From: Sam Ruby <rubys@...>
Date: 2008-02-10T02:42:29+09:00
Subject: Re: STDIN encoding differs from default source file encoding

ts wrote:
>>>>>> "Y" == Yukihiro Matsumoto <matz@ruby-lang.org> writes:
> 
> Y> What's your locale and version info?
> 
> vgs% locale charmap
> ANSI_X3.4-1968
> vgs%
>  
> vgs% ./ruby -ve 'a = "\x81"; p a.encoding; a =~ /a/'
> ruby 1.9.0 (2008-02-09 revision 15320) [i686-linux]
> #<Encoding:ASCII-8BIT>
> vgs%
>  
> vgs% export LC_CTYPE="en_US.UTF-8"
> vgs%
>  
> vgs% locale charmap
> UTF-8
> vgs% 
> 
> vgs% ./ruby -ve 'a = "\x81"; p a.encoding; a =~ /a/'
> ruby 1.9.0 (2008-02-09 revision 15320) [i686-linux]
> #<Encoding:UTF-8>
> -e:1:in `<main>': broken UTF-8 string (ArgumentError)
> vgs% 

The difference is that Paul reported his results using IRB:

$ locale charmap
UTF-8

$ ruby -ve 'a = "\x81"; p a.encoding; a =~ /a/'
ruby 1.9.0 (2008-02-09 revision 15420) [i686-linux]
#<Encoding:UTF-8>
-e:1:in `<main>': broken UTF-8 string (ArgumentError)

$ irb
irb(main):001:0> a = "\x81"; p a.encoding; a =~ /a/
#<Encoding:ASCII-8BIT>
=> nil
irb(main):002:0>

- Sam Ruby