libgcj/9715: Not all required character encodings are supported

Tom Tromey tromey@redhat.com
Fri Feb 21 02:22:00 GMT 2003


>>>>> "Jesse" == Jesse Rosenstock <jmr@ugcs.caltech.edu> writes:

Jesse> In writing my patch to make String.getBytes use the java.nio encoders,
Jesse> I noticed that (at least on my redhat 7.2 system) the iconv based
Jesse> converter supports UTF-16, UnicodeBig, UnicodeLittle, UTF-16BE, and
Jesse> UTF-16LE; it just encodes these incorrectly, only encoding the first
Jesse> character and not putting a byte order mark.  Perhaps someone wants to
Jesse> look at that.

Could you submit a libgcj PR for that?

If this is required by Java but not implemented by iconv, then we need
to either write our own converters for these cases, or add a special
case of another kind somewhere.

Tom



More information about the Java mailing list