RE: ORBit_sequence helpers ...

[Murray Cumming Comneon com]

I believe these are meant to be internal APIs for use by language mappings
or ORBit itself.

Murray Cumming
murrayc@usa.net
www.murrayc.com 

> -----Original Message-----
> From: Gustavo J. A. M. " Carneiro [mailto:gjc@inescporto.pt] 
> Sent: Dienstag, 11. März 2003 19:04
> To: Michael Meeks
> Cc: Mark McLoughlin; orbit; Anders Carlsson
> Subject: Re: ORBit_sequence helpers ...
> 
> 
> 	In my _personal_ opinion, the use of ORBit APIs should 
> be minimized in GNOME.  One of the great things about CORBA 
> is that it is vastly documented, making CORBA code readable 
> to programmers outside the GNOME project.  By using ORBit 
> APIs, the code may become simpler or with better performance, 
> but also less readable[1].  In addition, because of this, 
> some people will acuse bonobo of being too much tied to ORBit.
> 	Just some thoughts...
> 
> [1] though not in this case in particular, since the function 
> names are self-descriptive; however, there are other cases 
> where ORBit internal APIs are used, and it's hard to know 
> what they mean.
> 
> On Ter, 2003-03-11 at 14:56, Michael Meeks wrote:
> > Hi guys,
> > 
> > 	So - I just added a set of 'ORBit_sequence' helpers to 
> HEAD; they do 
> > helpful things:
> > 
> > /* More friendly(?) sequence allocators */
> > gpointer       ORBit_sequence_alloc    (CORBA_TypeCode      
> sequence_tc,
> > 					CORBA_unsigned_long length);
> > void           ORBit_sequence_append   (gpointer            
> sequence,
> > 					gconstpointer       element);
> > void           ORBit_sequence_set_size (gpointer            
> sequence,
> > 					CORBA_unsigned_long length);
> > #define        ORBit_sequence_index(sequence,idx)
> > (sequence)->_buffer[(idx)]
> > void           ORBit_sequence_concat   (gpointer            
> sequence,
> > 					gconstpointer       append);
> > 
> > 	Which should make a sequence approaching as friendly as 
> a GArray, 
> > that should clean some nasty cruft out of bonobo-activation 
> at least - 
> > with GLists' being turned into sequences later etc. etc.
> > 
> > 	The patch is:
> > http://primates.ximian.com/~michael/sequence-helpers.diff
> > 
> > 	I'm slightly concerned that I'm mis-using _maximum - is 
> that supposed 
> > to be a 'bounded sequence' thing ? either way - we can trivially 
> > support the API by using the smart allocation n_elements 
> stuff with a 
> > little more work.
> > 
> > 	If anyone wants other Sequence / GArray type methods - 
> speak soon ...
> > 
> > 	HTH,
> > 
> > 		Michael.
> -- 
> Gustavo João Alves Marques Carneiro
> <gjc@inescporto.pt> <gustavo@users.sourceforge.net>
> 
> 
> _______________________________________________
> orbit-list mailing list
> orbit-list@gnome.org http://mail.gnome.org/mailman/listinfo/orbit-list
>