From owner-freebsd-ports  Sat Jan 31 19:00:02 1998
Return-Path: <owner-freebsd-ports@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id TAA00784
          for freebsd-ports-outgoing; Sat, 31 Jan 1998 19:00:02 -0800 (PST)
          (envelope-from owner-freebsd-ports@FreeBSD.ORG)
Received: (from gnats@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id TAA00771;
          Sat, 31 Jan 1998 19:00:01 -0800 (PST)
          (envelope-from gnats)
Date: Sat, 31 Jan 1998 19:00:01 -0800 (PST)
Message-Id: <199802010300.TAA00771@hub.freebsd.org>
To: freebsd-ports
Cc: 
From: Tim Vanderhoek <ac199@hwcn.org>
Subject: Re: ports/5602: bsd.port.mk computing checksum ALWAYS even when not needed
Reply-To: Tim Vanderhoek <ac199@hwcn.org>
Sender: owner-freebsd-ports@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.org
X-To-Unsubscribe: mail to majordomo@FreeBSD.org "unsubscribe freebsd-ports"

The following reply was made to PR ports/5602; it has been noted by GNATS.

From: Tim Vanderhoek <ac199@hwcn.org>
To: Mikhail Teterin <mi@aldan.algebra.com>
Cc: FreeBSD-gnats-submit@FreeBSD.ORG
Subject: Re: ports/5602: bsd.port.mk computing checksum ALWAYS even when not needed
Date: Sat, 31 Jan 1998 20:52:46 -0500 (EST)

 On Fri, 30 Jan 1998, Mikhail Teterin wrote:
 
 > 	The following seems to fix the problem. May not be the best solution
 > 	though:
 > 
 > --- /usr/share/mk/bsd.port.mk.orig	Sun Jan 11 14:18:57 1998
 > +++ /usr/share/mk/bsd.port.mk	Fri Jan 30 00:07:31 1998
 
 This patch isn't to the current version of bsd.port.mk.
 
 > @@ -1384,3 +1383,3 @@
 >  				OK="false"; \
 > -			elif [ "$$CKSUM" = "$$CKSUM2" ]; then \
 > +			elif [ "`${MD5} < $$file`" = "$$CKSUM2" ]; then \
 
 And you just managed to hit one of the very few lines that has
 changed in the new version.  :-)
 
 Anyways, the problem you describe won't occur if IGNOREFILES
 is correctly set in the port makefile (search for ^CKSUMFILES
 and read that portion of bsd.port.mk to see why).  The two elif
 checks before the one you change are "assert()"-type error cases
 and shouldn't happen.
 
 If you found a port where IGNOREFILES is wrong, please file a pr
 for that.
 
 
 --
  tIM...HOEk
 OPTIMIZATION: the process of using many one-letter variables names
               hoping that the resultant code will run faster.