This is the mail archive of the
java@gcc.gnu.org
mailing list for the Java project.
Re: libgcj/9715: Not all required character encodings are supported
- From: Tom Tromey <tromey at redhat dot com>
- To: Mark Wielaard <mark at klomp dot org>
- Cc: Jesse Rosenstock <jmr at ugcs dot caltech dot edu>, java at gcc dot gnu dot org
- Date: 18 Feb 2003 11:29:44 -0700
- Subject: Re: libgcj/9715: Not all required character encodings are supported
- References: <20030217022848.C5ECF84054@euro.ugcs.caltech.edu><1045463165.30205.614.camel@elsschot>
- Reply-to: tromey at redhat dot com
>>>>> "Mark" == Mark Wielaard <mark@klomp.org> writes:
Mark> Yes, see for how to translate between the old/new names:
Mark> http://java.sun.com/j2se/1.4.1/docs/guide/intl/encoding.doc.html
Mark> ISO8859_15 is just ISO-8859-15 which is supported by iconv. It just
Mark> doesn't seem to know about this particular alias.
We could revamp the aliasing support in gnu.gcj.convert to pass
canonical names to iconv -- if we were sure that iconv would always
know the canonical names. Or we could add a second aliasing
mechanism.
Or we could add an internal 8859_15 converter. That would be easy
enough.
Mark> Cool! I see we already have the decoders in gnu.java.nio.charset.
Mark> It would be nice if gnu.gcj.convert could use those and vice versa.
Long term I've been hoping we could switch to java.nio and remove
gnu.gcj.convert. Whether this is possible depends on performance and
also whether we can continue to fall back to iconv.
Tom