Damien,

I found a good starting Seaside project.  It requires only minimal
security to start, and while a little clumsy, it will actually be useful
to me, and will be a good demo piece for "selling" future Seaside
projects that will have to be very secure.  Long story.

The good news is that it has me chipping away at the daunting task of
porting lots of infrastructure code, and as a confessed streams junkie,
I ran directly into the problems we have discussed.  Since I "can't
wait" for, nor do I expect a robust solution[*] to the silent errors
problems of #next and friends, I will simply deprecate them for my own
use.  Toward that end, #nextOne, #nextMany:, and #nextAvailable: exist
in a change set.  I will no doubt discover the need to provide
additional overrides.  Some of my implementations are little more than
checked versions of what lurks in the image; hopefully we can do better
with Nile, or any and all are welcome to do better tomorrow (today for
that matter<g>).

It is probably a little premature, but you are welcome to review what I
have, and I would very much like to discuss file stream semantics.  I
will add some Dolphin-influenced entry points, but some
changes/additions are going to be necessary.  If nothing else, file name
manipulation will change for the benefit of Linux.

I understand that streams are set for milestone 3, so feel free to stick
to that.  At the risk of appearing to want to rewrite the schedule, I'm
stuck until I have at least a temporary solution, and hopefully one that
provides semantics that I can map onto whatever we ultimately create. 
Your input would be welcome.  Let me know whether and how you would like
to proceed.

Bill


[*] if we had Cincom's backing, I would be more optimistic, but with
them clinging to backward compatibility, adding selectors seems the best
approach to get moving.





Wilhelm K. Schwab, Ph.D.
University of Florida
Department of Anesthesiology
PO Box 100254
Gainesville, FL 32610-0254

Email: [EMAIL PROTECTED]
Tel: (352) 846-1285
FAX: (352) 392-7029


_______________________________________________
Pharo-project mailing list
[email protected]
http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Reply via email to