X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
X-Virus-Status: Clean
X-Virus-Scanned: clamav-milter 0.98.4 at av01.lsn.net
Message-ID: <545104E3.2060008@ecosensory.com>
Date: Wed, 29 Oct 2014 10:16:51 -0500
From: John Griessen <john AT ecosensory DOT com>
User-Agent: Mozilla/5.0 (X11; Linux i686; rv:24.0) Gecko/20100101 Thunderbird/24.6.0
MIME-Version: 1.0
To: geda-user AT delorie DOT com
Subject: Re: [geda-user] schdiff
References: <544EAFDD DOT 1010200 AT ecosensory DOT com> <CACNnPRn1JMfVAAssZADq15paLWw-aEazZq_8N9cFGjy3_=UmiQ AT mail DOT gmail DOT com>
In-Reply-To: <CACNnPRn1JMfVAAssZADq15paLWw-aEazZq_8N9cFGjy3_=UmiQ@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Reply-To: geda-user AT delorie DOT com

On 10/28/2014 02:07 PM, Abhijit Kshirsagar wrote:
> Yes we use gschem as is. Graphical diffs (and well written commit
> mesages) are usually enough - we have rarely needed look closely at
> the diff of the sch.

OK.  How do you get the graphical diff?  Have you come up with an automated way?

> Plus we use multipage and hierarchical schematics to separate out
> logical entities for the most part, so that helps a lot too. IMO these
> would give a cleaner approach than having to text-diff the sch file at
> all...

Oh, I see now.  Have collaborators work on separate pages with
separations that keep them away from each other by design.

Sounds good.

On 10/28/2014 05:07 PM, Britton Kerin wrote:>
+1 +1 +1.  The question is, what can we do to get to the point where
hierarchical schematics are truly portable and sharable?
I know I make these same noises often,
sorry, but I to admit I have no real idea how to start on this.
I get the feeling no one else has a realistic
 > idea either.

To me, it needs the approach of chip design tools where
verilog or another schematic can be inside any module and
busses connect properly terminated to a module port
that is not limited to being a single conductor, but
can be a name like DACin[0:9].  Coding it seems hard, but
maybe possible.  Agreeing on what to do is, uh... harder,
and mostly just waiting
on someone to do it and announce as a fait accompli...