Calling our experts on character encoding.

[Meskauskas Audrius]

CORBA supports two string and char types, the "wide" (usually 16 bit) 
encoding and the "narrow" (usually 8 bit). When sending a string or even a 
single character to another CORBA system, the message must include a short 
context record, indicating, how the text is encoded. The charsets are 
indicated by providing they standard "Registered Value" code (ASCII = 
0x00010020, UTF8 = 0x05010001, UTF-16 = 0x00010109 and so on). It is 
important to ensure that CORBA supports all encodings, that are currently 
supported by Classpath.

The class, responsible for all this functionality in Classpath, is 
gnu.CORBA.GIOP.CharSets_OSF.

CharSets_OSF tries to detect the available encodings automatically, but the 
internal hashtables must contain the "Registered Value" code of the 
available encoding. I have included the codes I was able to find, but the 
current list may be not complete. If some currently supported encodings are 
not listed, please add them into that class, specifying the name and the 
matching "Registered Value" .

This is also submitted as task #4169.

Some resources:
ftp://ftp.opengroup.org/pub/code_set_registry/cs_registry1.2h
http://www.iana.org/assignments/character-sets


Audrius