Date: Sun, 27 Nov 2005 19:11:44 +0100 From: Alexander Leidinger <Alexander@Leidinger.net> To: freebsd-ports@freebsd.org, Ion-Mihai Tetcu <itetcu@people.tecnik93.com> Subject: Re: tips for using CVS for managing maintained ports ? Message-ID: <20051127191144.1a0c2bf4@Magellan.Leidinger.net> In-Reply-To: <20051127135120.5cb5fb5f@it.buh.tecnik93.com> References: <20051127135120.5cb5fb5f@it.buh.tecnik93.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 27 Nov 2005 13:51:20 +0200
Ion-Mihai Tetcu <itetcu@people.tecnik93.com> wrote:
> Managing all this by hand has became to costly so now I use RCS and a
> copy of the port in the same /usr/ports/category dir (I keep
> my /usr/ports up-to-date with cvsup); but I can't have subdirs (e.g.
> files) in RCS; and I would like to simplify the testing made by other
> RoFUG members on various platforms / OS versions.
You could get a copy of the CVS repository with CVS and check-out the
ports tree with cvs from this copy instead of using cvsup directly.
Then you have the full power of CVS at your hands and don't need to
worry about overwritten files. You also have a very fast way of
generating diffs or looking at the CVS history. Currently the entire
repo consumes 2.5G, ports only consumes 812M.
Another way would be doing 3 way merges. In my last job I was at home
only at the weekend, and I didn't had the possibility to use
cvsup/ssh/cvs at work. But I could use portsnap. So I kept 3 ports
trees. An old one, a development one, and a new one. The new one was
updates with portsnap. Then I did a diff between the old and the new,
merged the diffs into the development one and finally updated the old
one.
Here's how I did it:
---snip---
#!/bin/sh
#set -x
diff_exclude="--exclude INDEX --exclude INDEX-5 --exclude INDEX-6 --exclude INDEX.db --exclude CVS --exclude distfiles --exclude packages --exclude .portsnap.INDEX --exclude work"
cd /usr
if [ ! -e 3way.diff ]; then
echo '>>> Updating the Ports Collection'
portsnap update
echo '>>> Cleaning the Ports Collection from intermediate files'
find ports -name \*.orig -print -delete
# echo '>>> Cleaning the reference collection from intermediate files'
# find old_ports -name \*.orig -print -delete
echo '>>> Diffing Ports Collection'
diff -ruHdN ${diff_exclude} old_ports ports > 3way.diff
fi
cd devel_ports
echo '>>> Checking development version for conflicts'
patch -C --quiet -p1 < ../3way.diff
if [ $? -ne 0 ]; then
echo -n '>>> Code conflics, waiting 30 secs before proceeding...'
sleep 30
echo ' done'
echo '>>> Updating reference collection'
portsnap update -p /usr/old_ports
echo '>>> Patching development version, it will contain conflicts'
patch --quiet -p1 < ../3way.diff 2>&1 | tee /tmp/3way.log
else
echo '>>> Patching development version'
patch --quiet -p1 < ../3way.diff
echo '>>> Deleting intermediate/obsolete files'
find -d . \( -name \*.orig -o -name \*.rej -o -name .\#\* -o -empty \) -print -delete
rm ../3way.diff
echo '>>> Updating reference collection'
portsnap update -p /usr/old_ports
# echo '>>> Deleting intermediate files in the reference collection'
# find /usr/old_ports -name \*.orig -print -delete
echo '>>> Generating ports-WIP patch in /tmp'
cd /usr
diff -ruHdN ${diff_exclude} ports devel_ports > /tmp/ports-WIP.diff
fi
Bye,
Alexander.
--
Secret hacker rule #11: hackers read manuals.
http://www.Leidinger.net Alexander @ Leidinger.net
GPG fingerprint = C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20051127191144.1a0c2bf4>