X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Date: Fri, 10 Jul 2015 04:59:48 +0200 (CEST)
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] Back annotation
In-Reply-To: <559EFE69.1040601@zoot.drehmel.com>
Message-ID: <alpine.DEB.2.00.1507100447390.6924@igor2priv>
References: <CAM2RGhTpfbqM7zNn72TBOjeL7B7LPT1PxSEU3+9aDdChFrPFTg AT mail DOT gmail DOT com> <alpine DOT DEB DOT 2 DOT 00 DOT 1507090507530 DOT 6924 AT igor2priv> <alpine DOT DEB DOT 2 DOT 11 DOT 1507091329350 DOT 3444 AT nimbus> <559E86A4 DOT 3040109 AT ecosensory DOT com> <DC183265-AD4B-4707-970F-1EE5D3578126 AT noqsi DOT com>
 <201507091843 DOT t69IhGF6028321 AT envy DOT delorie DOT com> <6392CE1A-AFA0-4D62-979C-3F35786422BD AT noqsi DOT com> <201507092127 DOT t69LRHRC001744 AT envy DOT delorie DOT com> <559EFE69 DOT 1040601 AT zoot DOT drehmel DOT com>
User-Agent: Alpine 2.00 (DEB 1167 2008-08-23)
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 Fri, 10 Jul 2015, Robert Drehmel wrote:

> On 07/09/2015 11:27 PM, DJ Delorie wrote:
>>> #3 is certainly growing on me. I find myself dealing with multiple
>>> layout contractors, and one of them wants footprint names like
>>> "BGA484C100P22X22_2300X2300X260". I don't think those belong in the
>>> schematics, and the others are happy with "BGA484". So, it's a
>>> flow-dependent mapping.
>> 
>> That idea was a side-effect of my "component database" blue-sky.  We really
>> want *three* main tools:
>> 
>> * schematic capture (gschem)
>> * mapping to a backend (netlister + component_db + project_ruleset)
>> * backend (pcb/sim/etc)
>> 
>> The mapping would map symbolic information (pins A,B,Y, value, etc) to
>> physical information (package-specific pinouts, simulation models,
>> etc) based on whatever relevent local rules apply.  Most of this info
>> is what's back-annotated anyway, but the backend can provide its
>> as-built data to the netlister on the fly, to merge with new schematic
>> info.
>
> A couple of years ago I wrote a parser for the pin mapping syntax you
> propose (http://www.delorie.com/pcb/pin-mapping.html) for my component
> database layer (handled in pcb).  I used a gschem-compatible file format
> and one file each for a family of components, e.g.


I have something similar too, brought one step further. I have a pinmap 
attribute with a sligtly different syntax doing the same, and a devmap 
attribute. Devmap contains a file name; the file contains random 
attributes specific to an actual device (by convention always including a 
pinmap attribute and a footprint attribute).

I have a an alternative gsch2pcb (called gsch2pcb_devmap) that first 
appliies the extra attributes, then resolves pinmaps and then run the 
original gsch2pcb on the result.

(Unfortunately it operates on the schematics too, while it should be a 
gnetlist backend. It's because of scheme: the current setup is hackish, 
but works and I use it in production; if I  started to write a proper 
backend in scheme, I wouldn't have anything working until 2030.)

Anyway, in this setup devmap is sort of a high level description, using 
a database (not in the sql sense). I place a generic n-chan fet symbol and 
instead of fooling around with pins and footprints I just attach a 
devmap=2n7002_sot23 then there's a text file called 2n7002_sot23.

From my user perspective it's simple, easy to follow, easy to maintain, 
easy to set up. My devmap files are in svn, so I don't have to reinvent 
library sync methods. I need to play with footpritns and pin mappings only 
the first time I use a device, then I can trust the devmap file.

My devmap files also contain attributes for ordering numbers at my 
favorite parts shops. It's much easier to maintain them than having them 
scattered in billions of schematics and I still don't need to use heavy 
symbols.

Regards,

Igor2