Re: gEDA-dev: Broken slotting

[Peter Clifton <pcjc2-at-NOSPAMcam_dot_ac_dot_uk>]

On Wed, 2007-09-12 at 05:59 -0400, John Doty wrote:
> On Sep 11, 2007, at 11:02 PM, Steve Meier wrote:
> 
> > I have been violating the rule "All pins must have a pinseq=#  
> > attribute
> > attached to the pin object. " regularily. I reason, that if a  
> > device has
> > only one gate then no slot list is needed. Now I wonder if the slot  
> > list
> > is needed for spice?
> 
> What a Spice netlister needs is a way to figure out in what order the  
> pins are for the device or subcircuit instance. It uses pinseq for  
> this purpose. This is independent of the use of pinseq for slotting.
> 
> It seems to me that eventually we'll want to move the slotting and  
> pin assignment stuff to a database (see the lm7912 thread) to support  
> different packaging and usage of symbols. In that case, each pin will  
> need a unique identifier, and pinseq serves that purpose. But we  
> shouldn't keep using it directly for Spice (or anything else).  
> Rather, it should key the actual parameters, as it does in slotting.  
> Right now, pinseq is confusingly overloaded, but it need not be.

Ok, so how about this fix for now...

Keep pinseq as always was in libgeda + gschem, and not update it.

But have spice-sdb take into account slot=? when assigning pin order.
The only thing I can think which might need adding is guile-accessible
API to retrieve the slot number and pin-count on the component. These
may already be do-able though.

-- 
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)



_______________________________________________
geda-dev mailing list
geda-dev@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev