From: Wayne Brissette <wbrisett@...>
Date: 2013-01-08T20:08:40+09:00
Subject: Re: gsub with multibyte characters


On Jan 8, 2013, at 1:13 AM, Harry Kakueki wrote:

> On Tue, Jan 8, 2013 at 7:58 AM, Wayne Brisette <wbrisett@att.net> wrote:
>> I'm running into a problem trying to read a line of text that has some multibyte
>> characters gsub is failing. Is there a trick to finding and replacing multibyte
>> characters?
>> 
> 
> Do you have something like this as the first line of your code?
> 
> # encoding: UTF-8
> 

I have added this but when I do, I end up with 
'gsub!: Invalid byte sequence in UTF-8 (ArgumentError). 

Ironically if I use macruby's irb, I can do the gsub without any problems. 

Why would macruby allow this, but not the MRI version of Ruby? 


Wayne