[Biocorba-l] report on Anys in CORBA::ORBit

[Martin Senger]

On Tue, 12 Jun 2001, Ewan Birney wrote:

> 
> Hmmm. This confirms my suspicisions. I would suspect a similar mess in
> fnorb/omniorb python.
>
   Yep, it may be. But we still do not know if the problem is in
Orbit itself or in CORBA::ORBit. Philip is testing now a pure ORBit client
(in C) to find the answer for that.

> I think name value pairs are evil outside of defined dictionaries and
> we should be against them anyway.
>
   This is not related to using or npt using Anys. I agree that name in
name-value pairs should be controlled.
 
> anys just make life difficult and annoying. My 2p.
>
  I have just done these tests on Anys because I feel that they are
inevitable. I do not think that in OO arena the strings are sufficient.

   In the analysis part, I will use, at the moment, an array of Anys - but
the perl-client user will not notice, it will be wrapped in a perl module
BSA::Novella::Client. Therefore, I will be able to change the internals of
this module when [CORBA::]ORBit people fix the problems with Anys in
structs.
   I hope to have Novella ready soon...
 
   Martin

-- 
Martin Senger

EMBL Outstation - Hinxton                Senger@EBI.ac.uk     
European Bioinformatics Institute        Phone: (+44) 1223 494636      
Wellcome Trust Genome Campus             (Switchboard:     494444)
Hinxton                                  Fax  : (+44) 1223 494468
Cambridge CB10 1SD
United Kingdom                           http://industry.ebi.ac.uk/~senger