X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Date: Tue, 2 Jan 2018 20:30:49 +0100 (CET)
From: Roland Lutz <rlutz AT hedmen DOT org>
To: karl AT aspodata DOT se, geda-user AT delorie DOT com
Subject: Re: [geda-user] why pinnumber in sym pins?
In-Reply-To: <20171226122009.366BF804CEDE@turkos.aspodata.se>
Message-ID: <alpine.DEB.2.20.1801022021340.14184@nimbus>
References: <20171226122009 DOT 366BF804CEDE AT turkos DOT aspodata DOT se>
User-Agent: Alpine 2.20 (DEB 67 2015-01-07)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
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, 26 Dec 2017, karl AT aspodata DOT se wrote:
> Why must one have a pinnumber attribute for each pin i a sym file:
> http://wiki.geda-project.org/geda:master_attributes_list#pinnumber

The pin number is the main identifier of a pin inside a component.
Having a pin without an identifier would only make sense if the behavior
of that pin is defined in another way, and you never need to refer to
that pin by its identifier.

One example of that would be power/ground symbols; the pinnumber= 
attribute on their sole pin is obsolete for the new netname= notation. 
Another example would be port symbols in subschematics.

> Wouldn't it suffice to have pinseq and pinlabel ?

pinseq= is only relevant for slotted symbols (where it serves as a way to 
assign a pinnumber to the pin depending on the slot) and SPICE simulation 
(where it is used as a hacky way to split up a package into individual 
gates).  pinlabel= is only relevant for subschematic symbols.

> Wouldn't it be useful to allow  to have just a function (pinlabel)
> attached to the pin without knowing which pinnumber it will end up on ?

You would still need some way of mapping the function to the pinnumber.

> So, it would be nice if one could map sym -> fp with just the "pinlabel" 
> (gschem) -> pin "Name" (pcb). But pcb doesn't seem to support that idea.

In fact, the term "pinnumber" may be misleading because it doesn't need to 
be a number.  You could use "S", "G", and "D" as pinnumbers for a 
transistor symbol, or "shield" as a pinnumber for a shield pin, and then 
map that to a physical position via the footprint.

> E.g. dc/dc converters with single output usually have four pins +/-Vin 
> and +/-Vout in all sorts of packages where the pinnumber (from the 
> datasheet) can be anything but 1,2,3,4, but their pin names are 
> +/-Vin/Vout.

> Any suggestions or pointers ?

In this example, just use "+Vin", "-Vin" and so on (or some PCB-friendly 
variant of that) as the pinnumbers.

Roland