python django scrapy 将项目返回到 Controller

Question

我需要做一些简短的实时抓取并将结果数据返回到我的 Django REST Controller 中。

尝试使用 scrapy:

import scrapy
from scrapy.selector import Selector

from . models import Product

class MysiteSpider(scrapy.Spider):
    name = "quotes"
    start_urls = [
        'https://www.something.com/browse?q=dfd',
    ]
    allowed_domains = ['something.com']

    def parse(self, response):
        items_list = Selector(response).xpath('//li[@itemprop="itemListElement"]')

        for value in items_list:
            item = Product()
            item['picture_url'] = value.xpath('img/@src').extract()[0]
            item['title'] = value.xpath('h2').text()
            item['price'] = value.xpath('p[contains(@class, "ad-price")]').text()
            yield item

项目模型

import scrapy


class Product(scrapy.Item):
    name = scrapy.Field()
    price = scrapy.Field()
    picture_url = scrapy.Field()
    published_date = scrapy.Field(serializer=str)

根据Scrapy的架构，items会被返回到Item Pipeline ( https://doc.scrapy.org/en/1.2/topics/item-pipeline.html ) 用于将数据存储到DB/保存到文件等。

但是我一直被这个问题所困扰——如何通过 Django REST APIview 返回已抓取的项目列表？

预期用法示例:

from rest_framework.views import APIView
from rest_framework.response import Response

from .service.mysite_spider import MysiteSpider

    class AggregatorView(APIView):
        mysite_spider = MysiteSpider()

        def get(self, request, *args, **kwargs):

            self.mysite_spider.parse()

            return Response('good')

Answer 1

我实际上并没有测试与 Django REST 框架的集成，但以下代码片段将允许您从 python 脚本运行 Spider，收集生成的项目以便稍后处理它们。

from scrapy import signals
from scrapy.crawler import Crawler, CrawlerProcess
from ... import MysiteSpider

items = []
def collect_items(item, response, spider):
    items.append(item)

crawler = Crawler(MysiteSpider)
crawler.signals.connect(collect_items, signals.item_scraped)

process = CrawlerProcess()
process.crawl(crawler)
process.start()  # the script will block here until the crawling is finished

# at this point, the "items" variable holds the scraped items 

郑重声明，这可行，但可能有更好的方法:-)

进一步阅读:

• https://doc.scrapy.org/en/latest/topics/practices.html#run-scrapy-from-a-script
• https://doc.scrapy.org/en/latest/topics/signals.html