[PD] dynamic gui masking problem

Roman Haefeli reduzierer at yahoo.de
Tue Apr 6 09:59:14 CEST 2010


On Sat, 2010-04-03 at 19:40 -0400, Mathieu Bouchard wrote:
> > By using [canvasdelete] from the iemguts. It allows to delete certain 
> > objects of a certain canvas by their index number (for instance, 'delete 
> > 12'). The canvas needs a [canvasdelete] in order to add the 'delete' 
> > method to the canvas and a [sendcanvas] to which the 'delete <id>' 
> > message is sent.
> 
> How do you intend to manage those numbers from Pd without ever getting 
> mixed up ?

It's actually fairly easy. The GOP's top-level canvas contains a fixed
number of objects, that are the required minimum for being able to
dynamically delete objects: [canvasdelete], [r
$0.sendcanvas]-[sendcanvas], [pd guts], [namecanvas $0.myabs].
So the first dynamically created object will have index 5, the next #6
etc. However, what matters is the number of fixed objects. Let's say I
create dynamically 7 additional objects, I just need to send seven times
'delete 5' (the number '5' protects the first 5 objects from being
deleted)  to [sendcanvas] in order to remove them.  

> 
> > Then I ran into another problem: I was creating [cnv] objects in a GOP 
> > and moving them out of the GOP area, which is a way to display something 
> > in the parent patch outside the GOP area. However, when deleting such 
> > gui objects, they remain visible in the parent (as if the parent 
> > wouldn't be refreshed).
> 
> Why do you want to rely on a bug ?

I don't care at this point, whether it is considered a bug or not.
Personally, I cannot fix it anyway. It may won't be fixed soon anyway.
Although you consider it a bug, I find it a feature. By using mostly Pd
when programming, I am kind of accustomed to using kludges. Pick one of
those reasons.

Roman
 


	
		
___________________________________________________________ 
Der frühe Vogel fängt den Wurm. Hier gelangen Sie zum neuen Yahoo! Mail: http://mail.yahoo.de




More information about the Pd-list mailing list