GitHub currently disables the rendering of emoji in large document, that's why you see the strange :: stuff (they are emojis!).
In Erik Michaels-Ober's great talk, 'Writing Fast Ruby': Video @ Baruco 2014, Slide, he presented us with many idioms that lead to faster running Ruby code. He inspired me to document these to let more people know. I try to link to real commits so people can see that this can really have benefits in the real world. This does not mean you can always blindly replace one with another. It depends on the context (e.g. gsub versus tr).
Each idiom has a corresponding code example that resides in code.
All results listed in README.md are 🏃 with Ruby 2.2.0p0 on OS X 10.10.1. Machine information: MacBook Pro (Retina, 15-inch, Mid 2014), 2.5 GHz Intel Core i7, 16 GB 1600 MHz DDR3. Your results may vary, but you get the idea. : )
Let's write faster code, together! <3
👬 👭 👬 👬 👭 👫 👭 👬 👭 👫 👬 👫 👯 👫 👬 👬 👬 👬 👫 👬 👭 👭 👫 👫 👬 👭 👬 👯 👫
Use benchmark-ips (2.0+).
require 'benchmark/ips'
def fast
end
def slow
end
Benchmark.ips do |x|
x.report('fast code description') { fast }
x.report('slow code description') { slow }
x.compare!
endParallel Assignment vs Sequential Assignment code
Parallel Assignment allocates an extra array.
$ ruby -v code/general/assignment.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Parallel Assignment 99.146k i/100ms
Sequential Assignment 127.143k i/100ms
-------------------------------------------------
Parallel Assignment 2.522M (± 7.5%) i/s - 12.592M
Sequential Assignment 5.686M (± 8.6%) i/s - 28.226M
Comparison:
Sequential Assignment: 5685750.0 i/s
Parallel Assignment: 2521708.9 i/s - 2.25x slower
begin...rescue vs respond_to? for Control Flow code
$ ruby -v code/general/begin-rescue-vs-respond-to.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
begin...rescue 29.452k i/100ms
respond_to? 106.528k i/100ms
-------------------------------------------------
begin...rescue 371.591k (± 5.4%) i/s - 1.855M
respond_to? 3.277M (± 7.5%) i/s - 16.299M
Comparison:
respond_to?: 3276972.3 i/s
begin...rescue: 371591.0 i/s - 8.82x slower
define_method vs module_eval for Defining Methods code
$ ruby -v code/general/define_method-vs-module-eval.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
module_eval with string 125.000 i/100ms
define_method 138.000 i/100ms
-------------------------------------------------
module_eval with string 1.130k (±20.3%) i/s - 5.500k
define_method 1.346k (±25.9%) i/s - 6.348k
Comparison:
define_method: 1345.6 i/s
module_eval with string: 1129.7 i/s - 1.19x slower
Array#bsearch vs Array#find code
$ ruby -v code/array/bsearch-vs-find.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
find 1.000 i/100ms
bsearch 42.216k i/100ms
-------------------------------------------------
find 0.184 (± 0.0%) i/s - 1.000 in 5.434758s
bsearch 577.301k (± 6.6%) i/s - 2.913M
Comparison:
bsearch: 577300.7 i/s
find: 0.2 i/s - 3137489.63x slower
Array#count vs Array#size code
$ ruby -v code/array/count-vs-size.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
#count 130.991k i/100ms
#size 135.312k i/100ms
-------------------------------------------------
#count 6.697M (± 7.1%) i/s - 33.403M
#size 7.562M (± 9.1%) i/s - 37.481M
Comparison:
#size: 7562457.4 i/s
#count: 6696763.0 i/s - 1.13x slower
Array#shuffle.first vs Array#sample code
Array#shuffleallocates an extra array.
Array#sampleindexes into the array without allocating an extra array.
This is the reason why Array#sample exists.
