Ok, the next thing I'll patch for you is to add the auth_connect() calls. I figure since I've got my sources changed anyways, I'll just diff against the distribution sources for you ;-)
Beyond that, I have the preliminary compiles-but-doesn't-work LDAP code posted already, and am now working on getting it to properly function ;-) I'm also writing an 'authtest.c' as a test suite that will link against a given authentication "plugin" and beat on all the routines a little bit. It'll be neat to have a verification process for submitted authentication backends :-) Aaron On Mon, 16 Dec 2002, Roel Rozendaal - IC&S wrote: > Great! i'll be looking into it shortly. > > Aaron Stone heeft op zondag, 15 dec 2002 om 22:17 (Europe/Amsterdam) > het volgende geschreven: > > > Roel, > > > > Here's a diff that has a pair of #ifndef blocks in it, one for pgsql > > and > > one for mysql, so that if DBMail is set to use the same connection for > > db > > and for auth, then the auth_connect() and auth_disconnect() functions > > will > > simply return 0, but if the connections are separate, then they will > > open > > up a new link to the database. This should provide the infrastructure > > required to add an auth_connect() along with every db_connect() call > > without making a mess of the same connection status. > > > > Aaron > > <use_same_conn_1.diff> > > _______________________________________________ > Dbmail mailing list > Dbmail@dbmail.org > https://mailman.fastxs.nl/mailman/listinfo/dbmail >
