Hi Thomas,

I tried to do a similar thing in 'clone', which is an abstraction cloner that runs several instances of one abstraction (very useful for CPU tests like the one described in my last post ;-). Whenever one tries to reload an abstraction with a different number of inlets or outlets Pd would need to recreate the object. This only worked for me once: on object creation time... - When I change the abstraction Pd does not know about this since the abstraction is completely handled by clone. I haven't found a way to tell Pd to recreate a object box to reflect the changes (this is only supported for abstractions). To do so Pd would have to call functions in the object similar to the *_new() and *_free() functions to handle the changes. May be it's possible to add something like that to the Pd sources but I don't knwo where to start...

BTW, clone can be found here: http://www.akustische-kunst.org/puredata/clone/

best, Olaf

Thomas Grill schrieb:

Hi Jerome, this is true... i wonder why i haven't thought of that before. However, the problem is that PD (or at least flext) doesn't support a change of the inlet/outlet configuration without a recreation of the object. This means that the "reload" message can't change to number of inlets if that changes in the script.

Hmmm, has anyone tried to change the inlets/outlets message-based and can PD cope with that at all?

best greetings, Thomas

----- Original Message ----- From: "Jerome Etienne" jme@off.net To: "Thomas Grill" t.grill@gmx.net Sent: Thursday, November 13, 2003 8:51 PM Subject: bug reoprt for py/pyext

...

Hello, i think your object is veryusefull as it allow pd user to do scripting and gui much easier than with plain puredata. this email is just a small bug report.

"reload" message doesnt update the number of inlet in the number of inlet has changed in the python script.

---

PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list