Re: bonobo-activation; freeing base services ...

[Michael Meeks <michael ximian com>]

Hi Maciej,

On Mon, 29 Oct 2001, Maciej Stachowiak wrote:
> There's already got to be a nonstandard
> "give_me_all_the_objects_that_have_not_been_freed_yet" function or the
> like, right?

        No. This is an ORB internal, and invoked via CORBA_ORB_destroy,
the ref tracking works via hooks in CORBA_Object_duplicate / release, and
ORBit_RootObject_init ( called to construct ORB resources internaly ).

> It doesn't seem unreasonable to add a helper function to make that one
> work better.

        So this would be adding a new, non-standard method - this voids
much of the below.

> I would rather do `nasty' things in the platform libraries to make the
> API seen by developers simpler.

        Ok - fine,

> Your argument boils down to: "I am doing something bad. But I don't
> like your idea for how to fix it, therefore unless you accept my
> solution (despite thinking it's bad design), then it's all your
> fault."

        Sigh; I am doing something bad ? I'm making my regression tests
far more powerful, and providing an invaluable programmer tool.

> If you're really unwilling to add a way to make the reference leak
> debugger ignore objects that have the same lifetime as the program,

        I am.

> then maybe we could consider naming the function something other than
> bonobo_activation_shutdown. Perhaps bonobo_activation_debug_refcounts,
> so it's clear what it does and that it's optional, so people won't
> worry about calling it on abnormal exits or anything crazy like that,

        Right - well, I tend to think that shutdown describes the function
fairly well; but how about we don't declare it in the public headers - but
have a gentlemans agreement not to change the signature ? then I can whack
an:

        extern int bonobo_activation_shutdown (void);

        in bonobo-main.c and no-one will know any better ?

        Some other names might be, bonobo_activation_release_resources,
bonobo_activation_destroy_orb etc. that is unless we want to deliberately
assign a somewhat meaningless name, in which case I think George would
have us use bonobo_activation_gegl or something :-)

        So - can you confirm that you're passably happy with this as a way
to go / suggest the method name you prefer and I'll re-work ( and
preferably commit ) the patch in that form.

        Thanks,

                Michael.

-- 
 mmeeks gnu org  <><, Pseudo Engineer, itinerant idiot