From owner-cvs-user  Tue Aug 20 14:32:45 1996
Return-Path: owner-cvs-user
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id OAA06038
          for cvs-user-outgoing; Tue, 20 Aug 1996 14:32:45 -0700 (PDT)
Received: from mail.barrnet.net (mail.barrnet.net [131.119.246.7])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id OAA05608;
          Tue, 20 Aug 1996 14:27:55 -0700 (PDT)
Received: from rover.village.org (rover.village.org [204.144.255.49]) by mail.barrnet.net (8.7.5/MAIL-RELAY-LEN) with ESMTP id OAA25380; Tue, 20 Aug 1996 14:27:52 -0700 (PDT)
Received: from rover.village.org (localhost [127.0.0.1]) by rover.village.org (8.7.5/8.6.6) with ESMTP id PAA01081; Tue, 20 Aug 1996 15:24:33 -0600 (MDT)
Message-Id: <199608202124.PAA01081@rover.village.org>
To: Peter Wemm <peter@spinner.DIALix.COM>
Subject: Re: cvs commit: src/contrib/libpcap FREEBSD-upgrade 
Cc: Paul Traina <pst@freefall.freebsd.org>,
        CVS-committers@freefall.freebsd.org, cvs-all@freefall.freebsd.org,
        cvs-user@freefall.freebsd.org
In-reply-to: Your message of Tue, 20 Aug 1996 06:59:40 +0800
Date: Tue, 20 Aug 1996 15:24:32 -0600
From: Warner Losh <imp@village.org>
Sender: owner-cvs-user@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

: First, the -ko expand option was added to preserve the $Header$ lines in 
: the code..  The whole point of contrib is to have minimal diffs relative 
: to the "real thing" so that anybody can come in off the street and diff 
: the sources against the real release.  Specifying -ko on import for 
: src/contrib wasn't something we mentioned in the contrib guidelines, but 
: we probably should.

In CVS -ko is useless in 1.6.3, but works great in 1.8.1 for keeping
vendor branches up to date on imports.  Otherplaces it works great.

That is
	cvs import -ko
works in 1.8.1, but will leave the $xx$ stuff expanded in 1.6.3.  I
found this out the hard way keeping Linux in a vendor branch.  It is
*CRITICAL* that you do this.  Also, be on the lookout for the $Log$
directive, since it is *EVIL* and tends to screw you up when trying to
keep a series of diffs in a CVS tree.  Another reason -ko is so
critical.

Also, you may not notice the problems with -ko on the first import,
but subsequent ones have bitten me (since I didn't do a fresh checkout
after the import, since I had assumed that -ko would do the right
thing, which it fails to do in 1.6.3).

Warner

P.S.  The actual sequence is
	tar xvf something.tar
	cvs import -ko		# keywords expanded in the working dirs
	patch < some-patch
	cvs import -ko		# Bang, you're dead, you have
				# corrupted the $xx$ keywords
This has been fixed in 1.8.1.