Readline differences from 1.9.2 to 1.9.3?

Here's my test program:

···

----------------------------------------------
#!/usr/bin/env ruby

require 'readline'

ans = Readline.readline("? ", false)
puts ans
puts ans. length
----------------------------------------------
and here's my test session with Ruby 1.9.2 and Ruby 1.9.3

11:00 cards:>rvm use 1.9.2
Using /Users/bryanloc/.rvm/gems/ruby-1.9.2-p320
11:00 cards:>./tester.rb
? 出

1
11:00 cards:>rvm use 1.9.3
Using /Users/bryanloc/.rvm/gems/ruby-1.9.3-p392
11:01 cards:>./tester.rb
? \U+FFE5\U+FFBA

0

OK, am I doing something wrong, or do I need to do something special to get Chinese characters?
A simple call of gets(), i.e. not using readline, works fine, and returns the Chinese character.

Thanks in advance,

Bryan

Are you on OS X?

I am using ruby-2.0.0-p195, and if I install it naïvely I see:

ratdog:tmp mike$ ./try.rb
? \U+FFE5\U+FFBA

0
ratdog:tmp mike$ rbenv version
2.0.0-p195 (set by /Users/mike/.rbenv/version)

If I re-install using the helpful hints at Home · rbenv/ruby-build Wiki · GitHub (making sure I have the homebrew openssl and readline installed...)

ratdog:tmp mike$ env RUBY_CONFIGURE_OPTS=--with-openssl-dir=`brew --prefix openssl` CONFIGURE_OPTS=--with-readline-dir=`brew --prefix readline` rbenv install 2.0.0-p195
Downloading ruby-2.0.0-p195.tar.gz...
[...]
ratdog:tmp mike$ ./try.rb ? 出

1
ratdog:tmp mike$ rbenv version
2.0.0-p195 (set by /Users/mike/.rbenv/version)

Hope this helps,

Mike

···

On 2013-06-30, at 5:06 AM, Bryan Lockwood <lockwood.bryan@mac.com> wrote:

Here's my test program:
----------------------------------------------
#!/usr/bin/env ruby

require 'readline'

ans = Readline.readline("? ", false)
puts ans
puts ans. length
----------------------------------------------
and here's my test session with Ruby 1.9.2 and Ruby 1.9.3

11:00 cards:>rvm use 1.9.2
Using /Users/bryanloc/.rvm/gems/ruby-1.9.2-p320
11:00 cards:>./tester.rb
? 出

1
11:00 cards:>rvm use 1.9.3
Using /Users/bryanloc/.rvm/gems/ruby-1.9.3-p392
11:01 cards:>./tester.rb
? \U+FFE5\U+FFBA

0

OK, am I doing something wrong, or do I need to do something special to get Chinese characters?
A simple call of gets(), i.e. not using readline, works fine, and returns the Chinese character.

Thanks in advance,

Bryan

--

Mike Stok <mike@stok.ca>
http://www.stok.ca/~mike/

The "`Stok' disclaimers" apply.