—— @sferik rails/rails#17245
$ ruby -v code/array/shuffle-first-vs-sample.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Array#shuffle.first 25.406k i/100ms
Array#sample 125.101k i/100ms
-------------------------------------------------
Array#shuffle.first 304.341k (± 4.3%) i/s - 1.524M
Array#sample 5.727M (± 8.6%) i/s - 28.523M
Comparison:
Array#sample: 5727032.0 i/s
Array#shuffle.first: 304341.1 i/s - 18.82x slower
Enumerable#each + push vs Enumerable#map code
$ ruby -v code/enumerable/each-push-vs-map.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Array#each + push 9.025k i/100ms
Array#map 13.947k i/100ms
-------------------------------------------------
Array#each + push 99.634k (± 3.2%) i/s - 505.400k
Array#map 158.091k (± 4.2%) i/s - 794.979k
Comparison:
Array#map: 158090.9 i/s
Array#each + push: 99634.2 i/s - 1.59x slower
Enumerable#each vs for loop code
$ ruby -v code/enumerable/each-vs-for-loop.rb
ruby 2.2.0preview1 (2014-09-17 trunk 47616) [x86_64-darwin14]
Calculating -------------------------------------
For loop 17.111k i/100ms
#each 18.464k i/100ms
-------------------------------------------------
For loop 198.517k (± 5.3%) i/s - 992.438k
#each 208.157k (± 5.0%) i/s - 1.052M
Comparison:
#each: 208157.4 i/s
For loop: 198517.3 i/s - 1.05x slower
Enumerable#each_with_index vs while loop code
$ ruby -v code/array/each_with_index-vs-while-loop.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
each_with_index 11.496k i/100ms
While Loop 20.179k i/100ms
-------------------------------------------------
each_with_index 128.855k (± 7.5%) i/s - 643.776k
While Loop 242.344k (± 4.5%) i/s - 1.211M
Comparison:
While Loop: 242343.6 i/s
each_with_index: 128854.9 i/s - 1.88x slower
Enumerable#map...Array#flatten vs Enumerable#flat_map code
-- @sferik rails/rails@3413b88, Replace map.flatten with flat_map, Replace map.flatten(1) with flat_map
ruby -v code/enumerable/map-flatten-vs-flat_map.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Array#map.flatten(1) 3.315k i/100ms
Array#map.flatten 3.283k i/100ms
Array#flat_map 5.350k i/100ms
-------------------------------------------------
Array#map.flatten(1) 33.801k (± 4.3%) i/s - 169.065k
Array#map.flatten 34.530k (± 6.0%) i/s - 173.999k
Array#flat_map 55.980k (± 5.0%) i/s - 283.550k
Comparison:
Array#flat_map: 55979.6 i/s
Array#map.flatten: 34529.6 i/s - 1.62x slower
Array#map.flatten(1): 33800.6 i/s - 1.66x slower
Enumerable#reverse.each vs Enumerable#reverse_each code
Enumerable#reverseallocates an extra array.
Enumerable#reverse_eachyields each value without allocating an extra array.
This is the reason whyEnumerable#reverse_eachexists.
-- @sferik rails/rails#17244
$ ruby -v code/enumerable/reverse-each-vs-reverse_each.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Array#reverse.each 16.746k i/100ms
Array#reverse_each 18.590k i/100ms
-------------------------------------------------
Array#reverse.each 190.729k (± 4.8%) i/s - 954.522k
Array#reverse_each 216.060k (± 4.3%) i/s - 1.078M
Comparison:
Array#reverse_each: 216060.5 i/s
Array#reverse.each: 190729.1 i/s - 1.13x slower
Enumerable#detect vs Enumerable#select.first code
$ ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Enumerable#select.first 8.515k i/100ms
Enumerable#detect 33.885k i/100ms
-------------------------------------------------
Enumerable#select.first 89.757k (± 5.0%) i/s - 1.797M
Enumerable#detect 434.304k (± 5.2%) i/s - 8.675M
Comparison:
Enumerable#detect: 434304.2 i/s
Enumerable#select.first: 89757.4 i/s - 4.84x slower
Enumerable#select.last vs Enumerable#reverse.detect code
$ ruby -v code/enumerable/select-last-vs-reverse-detect.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Enumerable#reverse.detect 62.636k i/100ms
Enumerable#select.last 11.687k i/100ms
-------------------------------------------------
Enumerable#reverse.detect 1.263M (± 8.2%) i/s - 6.326M
Enumerable#select.last 119.387k (± 5.7%) i/s - 596.037k
Comparison:
Enumerable#reverse.detect: 1263100.2 i/s
Enumerable#select.last: 119386.8 i/s - 10.58x slower
Enumerable#sort vs Enumerable#sort_by code
$ ruby -v code/enumerable/sort-vs-sort_by.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Enumerable#sort 1.158k i/100ms
Enumerable#sort_by 2.401k i/100ms
-------------------------------------------------
Enumerable#sort 12.140k (± 4.9%) i/s - 61.374k
Enumerable#sort_by 24.169k (± 4.0%) i/s - 122.451k
Comparison:
Enumerable#sort_by: 24168.9 i/s
Enumerable#sort: 12139.8 i/s - 1.99x slower
Hash#[] vs Hash#dup code
Source: http://tenderlovemaking.com/2015/02/11/weird-stuff-with-hashes.html
Does this mean that you should switch to Hash[]? Only if your benchmarks can prove that it’s a bottleneck. Please please please don’t change all of your code because this shows it’s faster. Make sure to measure your app performance first.
