From owner-freebsd-doc Mon Jul 12 15:28:28 1999 Delivered-To: freebsd-doc@freebsd.org Received: from nothing-going-on.demon.co.uk (nothing-going-on.demon.co.uk [193.237.89.66]) by hub.freebsd.org (Postfix) with ESMTP id 5F10D152B2; Mon, 12 Jul 1999 15:28:13 -0700 (PDT) (envelope-from nik@nothing-going-on.demon.co.uk) Received: (from nik@localhost) by nothing-going-on.demon.co.uk (8.9.3/8.9.2) id XAA18383; Mon, 12 Jul 1999 23:09:01 +0100 (BST) (envelope-from nik) Date: Mon, 12 Jul 1999 23:09:00 +0100 From: Nik Clayton To: doc@freebsd.org, jdp@freebsd.org, wosch@freebsd.org Subject: Timetable for doc/ repository changes Message-ID: <19990712230859.K54911@catkin.nothing-going-on.org> References: <19990712122735.E2353@lehman.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: <19990712122735.E2353@lehman.com>; from Nik Clayton on Mon, Jul 12, 1999 at 12:27:35PM +0100 Organization: Nik at home, where there's nothing going on Sender: owner-freebsd-doc@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Folks [ Sent to -doc, John Polstra as the repository manager who'll be looking after this, and Wolfram Schneider, as this will affect the website ] Hopefully everyone's now had plenty of time to look at the script I posted a few weeks back which demonstrated exactly what changes are being made to the CVS tree. I've (I believe) calmed the fears of the Japanese translation team that the installation directory of doc/ja/ will be going away at any point in the near future, and we are hopefully in sufficient agreement that we can now progress to the actual implementation. In addition, I think (but would like confirmed) that John agrees that sliding or renaming the release* tags in doc/ja/ is now not necessary, because of how the tags are intended to be used. Because of the nature of the change, there will be a window of time during which the doc/ repository will fail to build successfully. This is unavoidable, as various changes will need to be made to the Makefiles after the files have been moved for everything to work. Obviously, if the repository fails to build it's much harder for people to continue working on the documentation, and anything that depends on the doc/ repository (for example, the web site builds) will also be affected. So I want to try and minimise this window as much as possible. For the specific case of the main web site, I'll update the "webbuild" script so that it uses an archive of the docs, rather than the live doc repository. This will allow the main FreeBSD web site to continue to build as normal, so there shouldn't be any problems there. However, there may be issues with mirrows of the FreeBSD web site, and I will be contacting the site mirror maintainers shortly to alert them of the problem. I estimate that fixing up doc/ after the repository copies have been completed will take me most of an evening. Generally, my free evenings in any given week are Monday -- Free all evening Tuesday -- Not free Wednesday -- Free all evening Thursday -- Free on all alternate Thursdays (next free is the 22nd) Friday -- Generally not free Saturday -- Generally not free Sunday -- Normally free Basically, this means any Monday, Wednesday, or Sunday evening are best for this to happen (from my point of view). John, how does this fit in with your schedule? Once we've decided a date for this to happen, I think the order of events is as follows; Action Responsible ------------------------------------------------------------------ Check out a complete copy of doc/ somewhere nik permanent on hub. Adjust "webbuild" to use this checked out copy, nik instead of trying to check out a fresh copy each time. Edit the AVAIL file, so that only I can commit to jdp doc/ Carry out the repository copy. jdp Fixup the Makefiles, perform any "cvs rm" commands nik necessary to get doc/ to build. Commit these changes. Announce a tentative all-clear. Get others to test nik the changes (Japanese team? Spanish team?) Reverse the change to AVAIL, so that others can now nik commit back to doc/ Anything that I've missed? N -- [intentional self-reference] can be easily accommodated using a blessed, non-self-referential dummy head-node whose own object destructor severs the links. -- Tom Christiansen in <375143b5@cs.colorado.edu> To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-doc" in the body of the message