Count unique elements in Array

count_unique_elements_in_array.rb

require 'benchmark'
require 'open3'
require 'histogram/array'

array = Array.new(10_000_000) { rand(60) }

puts 'uniq, count:'
uniq_count = Benchmark.measure do
  array.uniq.map { |key| array.count(key) }
end
puts uniq_count

puts 'system call (awk):'
system_call = Benchmark.measure do
  sys_count = "awk '{dups[$1]++} END{for (num in dups) {print num,dups[num]}}'"
  Open3.popen3(sys_count) do |i, o, _e, _t|
    i.write array
    i.close
    o.read
  end
end
puts system_call

puts 'histogram:'
histogram = Benchmark.measure do
  array.histogram
end
puts histogram

puts 'each_with_object:'
each_with_object = Benchmark.measure do
  array.each_with_object({}) { |x, mem| mem.key?(x) ? mem[x] += 1 : mem[x] = 1 }
end
puts each_with_object

puts 'count hash:'
count_hash = Benchmark.measure do
  count = Hash.new(0)
  array.each { |val| count[val] += 1 }
  count
end
puts count_hash

puts 'group_by:'
group_by = Benchmark.measure do
  array.group_by { |v| v }.map { |k, v| [k, v.size] }
end
puts group_by

# sample output:
# uniq, count:
#  17.740000   0.010000  17.750000 ( 17.761145)
# system call (awk):
#   1.750000   0.030000   7.400000 (  7.404668)
# histogram:
#   4.690000   0.000000   4.690000 (  4.706780)
# each_with_object:
#   1.740000   0.000000   1.740000 (  1.742432)
# count hash:
#   1.110000   0.000000   1.110000 (  1.125301)
# group_by:
#   1.000000   0.050000   1.050000 (  1.049218)