$ ruby -v code/hash/bracket-vs-dup.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Hash[] 29.403k i/100ms
Hash#dup 16.195k i/100ms
-------------------------------------------------
Hash[] 343.987k (± 8.7%) i/s - 1.735M
Hash#dup 163.516k (±10.2%) i/s - 825.945k
Comparison:
Hash[]: 343986.5 i/s
Hash#dup: 163516.3 i/s - 2.10x slower
Hash#fetch with argument vs Hash#fetch + block code
$ ruby -v code/hash/fetch-vs-fetch-with-block.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Hash#fetch + arg 15.650k i/100ms
Hash#fetch + block 130.271k i/100ms
-------------------------------------------------
Hash#fetch + arg 184.562k (± 5.2%) i/s - 923.350k
Hash#fetch + block 5.880M (± 7.5%) i/s - 29.311M
Comparison:
Hash#fetch + block: 5880209.2 i/s
Hash#fetch + arg: 184562.0 i/s - 31.86x slower
Hash#each_key instead of Hash#keys.each code
Hash#keys.eachallocates an array of keys;
Hash#each_keyiterates through the keys without allocating a new array.
This is the reason whyHash#each_keyexists.
—— @sferik rails/rails#17099
$ ruby -v code/hash/keys-each-vs-each_key.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Hash#keys.each 56.690k i/100ms
Hash#each_key 59.658k i/100ms
-------------------------------------------------
Hash#keys.each 869.262k (± 5.0%) i/s - 4.365M
Hash#each_key 1.049M (± 6.0%) i/s - 5.250M
Comparison:
Hash#each_key: 1049161.6 i/s
Hash#keys.each: 869262.3 i/s - 1.21x slower
Hash#merge! vs Hash#[]= code
$ ruby -v code/hash/merge-bang-vs-\[\]=.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Hash#merge! 1.023k i/100ms
Hash#[]= 2.844k i/100ms
-------------------------------------------------
Hash#merge! 10.653k (± 4.9%) i/s - 53.196k
Hash#[]= 28.287k (±12.4%) i/s - 142.200k
Comparison:
Hash#[]=: 28287.1 i/s
Hash#merge!: 10653.3 i/s - 2.66x slower
Hash#merge vs Hash#merge! code
$ ruby -v code/hash/merge-vs-merge-bang.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Hash#merge 39.000 i/100ms
Hash#merge! 1.008k i/100ms
-------------------------------------------------
Hash#merge 409.610 (± 7.6%) i/s - 2.067k
Hash#merge! 9.830k (± 5.8%) i/s - 49.392k
Comparison:
Hash#merge!: 9830.3 i/s
Hash#merge: 409.6 i/s - 24.00x slower
Block vs Symbol#to_proc code
Symbol#to_procis considerably more concise than using block syntax.
...In some cases, it reduces the number of lines of code.
—— @sferik rails/rails#16833
$ ruby -v code/proc-and-block/block-vs-to_proc.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
Block 4.632k i/100ms
Symbol#to_proc 5.225k i/100ms
-------------------------------------------------
Block 47.914k (± 6.3%) i/s - 240.864k
Symbol#to_proc 54.791k (± 4.1%) i/s - 276.925k
Comparison:
Symbol#to_proc: 54791.1 i/s
Block: 47914.3 i/s - 1.14x slower
Proc#call vs yield code
$ ruby -v code/proc-and-block/proc-call-vs-yield.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
block.call 70.663k i/100ms
yield 125.061k i/100ms
-------------------------------------------------
block.call 1.309M (± 5.7%) i/s - 6.572M
yield 6.103M (± 7.7%) i/s - 30.390M
Comparison:
yield: 6102822.9 i/s
block.call: 1309452.1 i/s - 4.66x slower
String#casecmp vs String#downcase + == code
$ ruby -v code/string/casecmp-vs-downcase-\=\=.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
String#downcase + == 101.900k i/100ms
String#casecmp 109.828k i/100ms
-------------------------------------------------
String#downcase + == 2.915M (± 5.4%) i/s - 14.572M
String#casecmp 3.708M (± 6.1%) i/s - 18.561M
Comparison:
String#casecmp: 3708258.7 i/s
String#downcase + ==: 2914767.7 i/s - 1.27x slower
String Concatenation code
$ ruby code/string/concatenation.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
String#+ 96.314k i/100ms
String#concat 99.850k i/100ms
String#append 100.728k i/100ms
"foo" "bar" 121.936k i/100ms
-------------------------------------------------
String#+ 2.731M (± 4.6%) i/s - 13.677M
String#concat 2.847M (± 5.2%) i/s - 14.279M
String#append 2.972M (± 6.1%) i/s - 14.807M
"foo" "bar" 4.951M (± 6.2%) i/s - 24.753M
Comparison:
"foo" "bar": 4950955.3 i/s
String#append: 2972048.5 i/s - 1.67x slower
String#concat: 2846666.4 i/s - 1.74x slower
String#+: 2730980.7 i/s - 1.81x slower
String#match vs String#start_with?/String#end_with? code (start) code (end)
⚠️
Sometimes you cant replace regexp withstart_with?,
for example:"a\nb" =~ /^b/ #=> 2but"a\nb" =~ /\Ab/ #=> nil.
