1. 创建项目
scrapy startproject qiumeimei
2. 建蜘蛛文件qiumei.py
cd qiumeimei
scrapy genspider qiumei
3. 考虑到只需要下载图片，先在items.py定义字段
?123456import scrapy class QiumeimeiItem(scrapy.Item): # define the fields for your item here like: img_path = scrapy.Field() pass
4. 写蜘蛛文件qiumei.py
?12345678910111213141516171819202122# -- coding: utf-8 --import scrapy from qiumeimei.items import QiumeimeiItemclass QiumeiSpider(scrapy.Spider): name = 'qiumei' # allowed_domains = 【''】 start_urls = 【''】 def parse(self, response): img_url = response.css('.main>p>img::attr(data-lazy-src)').extract() # print(img_url) for url in img_url: # print(url) item = QiumeimeiItem() item【'img_path'】 = url yield item next_url = response.css('.pagination a.next::attr(href)').extract_first() if next_url: yield scrapy.Request(url=next_url,callback=self.parse)
5. 管道文件pipelines.py 这里图片是全部放在了一个文件夹里，在settings.py中定义了一个路径，见下文第6步：
?12345678910111213141516171819202122232425262728import os,scrapyfrom scrapy.pipelines.images import ImagesPipelinefrom qiumeimei.settings import IMAGES_STORE as images_storeclass QiumeimeiPipeline(ImagesPipeline): def get_media_requests(self, item, info): img_path = item【'img_path'】 # print(000) yield scrapy.Request(url=img_path) def item_completed(self, results, item, info): old_name_list = 【x【'path'】 for t, x in results】 old_name = images_store + old_name_list【0】 # print(111) #图片名称 from datetime import datetime i = str(datetime.now()) # print(222) img_path = item【'img_path'】 img_type = img_path.split('.')【-1】 img_name = i【:4】+i【5:7】+i【8:10】+i【11:13】+i【14:16】+i【17:19】+i【20:】 #图片路径 所有图片放在一个文件夹里 # print(333) path = images_store + img_name +'.'+ img_type print(path+' 已下载...') os.rename(old_name,path) return item
6. 设置文件settings.py
?1234567891011USER_AGENT = 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/65.0.3325.181 Safari/537.36'# Obey robots.txt rulesROBOTSTXT_OBEY = False #图片路径，会自动创建IMAGES_STORE = './images/' #开启管道ITEM_PIPELINES = { 'qiumeimei.pipelines.QiumeimeiPipeline': 300,}//代码效果参考：http://www.ezhiqi.com/bx/art_3651.html
已成功：
