From: Robert Klemme <shortcutter@...>
Date: 2007-08-07T14:15:08+09:00
Subject: Re: Determining the common prefix for several strings

On 07.08.2007 03:16, Pe�a wrote:
> From: Robert Klemme [mailto:shortcutter@googlemail.com] 
> # I have added some interesting benchmarking:
> # 
> # $ ./prefix.rb
> #                                     user     system       total        real
> # 1000 * list.rx                  0.016000   0.000000    0.016000 (  0.014000)
> # 1000 * list.rx with conversion  0.031000   0.000000    0.031000 (  0.030000)
> # 1 * large.rx                    0.000000   0.000000    0.000000 (  0.000000)
> # 1 * large.rx with conversion    0.000000   0.000000    0.000000 (  0.000000)
> 
> Robert, thanks for the benchmark. I included Brian|Martin's solution in the mix but still the regex scheme is way a lot better w regards to speed.
> 
> Could some kind soul enlighten me why the regex solution works better, and even best using large strings?

My theory is that it's completely done in C and there is only *one* 
invocation.  All other solutions need some form of iteration implemented 
in Ruby.

Kind regards

	robert