From: "Brian Schröder" <ruby.brian@...>
Date: 2005-05-13T19:15:40+09:00
Subject: Re: Still umlauts

On 13/05/05, Jonas Hartmann <Mail@jonas-hartmann.com> wrote:> Bertram Scharpf wrote:> > Hi,> >> > Am Freitag, 13. Mai 2005, 04:34:00 +0900 schrieb Josef 'Jupp' SCHUGT:> >> >>At Tue, 10 May 2005 06:58:30 +0900, Bertram Scharpf wrote:> >>> >>>does this no longer work?> >>> >>You forgot to define the meaning of 'no longer works':> >>> >> - What precisely do you mean by 'it works'?> >> >> > Sorry, that was not actually elaborate. Now I think it> > never worked at all.> >> >> >>>--------------------> >>>#!/usr/bin/env ruby> >>># -*- coding: UTF-8 -*-> >>>> >>>puts $KCODE> >>>puts "�������"> >>>--------------------> >>>> >>>I tried it with Ruby 1.8.2 and 1.9, Debian Linux.> >>>What do I miss?> >> >> > In the meantime I received an answer in ruby-core and it> > seems Matz just _planned_ to implement it but didn't have> > the time yet.> >> > The problem arises when my program is run on SuSE Linux> > where the default encoding is UTF-8.> >> > A better way to test in which encoding you reside is:> >> >     "��" =~ /./> >     puts $&.length> >> > This gives 1 in `None' and 2 in `UTF-8'.> >> >> >>When in doubt *set* $KCODE explicitly.> >> >> > This has no influence on how the source code is read. The> > string "�" produces an error when `ruby -Ku' is called.> >> > Of course I should have written "\xc3\xa4" and "\xc3".> > Is this the only way to handle strings UTF-8 in ruby?> > >> > Sorry again for the noise.> >> > Bertram> >> >> > regards>    jonas> > 
No, it is perfectly possible to use utf-8 encoded source-files. (Butyou should use ruby -Ku then).
But some UTF-8 things are a bit weird. E.g. the length of an UTF-8String is the length of its bytes not the number of characters, so Ihad to write my own utf-8_length routine.
regards,
Brian
-- http://ruby.brian-schroeder.de/
Stringed instrument chords: http://chordlist.brian-schroeder.de/