A simple little scraper for the YC List

yclist_scraper.rb

require 'open-uri'
require 'nokogiri'

module YCList
  class << self

    def url
      'http://yclist.com/'
    end

    def scrape
      doc = Nokogiri::HTML(open(url))

      @companies = doc.css('#companies table tbody tr').map do |row|
        Company.new(*row.children.map(&:text)[1..-1])
      end
    end

    def companies
      @companies || scrape && @companies
    end

    def all(filter = //, field = :name)
      companies.select { |c| c.send(field).match filter }
    end

    def one(*args)
      all(*args).sample
    end

    def alive
      status ''
    end

    def dead
      status 'Dead'
    end

    def exited
      status 'Exited'
    end

    def method_missing(m, *args, &block)
      fields = %i(name url yc_class status description year)
      all(args.first, m) if fields.include?(m)
    end
  end

  class Company
    attr_reader :name, :url, :yc_class, :status, :description, :year

    def initialize(name, url, yc_class, status, description)
      @name        = name
      @url         = url
      @yc_class    = yc_class
      @status      = status
      @description = description
      @year        = yc_class.gsub(/^W|S/, "20").to_i
    end

    def to_s
      name
    end

    alias :inspect :to_s
  end
end

Usage

# All of the companies
YCList.companies

# All companies with a description that includes the word "iPhone"
YCList.description 'iPhone'

# All the dead companies
YCList.dead

# The company name for one of the companies in the summer class of '05
YCList.one('S05', :yc_class).name