From: Robert Dober <robert.dober@...>
Date: 2009-07-30T01:21:45+09:00
Subject: Re: Indexing hash with longer strings faster?

2009/7/29 Mladen Jablanović <jablan@gmail.com>:
> You pointed to the right direction here!
>
> Although your example gives me 100 as result (all the hashes are
> unique), here's what suggests that hash collision indeed is the reason
> of the slowness:
>
>>> [*0..99].map{|i| [*0..99].map{|j| [i.to_s,j.to_s].hash}}.flatten.uniq.size
> => 1756
>>> [*0..99].map{|i| [*0..99].map{|j| [i.to_s + '00000',j.to_s + '00000'].hash}}.flatten.uniq.size
> => 10000

OMG my code was sloppy, but you got it right, here is the 1.9 code
506/18 > cat hashes.rb && ruby -v hashes.rb
#!/usr/local/bin/ruby -w
# encoding: utf-8
# file: /home/robert/log/ruby/ML/hashes.rb
p [*0..99].map{|i| [*0..99].map{|j| [i.to_s,j.to_s].hash}}.flatten.uniq.size
p [*0..99].map{|i| [*0..99].map{|j| [i.to_s + '00000',j.to_s +
'00000'].hash}}.flatten.uniq.size

# vim: sts=2 sw=2 ft=ruby expandtab nu :
ruby 1.9.1p243 (2009-07-16 revision 24175) [i686-linux]
10000
10000

as expected :).
Cheers
Robert
>
> Can you please just post the ruby 1.9 results here?
>
> Thanks!
>
>



-- 
Toutes les grandes personnes ont d’abord été des enfants, mais peu
d’entre elles s’en souviennent.

All adults have been children first, but not many remember.

[Antoine de Saint-Exupéry]