[omniORB] wchar and wstring

Sai-Lai Lo S.Lo@uk.research.att.com
21 Jul 1999 18:35:20 +0100


>>>>> Stefan Seefeld writes:

> as it appears, omniORB doesn't support wide character
> types. Will they be included in the 2.8.0 release ?

Yes it is not supported and will not be with 2.8.0 (as said in README.FIRST
in omniORB 2.8.0pre1).

To support wchar and wstring properly, one has to support CodeSet
conversion on the wire. For example, your native codeset is some non-unicode
codeset and the remote end only understands unicode. The on-the-wire
protocol has to do the translation. This is only defined since GIOP
1.1. And GIOP 1.1 is broken in that it does not provide sufficent
information for generic bridge to decode the data types. GIOP 1.2 fixed
this but the protocol is not supported in 2.8.0. It is work in progress.

Before you ask, I might as well state that other datatypes such as longlong
and longdouble are not supported mainly because I cannot see an
implementation that is portable across ORB implementations and language
environments. The C++ Fixed type mapping in CORBA 2.2 is broken and so on.
Stay away from using these types until the spec. has reached a certain
maturity.

In general, you may read that a new feature has been defined in the latest
and greatest OMG spec. The fact that a spec. has been published *does not*
mean that the spec. is *implementable*!!! In fact, I guess in all
likelihood that most of the OMG specifications are adopted without anyone
writing a single line of code! Take a look at the interceptor spec. in
CORBA 2.2. Now we have another RFP in progress to revise the interceptors!

Sai-Lai


-- 
Sai-Lai Lo                                   S.Lo@uk.research.att.com
AT&T Laboratories Cambridge           WWW:   http://www.uk.research.att.com 
24a Trumpington Street                Tel:   +44 1223 343000
Cambridge CB2 1QA                     Fax:   +44 1223 313542
ENGLAND