On Dec 22, 2008, at 11:58 AM, Per Øyvind Karlsen wrote:
What is the DISTURL then ? Home page of "distro", or some ID-like
thing ?
No idea, ask Jeff. ;)
RPMTAG_DISTURL is a never used implementation because the requestor
(and the organizations the requestor particpated in, rpmfind.org and
RHN),
never defined the protocol that is supposed to be attached to
RPMTAG_DISTURL.
Unless you are willing to commit to attach a protocol to DistURL:,
then DistURL:
is not what you want or need.
Adcd a new tag for whatever purpose you want. Adding new tags
(particularly
arbitrary tags) takes minutes and has never been a problem. The
complexity
comes with integrating additional semantics throughout rpm seamlessly,
bot DistURL: and DistEpoch: have similar semantic complexities. The
existence
of a defined tag that is never (or hardly) used hurts nothing
whatsoever.
I chose existing DISTTAG and implemented DISTEPOCH since this was
something
Jeff suggested looong time ago when we discussed it last time like
a year ago.
FWIW, I'm quite happy to see Disttag: used. My issues are with adding
the semantic before there is consensus on adoption. Your patch is
clearly (imho) superior to %{?dist} adding to Release: and an
instrumented
well-defined framework like DistEpoch: are superior to random results
dependent on macro configuration.
Just "Not Yet" and under an #ifdef RPM_VENDOR_FOO (or equivalent)
"opt-in" mechanism for now.
The patch is mostly okay afaict as well.
73 de Jeff
______________________________________________________________________
RPM Package Manager http://rpm5.org
Developer Communication List rpm-devel@rpm5.org
