Python and CORBA

[Martin von Loewis]

cbbrowne at news.hex.net (Christopher Browne) writes:

> The folks working on GNOME have made a particular "design preference,"
> by default, to mandate a particular authentication approach when GNOME
> apps export functionality via CORBA.
> 
> On the one hand, this "preference" is indeed not standard, which may
> be argued to be a Bad Thing.

It seems that not the ORBit developers made this preference, but the
packagers. For example, the Debian change log lists the change

orbit (0.5.0-5) frozen unstable; urgency=medium
   * Postinst for liborbit0 creates default /etc/orbitrc, if none exists.
     Default file disables tcp, for security (closes: Bug#52519).  More
     details in README.Debian.

Indeed, the IORBit 0.5.0 included with RedHat 6.2 does include IIOP
profiles in every object reference by default.

> I rather like orbit-python; it would be nicest for it to be "maximally
> compliant with the ``standard'' mapping," although my priority is for
> it to work well...

It is the fine details that matter. For example, I feel that

import CosNaming

is much cleaner than writing

orb.load_idl("naming.idl")

which immediately raises the question where to find naming.idl. Did
you ever try freezing an IDL file? :-)

Also, I feel that inheritance is cleaner than delegation for
implementing interfaces, but that might be a matter of taste. The best
thing probably would be if orbit-python provided all aspects of the
standard mapping, offering extensions where they feel the standard
mapping is inappropriate. Then users can chose whether they want to
write portable (across ORBs) code, or orbit-specific code.

Regards,
Martin