From owner-freebsd-ports  Sat Apr  1 06:48:05 1995
Return-Path: ports-owner
Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA12770 for ports-outgoing; Sat, 1 Apr 1995 06:48:05 -0800
Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id GAA12764 for <FreeBSD-Ports@FreeBSD.org>; Sat, 1 Apr 1995 06:48:02 -0800
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.6.9/8.6.9) id GAA11762; Sat, 1 Apr 1995 06:47:40 -0800
Date: Sat, 1 Apr 1995 06:47:40 -0800
Message-Id: <199504011447.GAA11762@silvia.HIP.Berkeley.EDU>
To: gary@palmer.demon.co.uk
CC: FreeBSD-Ports@FreeBSD.org
In-reply-to: <230.796709350@palmer.demon.co.uk> (message from Gary Palmer on Sat, 01 Apr 1995 05:09:10 +0100)
Subject: Re: checksum files
From: asami@cs.berkeley.edu (Satoshi Asami/=?ISO-2022-JP?B?GyRCQHUbKEI=?= =?ISO-2022-JP?B?GyRCOCsbKEIgGyRCOC0bKEI=?=)
Sender: ports-owner@FreeBSD.org
Precedence: bulk

 * Those of you who are masocists and also read the CVS commits list(s) will
 * have seen a large commit I made earlier to try and fix up most of the
 * remaining checksum-less ports. I've mainly left the print directory alone
 * 'cos of the problems Jean-Marc has had with the CTAN maintainers
 * re-timestamping files in the tarballs making the checksums fail when in
 * fact nothing has changed.

Great!

You might want to watch out for xfed, though.  It fetches all the
files by itself and tars them up (I think this happens in a couple of
other ports too, as I've seen a bunch of Fortran source in
/usr/ports/distfiles of thud).

Which means depending on the GZIP environment variable, you'll get a
different .tar.gz file.

Or maybe we can add "GZIP=--best" or something before tar...or let tar 
spew the file out to stderr and pipe it to gzip -9.  In fact, I think
I'm going to do that now. :)

Satoshi

P.S. The relevant line in Makefile....
          tar -C ${.CURDIR}/src -czf ${DISTDIR}/${TARFILE} xfed; \