Subject: Re: Versions and uname output
To: Todd Vierling <>
From: Lubomir Sedlacik <>
List: tech-pkg
Date: 06/10/2003 19:46:15
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Jun 10, 2003 at 01:35:58PM -0400, Todd Vierling wrote:
> On Tue, 10 Jun 2003, Lubomir Sedlacik wrote:
> : this sounds reasonable on the first time but it doesn't cover all
> : possible scenarios.  imagine some crucial fix after RELEASE which is
> : pulled up along with the _STABLE name change.
> Then it's certainly *not* feasible to match on _STABLE, because someone v=
> well could be using a source tree before this "crucial fix", which is *al=
> named _STABLE.  How do you know which _STABLE is in use?

i understood what you mean from your previous mail, but see above.  it's
a corner case but it could happen.

> These are slightly different, mind you, in that _STABLE is tacked onto the
> prior version number, and the above are tacked onto the following version
> number, i.e.:
> ...
> 1.6.1_BETA*
> 1.6.1_RC*
> 1.6.1
> 1.6.1_STABLE   <-- still named "1.6.1_*"
> 1.6.2_ALPHA*   <-- now named "1.6.2_*"
> 1.6.2_BETA*
> 1.6.2_RC*
> 1.6.2
> 1.6.2_STABLE
> 1.6.3_ALPHA*
> ...
> Make sure the pattern you plan to propose works properly given the above.

we seem to be talking about two different things here.  i "proposed" a
pattern which simply matched whole netbsd-1-6 branch which is exactly
what was needed to fix the packages (audio/xmcd, misc/dialog).  since
then this duscussion took different course.


-- Lubomir Sedlacik <>                   --
--                  <>                   --

Content-Type: application/pgp-signature
Content-Disposition: inline

Version: GnuPG v1.2.2 (NetBSD)