X-Authentication-Warning:	delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient:	geda-user AT delorie DOT com
X-Authenticated-IP:	207.224.51.38
Mime-Version:	1.0 (Apple Message framework v1084)
Subject:	Re: [geda-user] duplicate refdes values
From:	John Doty <jpd AT noqsi DOT com>
In-Reply-To:	<20111108100616.GA32119@yumi>
Date:	Tue, 8 Nov 2011 06:38:31 -0700
Message-Id:	<89AF1CB2-6703-4ABC-A43D-7EA37559426F@noqsi.com>
References:	<20111108100616 DOT GA32119 AT yumi>
To:	geda-user AT delorie DOT com
X-Mailer:	Apple Mail (2.1084)
X-MIME-Autoconverted:	from quoted-printable to 8bit by delorie.com id pA8DccQj023621
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

On Nov 8, 2011, at 3:06 AM, Cyril Hrubis wrote:

> Hi!
> I've got bitten by this once more and decided for fix. The problem is
> that I've, by a mistake, forgot to change refdes on one of my resistors
> and spend another ten minutes by figuring out what went wrong.
> 
> What I want to fix is that both gsch2pcb and pcb are silent about this
> and the pcb just picks up only one of them.
> 
> As far as I understand the gsch2pcb source it acts as a stream editor
> and has no memory of previous parts I was thinking of adding hash table
> of previous refdes values and make it print warning on duplicate
> entires. I could prepare patch for this.
> 
> The pcb is more compliacted though both by the source code and by a gui
> perspective. 

1. This has nothing to do with pcb: this is gnetlist's normal behavior.

2. This is not a bug, it is a feature. Duplicate refdes values are the way you express the fact that you are treating a single single physical device as broken into several sections.

3. The drc2 back end to gnetlist already complains about duplicated refdes values for non-slotted components. Checking a design for errors is what drc2 is for, and if you're not using multiple heterogeneous symbols for parts, drc2 will do what you want here. The problem is that this isn't necessarily an error: this is one of the many cases where drc2's model of design is narrower than the capabilities of the toolkit. In this case, it is hard to fix drc2, as the indications of true error (inconsistent pin connections) are hidden behind the unnecessarily opaque gnetlist back-end API.

John Doty              Noqsi Aerospace, Ltd.
http://www.noqsi.com/
jpd AT noqsi DOT com

- Raw text -