSPIDER_MODULE = 'Fiction.spiders' ROBOTSTXT_OBEY = False DEFAULT_REQUEST_HEADERS = { 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', 'Accept-Language': 'en', } ITEM_PIPELINES = { 'Fiction.pipelines.FictionPipeline': 300, }

三，总结

本篇文章只获取了全书网玄幻魔法类的前10页的小说，修改start_urls可以获取到剩余页面或者其他类别的小说。效果图如下：