hash_selector_pattern.rb

type = :info

time = Time.now
2_000_000.times do
  msg = case type
        when :success then 'alert-success'
        when :error then 'alert-danger'
        when :warn then 'alert-warning'
        when :info then 'alert-info'
        end
end
puts "case: #{Time.now - time}"

time = Time.now
2_000_000.times do
  msg = if type == :success then 'alert-success'
        elsif type == :error then 'alert-danger'
        elsif type == :warn then 'alert-warning'
        elsif type == :info then 'alert-info'
        end
end
puts "if: #{Time.now - time}"

time = Time.now
2_000_000.times do
  msg = { success: 'alert-success',
        error:   'alert-danger',
        notice:  'alert-info',
        warn:    'alert-warning' }[type]
end
puts "hash: #{Time.now - time}"

Darwin Kernel Version 13.1.0: Thu Jan 16 19:40:37 PST 2014; root:xnu-2422.90.20~2/RELEASE_X86_64 x86_64 i386 MacBookPro11,3 Darwin

jruby 1.7.10 (1.9.3p392) 2014-01-09 c4ecd6b on Java HotSpot(TM) 64-Bit Server VM 1.7.0_45-b18 [darwin-x86_64]

case: 0.283
if: 0.514
hash: 0.453
hash lookup: 0.133

why allocate hash 2_000_000 times?

Why would a single end user care if it took 2 million iterations to notice roughly a 0.15 second difference between a hash lookup vs a case statement? If the only noticeable difference is readability of the code, I'll stick with a case statement, thank you very much.