JRuby

Do a better job of determining "locale" encoding when default charset's name is not in our encoding tables

Details

  • Type: Bug Bug
  • Status: Open Open
  • Priority: Major Major
  • Resolution: Unresolved
  • Affects Version/s: JRuby 1.6RC2
  • Fix Version/s: None
  • Component/s: Ruby 1.9.2
  • Labels:
    None
  • Number of attachments :
    0

Description

See JRUBY-5525. There, we originally had a problem that if the default Java Charset's name was not found in our encoding tables (EncodingDB), we would cause a NPE. My fix was to failover to the 1.9-mode "locale" encoding, which ultimately just uses ASCII-8BIT when the default charset's name doesn't look up.

We should endeavor to do better than failing over to ASCII-8BIT when we can't match Java's default Charset to an Encoding.

Activity

Hide
Permalink
Charles Oliver Nutter added a comment -

For the MacRoman and MacCentralEurope cases, I'm not sure we've got much course of action. The aliases they present are all still "Mac" charsets:

~/projects/jruby ➔ java -Dfile.encoding=MacCentralEurope -jar lib/jruby.jar -rjava -e "puts java.nio.charset.Charset.defaultCharset().aliases().to_a"
MacCentralEurRoman
x-mac-centraleurope
MacCentralEuropean
x-mac-centraleurroman
x-mac-ce
MacCentralEurope
x-mac-centraleuropean

~/projects/jruby ➔ java -Dfile.encoding=MacRoman -jar lib/jruby.jar -rjava -e "puts java.nio.charset.Charset.defaultCharset().aliases().to_a"
MacRoman
mac
csMacintosh
x-MacRoman
x-mac-roman

I'd like to get some charsets from other locales and platforms to see if it's worth trying to use aliases to do a second search of encodings.

Show
Charles Oliver Nutter added a comment - For the MacRoman and MacCentralEurope cases, I'm not sure we've got much course of action. The aliases they present are all still "Mac" charsets: ~/projects/jruby ➔ java -Dfile.encoding=MacCentralEurope -jar lib/jruby.jar -rjava -e "puts java.nio.charset.Charset.defaultCharset().aliases().to_a" MacCentralEurRoman x-mac-centraleurope MacCentralEuropean x-mac-centraleurroman x-mac-ce MacCentralEurope x-mac-centraleuropean ~/projects/jruby ➔ java -Dfile.encoding=MacRoman -jar lib/jruby.jar -rjava -e "puts java.nio.charset.Charset.defaultCharset().aliases().to_a" MacRoman mac csMacintosh x-MacRoman x-mac-roman I'd like to get some charsets from other locales and platforms to see if it's worth trying to use aliases to do a second search of encodings.

People

  • Assignee:
    Thomas E Enebo
    Reporter:
    Charles Oliver Nutter
Vote (0)
Watch (0)

Dates

  • Created:
    Updated: