X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
X-Original-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=gmail.com; s=20120113;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        bh=P1FYO0kbYjQE7UDWhwOIuYNdr92QBRRsfIiih6gH53E=;
        b=UZHijpPE+95ywUtnCxNC2x5XNjK2JQRoP9SDD26wp97UijEF5CAw2LlWpA32cvxQbg
         a6Qh7AR61ujnm4HjrmoR+NqWGagOaSgQWpTh89DBjQbzNrvdG82D8g3HJS37/24UwTu7
         iiejAsd9v3LWzxPXDigr55+H/2u/Jj8iNAjwloL6ahBykWTWO+0QiwbY3onfAr2oVnPD
         HMkHF1GCu6+Dhba1S7QVK8Qc0z1oEG/iXSmJvL5DQgleXAmdlkwaTi432A/RWSzLaiQg
         OP6t0Jm6QlJkUX2Y8PKZXBOcrPKA/Wh84j+mx13RCiZHEBksmD7ODP7YLbkUBnwMlFC0
         LDOQ==
X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed;
        d=1e100.net; s=20130820;
        h=x-gm-message-state:mime-version:in-reply-to:references:date
         :message-id:subject:from:to:content-type;
        bh=P1FYO0kbYjQE7UDWhwOIuYNdr92QBRRsfIiih6gH53E=;
        b=k066iVt3uXaEevXxF1JZl05PDcopwuwivXNT2JOlUCxr5yWVJLsVMB6AbOZUAODZDj
         OwBsZtAXZSIgH1jzZ62jwNqJWzqzDsSUPRA9bSEJPVrxOAnO1bXgQ2nrzmRFhtj9n5u8
         iD2xAtCTb/d6k+3Z/JkP+/wMpNtMUic0g+vGD1pSefgvU78IqH8nsD6YEG/ANMtQ49l1
         vXQY9u/XGKQPp3r7htolS3im1WQPy+OlS3fiyX7rj0VygsbgoMu8cwRK9RtLCl/+eAkz
         VXttux/soqpZjaIe0oEC16uPp0CbHlBTvJRR6UX0TeCDpdA3yTSsZfbRzoTDbwRt59fJ
         o/QQ==
X-Gm-Message-State: AG10YOSxXUCkOvdDwzBFoCHo71J3HT77DXg4bu7u/dJ2ZqRGFwzTQjJjyAwtcaRWI1DgBWoK9MbbcpvbcS9sgg==
MIME-Version: 1.0
X-Received: by 10.194.6.98 with SMTP id z2mr19936739wjz.101.1453767808134;
 Mon, 25 Jan 2016 16:23:28 -0800 (PST)
In-Reply-To: <20160125224737.3995681053E4@turkos.aspodata.se>
References: <CAC4O8c_ACXqSG7nC56a+GsTf0sy1_4+N5ivZU37X+mb92k+n1Q AT mail DOT gmail DOT com>
	<201601252210 DOT u0PMA0uc031097 AT envy DOT delorie DOT com>
	<CAJXU7q81ec_4DRuhL9KGRm=zTCzXr+C-ghM03se5fiGWam8Yjw AT mail DOT gmail DOT com>
	<20160125224737 DOT 3995681053E4 AT turkos DOT aspodata DOT se>
Date: Mon, 25 Jan 2016 15:23:28 -0900
Message-ID: <CAC4O8c9NFEj25BMero6TSL3O-QnPTCZcgXnncQcm6zH-KNXt_Q@mail.gmail.com>
Subject: Re: [geda-user] pcb_fprintf mixes mm and mils for same output object
From: "Britton Kerin (britton DOT kerin AT gmail DOT com) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
To: geda-user AT delorie DOT com
Content-Type: text/plain; charset=UTF-8
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 Mon, Jan 25, 2016 at 1:47 PM,  <karl AT aspodata DOT se> wrote:
> Peter Clifton:
>> On 25 Jan 2016 22:11, "DJ Delorie" <dj AT delorie DOT com> wrote:
>> > The idea was to preserve the intent of the developer, without access
>> > to any data about what that intent was.
>> >
>> > Switching the file format to nanomaters would be "exact" but make it
>> > very difficult to understand the file format, especially when the
>> > design is mostly imperial units.
>>
>> Indeed, this is deliberate - and "probably" should be kept as is.
>>
>> Mixing metric and imperial is a bit of a mess, but it does make some sense.
>
> Yes, please respect the user and don't rewrite the file to some
> canonical form.

It's not a matter of respecting the user, if we were doing that we'd
look at the current mode and use that to decide what units to put in
the file.  Instead we do something random and variable based on how
the actual coordinates end up rounding.  The best this can ever do is
usually work for elements: if the part was specced in mils originally
that will probably give the shortest sig fig count, though of course
there's no guarantee of this.  In fact if the user typed in mils for a
part designed in metric, there's probably a shorter expression in
metric, so you'll end up getting that.
For non-element lines/arcs/etc. it might help with designs that
switched between imperial and metric grids during the course of the
layout work, but is otherwise useless and weird.

Going forward, if we want to preserve exactness for parts specced in
mils we should mark them somehow as having been specified that way.

Britton