python爬虫之scrapy文件下载

 我们在写普通脚本的时候，从一个网站拿到一个文件的下载url，然后下载，直接将数据写入文件或者保存下来，但是这个需要我们自己一点一点的写出来，而且反复利用率并不高，为了不重复造轮子，scrapy提供很流畅的下载文件方式，只需要随便写写便可用了。

 mat.py文件

# -*- coding: utf-8 -*-
import scrapy
from scrapy.linkextractor import LinkExtractor
from weidashang.items import matplotlib

class MatSpider(scrapy.Spider):
    name = "mat"
    allowed_domains = ["matplotlib.org"]
    start_urls = ['https://matplotlib.org/examples']

    def parse(self, response):
　　　　　　　#抓取每个脚本文件的访问页面，拿到后下载
        link = LinkExtractor(restrict_css='div.toctree-wrapper.compound li.toctree-l2')
        for link in link.extract_links(response):
            yield scrapy.Request(url=link.url,callback=self.example)

    def example(self,response):
　　　　　　#进入每个脚本的页面，抓取源码文件按钮，并和base_url结合起来形成一个完整的url
        href = response.css('a.reference.external::attr(href)').extract_first()
        url = response.urljoin(href)
        example = matplotlib()
        example['file_urls'] = [url]
        return example

pipelines.py

class MyFilePlipeline(FilesPipeline):
    def file_path(self, request, response=None, info=None):
        path = urlparse(request.url).path
        return join(basename(dirname(path)),basename(path))

settings.py

ITEM_PIPELINES = {
   'weidashang.pipelines.MyFilePlipeline': 1,
}
FILES_STORE = 'examples_src'

items.py

class matplotlib(Item):
    file_urls = Field()
    files = Field()

 run.py

from scrapy.cmdline import execute
execute(['scrapy', 'crawl', 'mat','-o','example.json'])