From owner-freebsd-ports Sun May 14 11:43:53 2000 Delivered-To: freebsd-ports@freebsd.org Received: from rock.ghis.net (rock.ghis.net [209.222.164.7]) by hub.freebsd.org (Postfix) with ESMTP id 115F837B5B1 for ; Sun, 14 May 2000 11:43:48 -0700 (PDT) (envelope-from will@blackdawn.com) Received: from argon.blackdawn.com (07-042.dial.008.popsite.net [209.69.77.42]) by rock.ghis.net (8.9.3/8.9.3) with ESMTP id LAA34933 for ; Sun, 14 May 2000 11:43:45 -0700 (PDT) Received: by argon.blackdawn.com (Postfix, from userid 1000) id 34459192F; Sun, 14 May 2000 14:43:34 -0400 (EDT) Date: Sun, 14 May 2000 14:43:34 -0400 From: Will Andrews To: FreeBSD Ports Subject: New patching policy proposal Message-ID: <20000514144334.D82488@argon.blackdawn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i X-Operating-System: FreeBSD 5.0-CURRENT i386 Sender: owner-freebsd-ports@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I've been thinking about how ports handles patches, and I've decided that it may be prudent to change our policy about how patches are made. It seems that many active development projects would be extremely happy to implement patches to their source code. Some reasons why we should: 1) We won't have to change our patches every time they update code. 2) Their users can now build CVS snapshots out of the box, without having to apply OS-specific patches. 3) We can reduce repository bloat by removing patches as they are integrated into the distributions. These reasons seem enough to me to implement a new policy on ports patching. Note that this is a proposal/idea-type thing and not something I'm forcing on you. It's just food for thought. When source code patches are made, we should use #ifdef (or #ifndef) __FreeBSD__ to select particular code sections for use under FreeBSD. If this is done, we can try implementing a script or somesuch that will send patches to development teams (based on Author email address links in pkg/DESCR) that are ready to be applied to a certain distfile. This will, hopefully, greatly reduce turnaround time in getting patches removed from our repository and into distfiles. Of course, not all patches are source code patches. We also have Makefile, configure, shell script, etc. patches. We can adopt different plans for each kind of diff. Then, say somebody updates a port. A script that runs (nightly? TBD) every so often will check the ports that have been changed recently to see if any patches were added. Then it'll look up the Author email address if available and automatically send these patches to that email address, requesting a return reply to the port's maintainer as well as ports@FreeBSD.org, in case said maintainer is inactive. What does everyone think on this one? Granted, it will take some time to convert patches to conform to this sort of standard, but I'm sure that if this is implemented, it will save a fair amount of time quadruply on developers', committers', users', and maintainers' parts. Respectfully, -- Will Andrews GCS/E/S @d- s+:+>+:- a--->+++ C++ UB++++ P+ L- E--- W+++ !N !o ?K w--- ?O M+ V-- PS+ PE++ Y+ PGP+>+++ t++ 5 X++ R+ tv+ b++>++++ DI+++ D+ G++>+++ e->++++ h! r-->+++ y? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message