Fast Ruby - Barcelona ruby conference video notes

baruco.org -notes

=================================
def slow(&block)
  block.call
end

def slow
  Proc.new.call
end

- The below one is 5 times faster then above, in execution 

def fast
  yield
end

=================================

(1..100).map{|i| i.to_s}
(1..100).map(&:to_s) #20 times faster - https://github.com/rails/rails/pull/16833

=================================

Hash#merge # Slow
Hash#merge! # Fast - 3 times faster then above

=================================

enum.each_with_object({}) do |e, h|
  h.merge!(e => e)
end

- The below one is 2 times faster then above

enum.each_with_object({}) do |e, h|
  h[e] = e
end

=================================

Hash#fetch
Hash#fetch with block as a second argument # Faster 2 times

=================================

String#gsub
String#sub # 50% faster, so if not needed we should not use gsub

=================================

String#gsub
String#tr # 5 times faster then gsub

=================================

Should not use exceptions for control flow, instead we can put conditions. that would be 10 times faster.

=================================

Array#each_with_index
while loop # 80% faster then above, but more lines of code then above

=================================