[Chicago] Next meeting...So far

[Allan Spale]

Hi everyone,

I have used stackless and I thought that it was the best. python. ever.
Running some of the sample "hackeysack" simulation code was very fast. All
was good...until I tried to debug the thing in Eclipse. Maybe it was an
Eclipse problem or maybe it was a pdb (Python debugger) problem, but once I
started to trace into code that used stackless, my debugger basically froze
or went off to something untraceable. I could not have code that I could not
debug, so I dejectedly dropped stackless. For those of you that are more
familiar with compilers and languages, the magic of stackless comes by
"mangling" the normal functioning of the C call stack (since Python is
written in C/C++)...which would explain why things just did not work when I
was debugging stackless calls in my Python code.

Just when I thought all my stackless dreams were shattered and gone, I came
across PEP 342 (http://www.python.org/dev/peps/pep-0342/) that indicated to
me that the essence of stackless was already implemented with the "yield"
statement. Granted, this is not completely stackless because if you dug deep
enough, there were some scheduling functions that would come in handy if you
did not want to handcode them yourself or somehow work pre-emption into the
mix. I have yet to compare these two side-by-side, but I am expecting
"yield" to be a lot cleaner than stackless. Additionally, learning how to
use generators effectively is a nice side-effect of learning to use "yield"
for co-routines. One other fun fact..."yield" should work with Jython.

Despite my support for "yield" and not stackless, I would still vote for
hearing a talk about stackless.


Allan

On Tue, Nov 4, 2008 at 11:01 AM, Pete <pfein at pobox.com> wrote:

> On Nov 3, 2008, at 9:14 AM, David Durham, Jr. wrote:
>
>> I do not currently have anything to present on.  I am willing to talk
>> about stackless python (and demo a PSP example), but I do not have
>> anything substantive at this time.  Also, no one really said until now
>> that there was interest. :)  I kind of figured that stackless was old
>> news to the python group.
>>
>
> I'd love to see a talk on Stackless, on a PSP or other hardware.  I'm
> always interested in alternative computation models.
>
> Sounds like we've got a few speakers at this point. Brian, can you
> summarize what we've got / still need to figure out?  I would, but I do
> enough of that sort of thing in my Day Job already. ;-)
>
> _______________________________________________
> Chicago mailing list
> Chicago at python.org
> http://mail.python.org/mailman/listinfo/chicago
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/chicago/attachments/20081104/fd2536b5/attachment.htm>