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 Dec 2015 08:28:54 +0100 (CET)
X-X-Sender:	igor2 AT igor2priv
To:	geda-user AT delorie DOT com
X-Debug:	to=geda-user AT delorie DOT com from="gedau AT igor2 DOT repo DOT hu"
From:	gedau AT igor2 DOT repo DOT hu
Subject:	Re: [geda-user] gsch2pcb after refdes-renum? (If implemented
	syncronization detail)
In-Reply-To:	<555EDA31-984C-4978-AE21-F74907CDC993@noqsi.com>
Message-ID:	<alpine.DEB.2.00.1512090826360.9035@igor2priv>
References:	<56658683 DOT 401 AT envinsci DOT co DOT uk> <CALT8Ef4=tMf=WmjYmp-2B3rN0SBmoqF5RkCoWZEm=+2hTBTENA AT mail DOT gmail DOT com> <20151207153821 DOT c2ac19e6f24b1776a3595e4a AT gmail DOT com> <CD4C755E-8E00-427D-909C-3F762A038844 AT gmail DOT com> <20151208091411 DOT c8968b0bedb705765529176c AT gmail DOT com>
	<B2F4DA99-3254-4C97-A408-854445A09C21 AT noqsi DOT com> <201512081819 DOT tB8IJBrt022764 AT envy DOT delorie DOT com> <20151208205451 DOT bb2478f8722e1a885822689d AT gmail DOT com> <201512082019 DOT tB8KJOlP030874 AT envy DOT delorie DOT com> <20151208222204 DOT 6566ebed300ca2683dd71bc8 AT gmail DOT com>
	<201512082139 DOT tB8LdLuL005789 AT envy DOT delorie DOT com> <20151208230637 DOT b8eb6c22230d5a3b977c3ca1 AT gmail DOT com> <201512082229 DOT tB8MTJBj010408 AT envy DOT delorie DOT com> <alpine DOT DEB DOT 2 DOT 00 DOT 1512090417270 DOT 9035 AT igor2priv> <555EDA31-984C-4978-AE21-F74907CDC993 AT noqsi DOT com>
User-Agent:	Alpine 2.00 (DEB 1167 2008-08-23)
MIME-Version:	1.0
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 Tue, 8 Dec 2015, John Doty wrote:

>
> On Dec 8, 2015, at 5:29 PM, gedau AT igor2 DOT repo DOT hu wrote:
>
>>
>>
>> On Tue, 8 Dec 2015, DJ Delorie wrote:
>>
>>>
>>>>> So already the refdes isn't a unique identifier for a symbol...
>>>>
>>>> In such case it is a slot and the slot number could also be used?
>>>
>>> Sometimes.  Other times, like large MCUs, there are no slots, just
>>> different symbols using the same refdes, which the netlisters just
>>> merge together.
>>>
>>> At best, "refdes + pinnumber" may be a unique identifier, *except*
>>> that changing the refdes and/or pin numbers changes the identifier...
>>> and those are the things we want to change.
>>
>> Refdes+pin sounds like a good identifier on PCB-side, but not on gschem-side.
>>
>> When I use stock dual-opamp-1.sym (slotted) pin+refdes it's not unique in the sense that there are two instances of the very same power pins (4 and 8) on the schematics.
>>
>> If I connect Vcc to pin 4 on slot 1 and gnd to pin 4 on slot 2, I get the following netlist (in the gsch2pcb flow):
>>
>> Vcc U1-4
>> GND U1-4
>>
>> Same happens without slotting. However, if I take only one slot and connect both Vcc and GND to the same pin 4, only one net remains:
>>
>> GND U1-4
>>
>> This suggests the pcb gnetlist backend (or maybe gnetlist) already uses a per gschem-component identifier and refdes or refdes-pin doesn't fully identify a component or a pin.
>
> What happens in this case depends on how the back end retrieves the connection information from the core. My check-duplicates back end will detect such cases. http://www.gedasymbols.org/user/john_doty/tools/check-duplicates.html
>

Clear, and I do understand and respect your point of view on why this is a 
good thing. However, my opinion on what the desired behavior should be 
differs, and I find the current situation (not this specific example, but 
the whole "we don't want to have a common way of identifying things") a 
bad thing that does increase confusion big times for a marginal gain on 
flexibility.

- Raw text -