Date: Mon, 25 Sep 2000 16:09:50 -0700 (PDT) From: Doug Barton <DougB@gorean.org> To: freebsd-ports@freebsd.org Subject: Random patch-name thoughts Message-ID: <Pine.BSF.4.21.0009251558170.10480-100000@dt051n37.san.rr.com>
next in thread | raw e-mail | index | archive | help
I was playing around with a port today and had one of those weird
flashes of insight. There are (at least) 3 classes of patches that we have
for ports
1. Patches to squeeze things into our own directory structure
(/usr/local/, <foo>/www/, etc.).
2. Patches to fix portability problems.
3. Patches to fix real bugs/security issues.
Now, in an ideal world, only class 1 would be permanent, since the other
patches would be contributed back to the source project and
incorporated. Also in an ideal world, most of the items in 1 would be
better handled with the appropriate configure options (where
available) and/or some sed magic in the Makefile. A good example of this
(if I may say so) is the textproc/htdig Makefile that billf and I worked
on.
Now, what's the point of this? I personally like to review patches
for a port before I install it. I think it might be valuable in terms of
history, ease of future debugging, etc. to classify patches with
distinctive category names. Personally I'd like to eliminate patches in
category 1 wherever possible, but that's a bias. Something like:
1. patch-local-aa
2. patch-port-aa
3. patch-bug-aa
3a. patch-security-aa
etc.
Normally I'd just sit on my hands with this kind of idea, but
since we're talking about "Ports, The Next Generation" anyway....
Doug
--
"The dead cannot be seduced."
- Kai, "Lexx"
Do YOU Yahoo!?
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0009251558170.10480-100000>