On 10/9/05, Todd Vierling <tv@duh.org> wrote:
> On Sun, 9 Oct 2005, Quentin Garnier wrote:
>
> > You mean, no indication besides common sense?  There are very little
> > cases where DIST_SUBDIR=3D${PKGNAME_NOREV} would make sense.
>
> No, I think the other way round doesn't typically make sense.  All the ca=
ses
> I can remember of:
>
> DIST_SUBDIR=3D${PKGNAME}
>
> were because the distfiles were *unversioned*, thus they had to be
> downloaded into a place that disambiguated the files when they changed
> [author-specified] version.  In every single such case, it really should =
be:
>
> DIST_SUBDIR=3D${PKGNAME_NOREV}
>
> I can't think of any cases off the top of my head where the PKGREVISION
> *should* be in the DIST_SUBDIR.

There are cases where the distfile is versioned but, nevertheless, the auth=
or
replaces it with a newer one sometime in the future.  The reasons vary, but
it's often to fix trivial things such as typos or small bugs.  Hence, we ne=
ed
to bump the revision to differentiate the versions and force a new fetch.

--
Julio M. Merino Vidal <jmmv84@gmail.com>
http://www.livejournal.com/users/jmmv/
The NetBSD Project - http://www.NetBSD.org/