X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Date: Wed, 9 Nov 2011 11:40:54 +0100
From: Gabriel Paubert <paubert AT iram DOT es>
To: geda-user AT delorie DOT com
Subject: Re: [geda-user] duplicate refdes values
Message-ID: <20111109104054.GA6261@iram.es>
References: <20111108100616 DOT GA32119 AT yumi> <89AF1CB2-6703-4ABC-A43D-7EA37559426F AT noqsi DOT com> <20111108180039 DOT GA6472 AT yumi> <201111081833 DOT pA8IXpek012142 AT envy DOT delorie DOT com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <201111081833.pA8IXpek012142@envy.delorie.com>
User-Agent: Mutt/1.5.18 (2008-05-17)
Reply-To: geda-user AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-user AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On Tue, Nov 08, 2011 at 01:33:51PM -0500, DJ Delorie wrote:
> 
> > I'm saying this is a bug, I'm saying that this is worth a warning. What
> > about warning about duplicated refdes for syms that has no slotting
> > defined?
> 
> Unfortunately, that's a very common case these days.  *Most* of my
> designs have at least one such component, usually an MCU that I've
> split into 2-3 blocks (power, bus, I/O).
> 
> The tricky part is being *reliable* about detecting overlapping
> refdes's vs an intentionally split-up component.  You have to consider
> all the possible situations and plan accordingly.

Hmm, typically, intentionally split-up components do not have the
same pin numbers on each part. But maybe some people occasionally
repeat some pins (I've split components, but never repeated pin
numbers myself).

A possible solution is to add an attribute to (all parts) of deliberately 
split-up components.

But I believe I'd prefer to check for duplicate (refdes,pinnumber), this
way you would also detect duplicate slot attribution (not that I use
many slotted components anymore), and identical split-up components in which
you forget to edit the refdes of one of the subparts.

No, I won't code it. This is only food for thought. My circuits are
far too cimple to need it.


> 
> Also, just "saying this is a bug" doesn't get stuff done.  If you want
> it done, you need to be willing to do it too.

Full ack.

	Gabriel