Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
A cute HTML scraper / data extraction tool in under 70 lines of code
Ruby
tree: 9dad30b322

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
examples
LICENSE
README.md
Rakefile
scraper.rb

README.md

Scraper

Scraper is a cute HTML screen-scraping tool.

require 'scraper'
require 'open-uri'

class BlogScraper < Scraper
  element :title

  elements 'div.hentry' => :articles do
    element 'h2' => :title
    element 'a/@href' => :url
  end
end

blog = BlogScraper.parse open('http://example.com')

blog.title
#=> "My blog title"

blog.articles.first.title
#=> "First article title"

blog.articles.first.url
#=> "http://example.com/article"

There are sample scripts in the "examples/" directory; run them with:

ruby -rubygems examples/<script>.rb

See the wiki for more on how to use Scraper.

Requirements

None. Well, Nokogiri is a requirement if you pass in HTML content that needs to be parsed, like in the example above. Otherwise you can initialize the scaper with an Hpricot document or anything else that implements at(selector) and search(selector) methods.

Something went wrong with that request. Please try again.