X-Authentication-Warning: delorie.com: mail set sender to geda-help-bounces using -f
X-Recipient: geda-help AT delorie DOT com
Date: Sat, 12 Dec 2020 13:00:32 +0100 (CET)
From: Roland Lutz <rlutz AT hedmen DOT org>
To: "Klaus Rudolph (lts-rudolph AT gmx DOT de) [via geda-help AT delorie DOT com]" <geda-help AT delorie DOT com>
Subject: Re: [geda-help] using net names on multiple sub schematics used by
 single symbol
In-Reply-To: <bb2887ac-06cb-1d7f-ff3f-a73c8b0bf8b4@gmx.de>
Message-ID: <alpine.DEB.2.21.2012121250160.1221@nimbus>
References: <c6376b29-a72c-6ae0-1b39-081ecb97ec1c AT gmx DOT de> <alpine DOT DEB DOT 2 DOT 21 DOT 2012041901490 DOT 1174 AT nimbus> <3e21c34b-571c-8762-7e68-f096bcf10a37 AT gmx DOT de> <alpine DOT DEB DOT 2 DOT 21 DOT 2012081605590 DOT 1246 AT nimbus> <20201209082005 DOT 8890C8512092 AT turkos DOT aspodata DOT se>
 <alpine DOT DEB DOT 2 DOT 21 DOT 2012091338400 DOT 1205 AT nimbus> <ce08e8ac-cdc1-9ed4-420a-c52a750baa3c AT gmx DOT de> <alpine DOT DEB DOT 2 DOT 21 DOT 2012111629350 DOT 12695 AT nimbus> <bb2887ac-06cb-1d7f-ff3f-a73c8b0bf8b4 AT gmx DOT de>
User-Agent: Alpine 2.21 (DEB 202 2017-01-01)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Reply-To: geda-help AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: geda-help AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

On Sat, 12 Dec 2020, Klaus Rudolph (lts-rudolph AT gmx DOT de) [via 
geda-help AT delorie DOT com] wrote:
>> That's still valid, but I agree that it would be a good idea to update 
>> these resources to use the newer convention.
>
> ... and maybe add a "deprecated" warning while processing these kind of
> schematics?

Using refdes= for ports isn't deprecated because there are some things 
which can't be achieved with a portname= attribute, like having a 
component be a port in some situations and a connector in others.

> No! I want an explicit access on the namespace. Every level of hierarchy
> should be seen as local, but each of these level should be in some way
> *explicit* addressable but giving additional information in some syntax
> to the name of the net. Something like net=<upper>#<some_net_name> or
> "net=<global>#<some_net_name> " whatever syntax you like.

But that's exactly what I/O ports are for!  So you want to have an 
"invisible pin" on the subschematic symbol which connects to a named net, 
and an "invisible port symbol" inside the subschematic which connects the 
port to a local net?

> But it would be nice if buses can be used as nets. Connecting them via a 
> element ( maybe a pin ) may connect the whole bus to the sub schematic. 
> If you have a design with some address and data bus and you can simply 
> connect all your sub schematic peripherals with the pins would be nice. 
> I am currently did not have such jobs to realize, but it looks "natural" 
> to me.

I have already implemented that as part of my (at the time) experimental 
netlister features, but I haven't merged it yet, mostly because of 
different conflicting conventions for pin numbering.

What pin numbering scheme do you use?

Roland