Hi all,
I added the slot and numslots check to the DRC.
Also, gnetlist reports errors when slot attribute is not a number and
using functions get-unique-slots or get-slots, so I changed gnetlist's
message so it can be identified as reported by gnetlist.

Attached is the patch. Regards,

Carlos

El jue, 23-10-2003 a las 04:08, Ales Hvezda escribi�:
> Hi Carlos,
> 
> [snip]
> >I think the current behaviour is not coherent. If I place a 7400
> >(multislotted), gschem will draw in the schematic the symbol
> >corresponding to slot 1, but no "numslots" neither "slot" attributes are
> >defined.
> 
>       True.  It is assumed that the default place is slot=1.  
> 
> >A mixed solution could be that gschem defines "slot" and "numslots"
> >attributes ONLY for multislotted parts. Then, the DRC backend could
> >assume (and it could do it only because that definition is automatic)
> >that parts without those attributes are non-multislotted parts.
> >
>       Yes.  If the slot or numslot attribute are not present (though
> numslots=0 should be present; if it is not, then that is a bug in the
> symbol and there are lots of such examples in the current library),
> then assume that the component is not slotted.
> 
>       If the slot=N attribute is present, then the component is 
> assumed to be slotted and should be checked to make sure N <= numslots
> etc...
> 
>       I just checked the symbol library.  All symbols that have a slot=1
> attribute also have a numslots=N attribute.  There are also symbols with
> numslots=0 and those do not have a slot= attribute.  There are no symbols
> which have a numslots=0 and also a slot=1 attribute (unless my script
> is wrong).
> 
>                                                       -Ales
> 
> ---Publicidad--------------------------------------------------------
> Juega con Ventura24.es, loter�a inteligente y multiplica tus
> posibilidades!! http://www.iespana.es/_reloc/email.ventura
> 

Attachment: gnetlist.diff.gz
Description: GNU Zip compressed data

Reply via email to