Kill this thread, it shouldn't have been one ( was: Re: CYRUS = ....)

[Rob Siemborski]

On Tue, 3 Feb 2004, Peter P. Benac wrote:

> I for one would love to be able to use a SQL like MySQL to backend this
> software.  I don't have a problem with Berkeley DB, but I know MySQL,
> Oracle, and Postgres to whole lot better.  If I knew which routines actually
> did the database work I don't think I'd have much trouble converting them to
> use an SQL.  It would save me a ton of time if these routines were
> documented.

Given that most of the interesting database access done by cyrus is done
by enumerating the great majority of the database, I don't believe this is
really a way to speed anything up (perhaps duplicate delivery and tls
sessions could benefit from it).

You'd need to write a cyrusdb backend.  It should be pretty
straightforward (see lib/cyrusdb.h for the API you would need to
implement).

-Rob

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Rob Siemborski * Andrew Systems Group * Cyert Hall 207 * 412-268-7456
Research Systems Programmer * /usr/contributed Gatekeeper