Samplify.rb

=begin

This fuction is designed to summarize the contents of a dictionary of arrays.
As many elements as possible will be grabbed for a specified size N while maintaining maximum diversity. 
Function will terminate as soon as the desired amount of items are gathered. 

Example IO:

posts = { 'Greg' => ['A','A','A'],
          'Steve' => [],
          'Elly' => ['C','C','C'],
          'George' => ['D']
}

samplify(2, posts) #Output: ['A','C']
samplify(5, posts) #Output: ['A','A','C','C','D']

- Nick Maccharoli 
March, 2015
=end

class Hash
    def purify!
        self.each{|h| self.delete(h[0]) if h[1].length == 0}
    end 
end

def samplify(sampleSize, dic)
        dic.purify!
        mixedItems = []
        grabPerRow = [(sampleSize / dic.keys.length), 1].max
        grabPerRow += (grabPerRow % 2)  
        setsVisited = 0 
        dic.keys.each_with_index{ |item, index|
                break if setsVisited == sampleSize
                currentRow = []
                begin
                        numberToTake = [sampleSize, item.length].min
                        currentRow = dic[item].take numberToTake
                        grabIndex = [(index * grabPerRow),mixedItems.length].min
                        currentRow.each{|element|
                                mixedItems[grabIndex] = element   
                                grabIndex += 1
                        }   
                        setsVisited += 1    
                rescue
                end 
        }   
        return mixedItems.slice(0,sampleSize)
end