Playing around with list summarization

number_list_summarization.rb

# Generate a list of all the possible port numbers with 1/80 randomly removed
# from the list.
ar = (1...(2 ** 16)).to_a.select { |i| rand(80) != 0 }

# Build a summary of the list
prev = ar[0]
summary = ar.slice_before do |e|
  prev, prev2 = e, prev
  prev2.succ != e
end.map do |a|
  (a.first == a.last) ? a.first : a.first..a.last
end

summary2 = ar.enum_for(:chunk).with_index do |state, index|
  state - index
end.map do |diff, group|
  (group.first == group.last) ? group.first : (group.first..group.last)
end

# Print out and compare the relative sizes as they would appear within a JSON
# document.
puts "Normal: #{ar.map(&:to_s).join(',').length} bytes"
puts "Summary: #{summary.map(&:to_s).join(',').length} bytes"
puts "Summary 2: #{summary2.map(&:to_s).join(',').length} bytes"

# Rebuild the original array
str_sum = summary.map(&:to_s).join(',')
original_nums = str_sum.split(',').map do |rep|
  first, last = rep.split('..')
  last.nil? ? first.to_i : (first.to_i..last.to_i).to_a
end.flatten

# Rebuild the original array
str_sum = summary2.map(&:to_s).join(',')
original_nums2 = str_sum.split(',').map do |rep|
  first, last = rep.split('..')
  last.nil? ? first.to_i : (first.to_i..last.to_i).to_a
end.flatten

puts original_nums == ar
puts original_nums2 == ar