[csw-maintainers] automated catalog promotion for packages

[Ben Walton]

Excerpts from Peter FELECAN's message of Sat Nov 12 01:07:15 -0500 2011:

> In my opinion, the entry epoch is by itself the the sufficient data and
> can be part of the package database and/or catalog.

Yes, this should be fine.  I don't think it belongs in the catalog
though so it should go in the pkgdb somewhere.  Looking it up via the
md5sum is the nicest method...it could be added to the pkginfo data or
as a separately available item.  Maciej is the best just of the
cleanest/nicest way to place this information for retrieval.

If we ever moved to an xml based catalog (I'm not proposing we do that
now), then I think I'd be more amenable to including the info directly
in the catalog as anything that didn't care about it could ignore it
without parser modifications, etc.


> > If I build against libxml2 and my new program passes the tests, a new
> > bug against libxml2 shouldn't hold me up.  But if libxml2 is also new
> > (defined by ticking clock here, but that may not be ideal?), and a bug
> > is filed against it, packages built against it should (?) be stopped
> > as well.  The idea here is that we don't want a situation where I need
> > a new SONAME to exist in the next catalog and that doesn't exist
> > because the package providing that was held up.
> >
> > To truly prevent this, we may need to really leverage the pkgdb
> > information.
> 
> This is still not clear and should be reworked. Maybe a sequence and
> state diagram is required here.

Sure.  I'll work on putting something visual together to illustrate
what I'm saying.

> > Aside from augmenting categories/types of bugs, we could also consider
> > merging all of the packages into a single item based on their bundle
> > tag.  The problem here is that this is relatively new so not all
> > packages have it.
> 
> Is the bundle concept supported today?

It is included in all packages built with GAR.  I don't know if
checkpkg is enforcing the existence of the line in pkginfo though.  It
should if we're going to start leveraging it.  Older packages won't
have this info but they also won't be up for promotion in this sense
so I think we can live with that.

> > I think Debian does something like this.  It's heading in the
> > direction of proper change logs  (a required Changelog.CSW in
> > $docdir?).  Do you see this as needing to be visible to the promotion
> > system?
> 
> As RedHat's RPMs

True.

> > As I mentioned above, we do need some mechanism to detect the
> > difference.  That can be either a drop (where we allow for the
> > cron interval to ensure this is seen by the back end) or a re-add
> > with a new version/md5sum.  The former is more fragile, imo.
> 
> I prefer the the MD5 signature but, in my opinion, the entry epoch
> is sufficient.

Ok.  I too prefer the md5 signature.

Thanks
-Ben

--
Ben Walton
Systems Programmer - CHASS
University of Toronto
C:416.407.5610 | W:416.978.4302