that might be ok — i *think* in the very early s7s  that was the only way to 
allocate record instances…
there may be some issue with random numbers and patterns:: iirc im using a c++ 
random generator, not a scheme one...

On Jun 11, 2015, at 8:09 AM, <[email protected]> <[email protected]> 
wrote:

> Not disqualifying anything else in cm, but having cm/scm/patterns.scm in
> snd would be a great start.
> 
> Running through patterns.scm now, it seems the main thing stopping it is
> a void 'define-record.  Replacing these with calls on define-record-type
> (srfi-17?) from snd/r7rs.scm makes it load and the test-patterns output
> expected results.
> 
> Q: would it break things in CM if define-record got replaced with
> define-record-type in pattern.scm?
> 
> To use it stand-alone, ie. in a running snd, it needs some things from
> cm/scm/s7.scm + utilities.scm.  I guess for now its ok to just 'inline'
> these dependencies, and aiming at a patterns.scm which to include with
> snd only.
> 
> -anders
> 
> _______________________________________________
> Cmdist mailing list
> [email protected]
> http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist
> 


_______________________________________________
Cmdist mailing list
[email protected]
http://ccrma-mail.stanford.edu/mailman/listinfo/cmdist

Reply via email to