⚠️
You can combinestart_with?andend_with?to replaceerror.path =~ /^#{path}(\.rb)?$/to this
error.path.start_with?(path) && error.path.end_with?('.rb', '')
—— @igas rails/rails#17316
$ ruby -v code/string/start-string-checking-match-vs-start_with.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
String#=~ 55.411k i/100ms
String#start_with? 113.854k i/100ms
-------------------------------------------------
String#=~ 910.625k (± 4.6%) i/s - 4.544M
String#start_with? 3.983M (± 5.5%) i/s - 19.924M
Comparison:
String#start_with?: 3983284.9 i/s
String#=~: 910625.0 i/s - 4.37x slower
$ ruby -v code/string/end-string-checking-match-vs-start_with.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
String#=~ 52.811k i/100ms
String#end_with? 100.071k i/100ms
-------------------------------------------------
String#=~ 854.830k (± 5.8%) i/s - 4.278M
String#end_with? 2.837M (± 5.5%) i/s - 14.210M
Comparison:
String#end_with?: 2836536.9 i/s
String#=~: 854830.3 i/s - 3.32x slower
String#gsub vs String#sub code
$ ruby -v code/string/gsub-vs-sub.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
String#gsub 35.724k i/100ms
String#sub 42.426k i/100ms
-------------------------------------------------
String#gsub 486.614k (± 5.4%) i/s - 2.429M
String#sub 611.259k (± 4.6%) i/s - 3.055M
Comparison:
String#sub: 611259.4 i/s
String#gsub: 486613.5 i/s - 1.26x slower
String#gsub vs String#tr code
$ ruby -v code/string/gsub-vs-tr.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
String#gsub 38.268k i/100ms
String#tr 83.210k i/100ms
-------------------------------------------------
String#gsub 516.604k (± 4.4%) i/s - 2.602M
String#tr 1.862M (± 4.0%) i/s - 9.320M
Comparison:
String#tr: 1861860.4 i/s
String#gsub: 516604.2 i/s - 3.60x slower
attr_accessor vs getter and setter code
https://www.omniref.com/ruby/2.2.0/files/method.h?#annotation=4081781&line=47
$ ruby -v code/general/attr-accessor-vs-getter-and-setter.rb
ruby 2.2.0p0 (2014-12-25 revision 49005) [x86_64-darwin14]
Calculating -------------------------------------
getter_and_setter 61.240k i/100ms
attr_accessor 66.535k i/100ms
-------------------------------------------------
getter_and_setter 1.660M (± 9.7%) i/s - 8.267M
attr_accessor 1.865M (± 9.2%) i/s - 9.248M
Comparison:
attr_accessor: 1865408.4 i/s
getter_and_setter: 1660021.9 i/s - 1.12x slower
Please! Edit this README.md then Submit a Awesome Pull Request!
Code example is wrong? 😢 Got better example? 😍 Excellent!
Please open an issue or Open a Pull Request to fix it.
Thank you in advance! 😉 🍺
Share this with your #Rubyfriends! <3
Brought to you by @JuanitoFatas
Feel free to talk with me on Twitter! <3
💝 💞 💝 💖 💙 💕 ❤️ 💗 💚 💞 💓 💛 💗 💓 💟 💙
-
Talk by Davy Stevenson @ RubyConf 2014.
-
Provides Big O notation benchmarking for Ruby.
-
Talk by Prem Sichanugrist @ Ruby Kaigi 2014.
-
Make your Rubies go faster with this command line tool.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
To the extent possible under law, @JuanitoFatas has waived all copyright and related or neighboring rights to "fast-ruby".
This work belongs to the community.