Set #any? vs. not #empty?

README.md

Calling #any? on Set is much slower than calling #empty? and negating it! Take a look at these benchmarks.

Warming up --------------------------------------
               #any?   182.420k i/100ms
         not #empty?     2.432M i/100ms
Calculating -------------------------------------
               #any?      1.834M (± 1.3%) i/s -      9.303M in   5.073017s
         not #empty?     24.254M (± 0.3%) i/s -    121.597M in   5.013433s

Comparison:
         not #empty?: 24254424.3 i/s
               #any?:  1834197.2 i/s - 13.22x  (± 0.00) slower

set-benchmark.rb

require 'benchmark/ips'
require 'set'

set = Set.new
10.times { set << rand }
p set

Benchmark.ips do |x|
  x.report("#any?") do |times|
    i = 0
    while i < times
      set.any?
      i += 1
    end
  end

  x.report("not #empty?") do |times|
    i = 0
    while i < times
      !set.empty?
      i += 1
    end
  end

  x.compare!
end