Here's a way I parse through keys in an S3 bucket, calculating depth based on number of /'s in the key. I then parse and throw folders/files that are too deep and show only the folder/file entries at the current level

Reading S3 keys as a file hierarchy

def entries(dir = '')
	entries = []
	
	# Depth of the directory path
  dir_depth = dir.count('/')
	
	@bucket.keys(:prefix => path).each do |entry|
	  # Entries cannot end with ~ (ascii 126) or starts with # or . (35 and 46 respectively)
	  if entry.name[-1] != 126 && entry.name[0] != 35 && entry.name[0] != 46 && entry.name != path
	    # If is a subdirectory (entries that are deeper than the current path must only be directories)
	    if (entry.depth == dir_depth + 1 && entry.is_folder?) || entry.depth == dir_depth
	      entries << entry
	    end
	  end
	end
end

module RightAws
  class S3
    class Bucket
      def common_prefixes(prefix, delimiter = '/')
        common_prefixes = []
        @s3.interface.incrementally_list_bucket(@name, { 'prefix' => prefix, 'delimiter' => delimiter }) do |thislist|          
          common_prefixes += thislist[:common_prefixes]
        end
        common_prefixes
      end
    end
    
    class Key
      # Is this key a folder
      def is_folder?
       self.name[-1] == 47
      end
      
      # Return the depth of the key
      def depth
       self.name.count('/')
      end
		end
	end
end