gistfile1.txt

# This script can be used to check craigslist to see if certain apartment complexes become available.
# You can customize the Apartments, Periods and Bedrooms below to change the parameters.

require 'rubygems'
require 'htmlentities'
require 'rss'
require 'open-uri'
require 'pry'

# We use these constants to check multiple spellings and meanings of the same concept.
Apartments = {
  "Beaumont" => [/beaumont/i, /baeumont/i], 
  "The Oaks" => [/oaks/i, /oak's/i], 
  "Abbott Pointe" => [/abbott point/, /abbot point/i, /abott point/]
}
Periods = [/spring/i, /jan/i, /dec/i, /second semester/i]
Bedrooms = [/2 bed/i, /2 br/i, /2br/i, /2 bd/i, /2bd/i]


class ApartmentFinder
  # Find apartments based on the rss feed urls given.
  def find(urls)
    urls.each do |url|
      open(url) do |rss|
        feed = RSS::Parser.parse(rss)
        find_from_feed(feed)
      end
    end
  end
  
  # Check each item in the feed
  def find_from_feed(feed)
    feed.items.each do |entry|
      checker = AdChecker.new(entry.title, entry.description, entry.link)
      checker.check_ad
    end
  end
end

class AdChecker
  def initialize(title, content, url)
    coder = HTMLEntities.new
    @title = coder.decode(title)
    @content = coder.decode(content)
    @url = url
  end
  
  # Checks whether or not the title or content match any of the regexes given.
  def check_title_and_content(regexes)
    regexes.any? { |regex| @title =~ regex || @content =~ regex }
  end
  
  def check_ad
    Apartments.each_key do |key|
      if check_title_and_content(Apartments[key])
        # If something matches the apartment hash, we output it
        print "Apartment found at #{key}"

        print " ( SPRING )" if check_title_and_content(Periods)
        print " ( 2 BEDROOM )" if check_title_and_content(Bedrooms)
        binding.pry
        puts "\n", @title, @url, "\n"
      end
    end
  end
end

apartment_finder = ApartmentFinder.new
apartment_finder.find([
  'http://lansing.craigslist.org/apa/index.rss', 
  'http://lansing.craigslist.org/sub/index.rss'
  ])