On 6/17/06, Ken Tilton <kentilton@gmail.com> wrote:

On 6/16/06, Ryan Forsythe < ryanf@cs.uoregon.edu> wrote:

On Jun 16, 2006, at 2:43 PM, Ken Tilton wrote:
> I downloaded it and it has the right size, but PicoZip (which
> normally does fine with tars and tgzs (but not always)) finds
> nothing to extract.

Weird. I've uploaded a zip of the export, maybe that'll work for
picozip: http://pycells.pyworks.org/pycells-r12.zip

Thanks, that worked.

One quick reaction: in Cells, observers are more clasically OO, they are defined as a class attribute. I am not sure if PyCells is doing the same after a quick glance, but I do see that the Cell data structure has a slot for observers, which sounds as if in PyCells they might be Cell-specific. Correct me if I am wrong.

I forgot to mention something (and I am CCing cells-devel to see what others think): the idea of instance/slot-specific observers does not seem like a bad idea. It certainly is consistent with Cells allowing different rules for the same slot of different instances. Normal observers really do have to be class-oriented (it does not how a color of a widget gets decided, a change still has to trigger an OS update) but if PyCells happens to support instance/slot-specific observers that would be OK.

The only problem I see is that you need a little new syntax. I suppose in Cells it would be:

(c-formua (:observer (observer-lambda () ......)...)

Does anyone recall ever needing an instance-specific observer?

kt