Or perhaps a few hours later ;-)
There's still a major todo left, and that's that I wrote the
auth_check_user_ext() function to work really well and nice, but forgot to
take care of auth_check_user(). At this point, I don't understand the
difference, except that one is used by injector.c and the other in pipe.c,
respectively. Is the list returned by auth_check_user() really just the
two lists, userids and forwards, from auth_check_user_ext() concatenated?
If that's the case, let's just delete the auth_check_user() function and
rewrite pipe.c to use _ext()... or to be lazy, do this:
int auth_check_user( const char *address, struct list *userids, int checks )
{
int occurences=0;
int c1, c2 }
int count1, count2;
struct list templist1, templist2;
struct element *tempelem1, *tempelem2;
char tempchar[AUTH_QUERY_SIZE];
list_init( templist1 );
list_init( templist2 );
occurences = auth_check_user_ext( address, templist1, templist2, checks );
tempelem1 = list_getstart( templist1 );
count1 = templist1.total_nodes;
for( c1 = 0; c1 < count1; c1++ )
{
/* _ext() gives us a list of numeric user ids, so convert to char */
snprintf( tempchar, AUTH_QUERY_SIZE, "%llu", tempelem1->data );
list_nodeadd( userids, tempchar );
tempelem1 = tempelem1->nextnode;
}
tempelem2 = list_getstart( templist2 ); count2 =
templist2.total_nodes; for( c2 = 0; c2 < count2; c2++ )
{
list_nodeadd( userids, tempelem2->data );
tempelem2 = tempelem2->nextnode;
} tempelem2 = list_getstart( templist2 );
}
Aaron
On Sun, 30 Mar 2003, Aaron
Stone wrote: count2 = templist2.total_nodes;> Just finished a full
rewrite, actually ;-) so the ldap sections are for( c2 = 0; c2 < count2;
c2++ )> now internally abstracted and a few extra config options
specifically with
{> drop-in Exchange replacement in mind... now supported are external
list_nodeadd( userids, tempelem2->data );> forwards through a Contact,
and distribution groups (naturally, it's
tempelem2 = tempelem2->nextnode;> recursive to allow a member of a
I'm busy right now cleaning up all of the printf()'s I stuck everywhere.
> Most will become trace(DEBUG) but a few were just to see the program flow.
> There's also a 12 byte leak in auth_adduser, but I can't figure out what's
> happening there; I've tried free()'ing everything in sight, but only get
> segfaults :-( I guess the good news is that the daemons never call
> auth_adduser() so it's practically a moot point.
>
> Thanks for asking, btw! I'll post the new version in a few days!
>
> Aaron
>
>
> On 27 Mar 2003, Ryan Butler wrote:
>
> > On Thu, 2003-03-27 at 04:27, Roel Rozendaal - IC&S wrote:
> > > Hi all,
> > >
> > > i regret to admit that i have lost pretty much track of the LDAP auth
> > > system :$
> > >
> > > What's the status? Is it ready to be included in cvs?
> > >
> > > regards roel
> > >
> > >
> > > ______
> >
> > I think it is/was ready to be committed and once its in there, Myself
> > and/or Jesse need to update autoconf to work with it.
> >
> >
> > Ryan
> >
> >
> > _______________________________________________
> > Dbmail-dev mailing list
> > Dbmail-dev@dbmail.org
> > http://twister.fastxs.net/mailman/listinfo/dbmail-dev
> >
>
> _______________________________________________
> Dbmail-dev mailing list
> Dbmail-dev@dbmail.org
> http://twister.fastxs.net/mailman/listinfo/dbmail-dev
>
