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=O1CiAe0ODFm2mFc9nrNpiry8qeVeacbOFB/9y0H9pF8=;
        b=HcjtpixdfoxfODpgekb9qbpw/FFLjNThc3jtnBUc1WhFB2GyttirVeRkzV4O3+dDZx
         1f2Oi4YDvlu2i8RkeliSGYFmZvpW07zC3zLj7WH4P9Qoh6OTYA6u9ljqamY/jh2jMF6l
         55d4TvXDjdRUCoJAEblCXI5sTFyPAKO98X7ItIcGECLE1ZpeaR8AZQxAEZ4QV+jpOU3Q
         BnHYJf2Z7T9geHhUA026HEqtm1n6T3iHtw1coZGAYuXfYuoXTBmh6cRIDHNeIhLng16q
         flyhciwSVhIohQIz+HEiNb1gp0WfL4JtMc6Ty+P3AgcTrglGiPGgouoG6NcBqozn/Kbq
         PJCg==
MIME-Version: 1.0
X-Received: by 10.107.47.26 with SMTP id j26mr11978271ioo.17.1437056880193;
 Thu, 16 Jul 2015 07:28:00 -0700 (PDT)
In-Reply-To: <mo6ijf$se4$1@ger.gmane.org>
References: <1507142204 DOT AA17911 AT ivan DOT Harhan DOT ORG>
	<55A58A48 DOT 80501 AT neurotica DOT com>
	<20150715014343 DOT GA6516 AT recycle DOT lbl DOT gov>
	<mo6ijf$se4$1 AT ger DOT gmane DOT org>
Date: Thu, 16 Jul 2015 10:28:00 -0400
Message-ID: <CAOuGh89bwQxit-TYpySeN-xFLtz-Q1eSPgcYXDdVe1GYyc2MXg@mail.gmail.com>
Subject: Re: [geda-user] Re: geda-gaf on FreeBSD and probably other architectures
From: "Bob Paddock (graceindustries 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

> Apparently, both, pcb and geda-gaf do not reproduce because of time
> stamps from cpp macros:
>         https://reproducible.debian.net/index_issues.html

As I was the one that added that 'feature' to PCB I can explain why.
For years there were no changes to the version number, it was just
stuck at 1.99 with a letter added.  Sometimes the letter got updated.

When I was working on PCB I wanted to know what version I was really
testing so I added the time stamp
to the help screen.  The alternative was to rapidly running out of
letters or to have no idea at all what version was really being run
compared to what was being developed.

It is annoying when you only change a comment in the code and the
binary changes.  :-(
It is also a violation of MISRA rules of C to use the time/date macros.

It is a trade off.  Ideally I want bit-by-bit reproducible AND a
timestamp in the code.  As far as I know an impossibility in this
current temporal 'Reality'.,,

ssdeep for computing context triggered piecewise hashes (CTPH), or
fuzzy hashes, might solve the temporal 'Reality' problem?:

http://www.forensicswiki.org/wiki/Ssdeep

http://ssdeep.sourceforge.net/


In my newer code I use a 'buildnumber' to know the differences between
versions, as I explain here:

"Automatic Serial Number for Test Driven Development. Makefile Tip #0"

http://blog.softwaresafety.net/2010/08/automatic-serial-number-for-test-driven.html