JRuby should respect encoding set in file header in 1.8 mode

[dgolombek]

~/src/test> jruby --version
jruby 1.7.6 (ruby-1.8.7p370) 2013-10-22 6004147 on Java HotSpot(TM) 64-Bit Server VM 1.6.0_65-b14-462-11M4609 [darwin-x86_64]
~/src/test> irb -Ku
jruby-1.7.6 :001 >  # encoding: ascii
jruby-1.7.6 :002 >   QSAFECHAR = Regexp.new("[ \t\x21\x23-\x7e\x80-\xff]")
 => /[  !#-~\200-\377]/
jruby-1.7.6 :003 > ALL_QSAFECHARS = /\A#{QSAFECHAR}*\z/
RegexpError: (RegexpError) empty range in char class: /\A(?-mix:[   !#-~\200-ÿ])*\z/
    from (irb):3:in `evaluate'
    from org/jruby/RubyKernel.java:1116:in `eval'
    from /Users/dgolombek/.rvm/rubies/jruby-1.7.6/lib/ruby/1.8/irb.rb:158:in `eval_input'
    from /Users/dgolombek/.rvm/rubies/jruby-1.7.6/lib/ruby/1.8/irb.rb:271:in `signal_status'
    from /Users/dgolombek/.rvm/rubies/jruby-1.7.6/lib/ruby/1.8/irb.rb:155:in `eval_input'
    from org/jruby/RubyKernel.java:1517:in `loop'
    from org/jruby/RubyKernel.java:1268:in `catch'
    from /Users/dgolombek/.rvm/rubies/jruby-1.7.6/lib/ruby/1.8/irb.rb:154:in `eval_input'
    from /Users/dgolombek/.rvm/rubies/jruby-1.7.6/lib/ruby/1.8/irb.rb:71:in `start'
    from org/jruby/RubyKernel.java:1268:in `catch'
    from /Users/dgolombek/.rvm/rubies/jruby-1.7.6/lib/ruby/1.8/irb.rb:70:in `start'
    from /Users/dgolombek/.rvm/rubies/jruby-1.7.6/bin/irb:13:in `(root)'
jruby-1.7.6 :004 > quit
~/src/test> irb
jruby-1.7.6 :001 > # encoding: ascii
jruby-1.7.6 :002 >   QSAFECHAR = Regexp.new("[ \t\x21\x23-\x7e\x80-\xff]")
 => /[  !#-~\200-\377]/
jruby-1.7.6 :003 > ALL_QSAFECHARS = /\A#{QSAFECHAR}*\z/
 => /\A(?-mix:[     !#-~\200-\377])*\z/

I've only seen this in 1.8 mode, it appears to work in 1.9. This showed up in the Vcard gem. I'll patch that gem to work around the issue. This is low priority but I figured I'd submit it anyways.

[enebo]

This works in 9k and we will never fix this in 1.7.x as it is winding down....resolving.