[Debtags-devel] tdb

Enrico Zini enrico@enricozini.org
Tue, 12 Apr 2005 00:43:54 +0200


--SLDf9lqlvOQaIe6s
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline

On Mon, Apr 11, 2005 at 01:12:09PM -0700, Erich Schubert wrote:

> I'd be fine with merging package names and tag names into one index,
> by using this prefix character; but I intentionally did not use
> strings in the database, but integer IDs.
> I can perform the standard set operations much easier and faster on these...
> At least I believe this is a lot faster. :-)
> And retrieving the names/descriptions for IDs is also pretty easy...

Yes, in the past I used integers as well.  However, now the tagcoll and
debtags code has gone much more complex, and maintaining an int <-> tag
and an int <-> package mapping is a layer of complexity that I'd like to
avoid.

Note that I'm already mapping Debtags' idea of packages to libapt's idea
of packages; tags with their offset in the vocabulary; everything to
their description and associated metadata [...] :)


Ciao,

Enrico

--
GPG key: 1024D/797EBFAB 2000-12-05 Enrico Zini <enrico@enricozini.org>

--SLDf9lqlvOQaIe6s
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQFCWv2q9LSwzHl+v6sRAk/RAJ96wVlJvGKzeiy7XC665jXVQOBhLACfd9Rr
pOLywD/PBZehaiqzOqYIuVs=
=PD6u
-----END PGP SIGNATURE-----

--SLDf9lqlvOQaIe6s--