Date: Sat, 19 Aug 2000 09:17:07 -0700 From: bmah@cisco.com (Bruce A. Mah) To: Will Andrews <andrews@technologist.com> Cc: Kris Kennaway <kris@FreeBSD.ORG>, Dirk Meyer <dirk.meyer@dinoex.sub.org>, freebsd-ports@FreeBSD.ORG Subject: Re: Indicating patch levels Message-ID: <200008191617.e7JGH7Q30715@bmah-freebsd-0.cisco.com> In-Reply-To: <20000817193048.A338@argon.gryphonsoft.com> References: <uj/HALBCtu@dmeyer.dinoex.sub.org> <Pine.BSF.4.21.0008160104060.90191-100000@freefall.freebsd.org> <20000817193048.A338@argon.gryphonsoft.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--==_Exmh_-1129676735P
Content-Type: text/plain; charset=us-ascii
If memory serves me right, Will Andrews wrote:
> On Wed, Aug 16, 2000 at 01:05:55AM -0700, Kris Kennaway wrote:
> > This is a fair suggestion - bsd.port.mk could only append the suffix if
> > it's >0 (I still prefer numerical revisions than arbitrarily-named
> > extensions - what happens if you have a secfix and a FreeBSD-specific
> > enhancement, do you call it -secfix1-enh1 or something? :)
I agree with Kris on the point of numerical revisions, in that it's much
easier for pkg_version to parse these rather than the arbitrary
revisions. It's also not clear whether, for example, "-enh1-bugfix2"
supercedes "-enh2-bugfix1" or not. Maybe this latter issue isn't a
problem though.
> Important FreeBSD-specific extensions can be given a PATCHLEVEL. I
> think something like this is in order:
>
> .if defined(PATCHLEVEL) && PATCHLEVEL > 0
> PKGNAME=${PORTNAME}-${PORTVERSION}-${PATCHLEVEL}
> .else
> PKGNAME=${PORTNAME}-${PORTVERSION}-${PATCHLEVEL}
> .endif
>
> Satoshi, I'm in favor of this sort of thing.
>
> Policy deciding whether to have PATCHLEVEL bumped will need to be
> decided. Kris, could you make your proposal on this policy?
My $0.2HK [1]: We want PATCHLEVEL to be bumped only for fairly
significant events [2]. There's two reasons for this: 1) We probably
don't want to give the users the impression that they should be chasing
every little "fixed a typo in this diagnostic message" patch. 2) People
generating updates for ports (me for instance) are going to forget to do
this sometimes anyways.
One problem with the bump for "fairly significant events" is that
someone needs to determine what "fairly significant" means. For ports
with an active MAINTAINER, this shouldn't be a problem. For the
others, it might be slightly problematic.
Bruce.
[1] I'm not actually in .hk, but that's the first creative currency
that I could think of.
[2] The two that immediately come to mind are fixing a published or
soon-to-be-published security vulnerability, or fixing compatability
with some underlying library which itself was updated.
--==_Exmh_-1129676735P
Content-Type: application/pgp-signature
-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
MessageID: iqi1V/vSWX6RhfmgASNjQWiHiUkX68l8
iQA/AwUBOZ6zA9jKMXFboFLDEQIM7QCeORWZSJyx9xXJP50zw49XCWQv+SYAn2TN
LmwCFkRCIXRr+FgyKD/B+F55
=M23s
-----END PGP SIGNATURE-----
--==_Exmh_-1129676735P--
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?200008191617.e7JGH7Q30715>