Scraping using the Scrapy framework

First you have to set up a new Scrapy project. Enter a directory where you’d like to store your code and run:

scrapy startproject projectName

To scrape we need a spider. Spiders define how a certain site will be scraped. Here’s the code for a spider that follows the links to the top voted questions on StackOverflow and scrapes some data from each page (source):

import scrapy

class StackOverflowSpider(scrapy.Spider):
    name = 'stackoverflow'  # each spider has a unique name
    start_urls = ['http://stackoverflow.com/questions?sort=votes']  # the parsing starts from a specific set of urls

    def parse(self, response):  # for each request this generator yields, its response is sent to parse_question
        for href in response.css('.question-summary h3 a::attr(href)'):  # do some scraping stuff using css selectors to find question urls 
            full_url = response.urljoin(href.extract())
            yield scrapy.Request(full_url, callback=self.parse_question)

    def parse_question(self, response): 
        yield {
            'title': response.css('h1 a::text').extract_first(),
            'votes': response.css('.question .vote-count-post::text').extract_first(),
            'body': response.css('.question .post-text').extract_first(),
            'tags': response.css('.question .post-tag::text').extract(),
            'link': response.url,
        }

Save your spider classes in the projectName\spiders directory. In this case - projectName\spiders\stackoverflow_spider.py.

Now you can use your spider. For example, try running (in the project’s directory):

scrapy crawl stackoverflow

Found a mistake? Have a question or improvement idea? Let me know.

Table Of Contents