The Ruby script I used to parse www.freewareppc.com

class Rating

  attr_reader :stars, :comment

  def initialize(stars, comment)

    @stars = stars

    @comment = comment

  end   

end



class Program

  attr_reader :name, :url, :ratings

  def initialize(name, url)

    @ratings = []

    @name = name

    @url = url

  end   

  def imageurl

    "#{url.gsub('freewareppc.com', 'freewareppc.com/images/products').gsub('shtml', 'gif')}"

  end

  def addRating(rating)

    @ratings << rating

  end

  def ratingsWithStars(stars)

    ratings.select {|rating| rating.stars == stars}.length

  end

  def <=>(other)

    result = other.ratings.length <=> @ratings.length

    return result if result != 0

    result = other.ratingsWithStars(5) <=> ratingsWithStars(5)

    return result if result != 0

    result = other.ratingsWithStars(4) <=> ratingsWithStars(4)

    return result if result != 0

    result = other.ratingsWithStars(3) <=> ratingsWithStars(3)

    return result if result != 0

    result = other.ratingsWithStars(2) <=> ratingsWithStars(2)

    return result if result != 0 

    ratingsWithStars(1) <=> other.ratingsWithStars(1)

  end

  def report

    return if ratings.length == 0

    puts ""

    puts "
#{name}
"

    puts ""

    puts "
"

    ratings.sort {|a,b| b.stars <=> a.stars }.each {|rating|

      puts "    " + ("*" * rating.stars) + " " + rating.comment + "
"

    }

  end

end

programs = []

IO.foreach("FreewarePPC-simplified.txt") { | line |

  if line =~ /\$\$\$(.*) \^ (.*)/ then

    programs << Program.new($1, $2.strip)

  elsif line =~ />>> (\**) \^ (.*)/ then

    programs.last.addRating(Rating.new($1.length, $2))

  else

    puts line

    throw :shouldNeverReachHere

  end

}



programs.sort.each {|program| program.report}