You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@directory.apache.org by Alex Karasulu <ak...@apache.org> on 2007/09/20 02:40:21 UTC
[ApacheDS] Completing all hooks for triggers
Ersin,
I was wondering what the plan is for completing all the trigger hooks.
While looking into
implementing a custom authenticator for delegating authentication to
external LDAP servers
(namely AD) it occurred to me that I could use a trigger and stored
procedure instead. I
think Emmanuel had this idea a little while ago.
I can implement a SP and fire it via an INSTEAD OF trigger on BIND rather
implementing a
custom authenticator. However we don't have INSTEAD OF triggers nor do we
have a hook
on the BIND operation. You planning on completing this stuff soon or is
this something that
you recommend I start looking into?
Alex
Re: [ApacheDS] Completing all hooks for triggers
Posted by Ersin Er <er...@gmail.com>.
On 9/20/07, Alex Karasulu <ak...@apache.org> wrote:
>
> Sounds like a plan. I want to drive forward with this as well as the
> specification process. I have some ideas on these topics namely the search
> issues.
I could not pass this..:
My recent thought lead me to the idea of Read Triggers instead of Search
Triggers... (more dots here)
Alex
>
> On 9/19/07, Ersin Er <er...@gmail.com> wrote:
> >
> > Hi,
> >
> > On 9/20/07, Alex Karasulu < akarasulu@apache.org> wrote:
> > >
> > > Ersin,
> > >
> > > I was wondering what the plan is for completing all the trigger
> > > hooks. While looking into
> > > implementing a custom authenticator for delegating authentication to
> > > external LDAP servers
> > > (namely AD) it occurred to me that I could use a trigger and stored
> > > procedure instead. I
> > > think Emmanuel had this idea a little while ago.
> >
> >
> > Basically I consider triggers as a user-side tool. A trigger is a
> > replacement for code developed at client side to keep things in sync. If we
> > want to use triggers for server extensions we need to go beyond the
> > standardization effort. I don't think that we can propose as standard those
> > all constructs needed by our server extension process.
> >
> > I can implement a SP and fire it via an INSTEAD OF trigger on BIND
> > > rather implementing a
> > > custom authenticator. However we don't have INSTEAD OF triggers nor
> > > do we have a hook
> > > on the BIND operation. You planning on completing this stuff soon or
> > > is this something that
> > > you recommend I start looking into?
> >
> >
> > As we've discussed before there are some problems with execution
> > semantics of INSTEAD OF triggers. It especially applies to the search
> > operation. If there are multiple triggers defined on an entry for a search
> > operation, which of them will be executed? This decision is important
> > because search operation returns a response to the client and this will be
> > provided by the stored procedure invoked by the trigger.
> >
> > Alex
> > >
> >
> > I can make a list of problems I have seen about implementing those hooks
> > so we can discuss on them. WDYT?
> >
> > --
> > Ersin Er
> > http://www.ersin-er.name
>
>
>
--
Ersin Er
http://www.ersin-er.name
Re: [ApacheDS] Completing all hooks for triggers
Posted by Alex Karasulu <ak...@apache.org>.
Sounds like a plan. I want to drive forward with this as well as the
specification process. I have some ideas on these topics namely the search
issues.
Alex
On 9/19/07, Ersin Er <er...@gmail.com> wrote:
>
> Hi,
>
> On 9/20/07, Alex Karasulu <ak...@apache.org> wrote:
> >
> > Ersin,
> >
> > I was wondering what the plan is for completing all the trigger hooks.
> > While looking into
> > implementing a custom authenticator for delegating authentication to
> > external LDAP servers
> > (namely AD) it occurred to me that I could use a trigger and stored
> > procedure instead. I
> > think Emmanuel had this idea a little while ago.
>
>
> Basically I consider triggers as a user-side tool. A trigger is a
> replacement for code developed at client side to keep things in sync. If we
> want to use triggers for server extensions we need to go beyond the
> standardization effort. I don't think that we can propose as standard those
> all constructs needed by our server extension process.
>
> I can implement a SP and fire it via an INSTEAD OF trigger on BIND rather
> > implementing a
> > custom authenticator. However we don't have INSTEAD OF triggers nor do
> > we have a hook
> > on the BIND operation. You planning on completing this stuff soon or is
> > this something that
> > you recommend I start looking into?
>
>
> As we've discussed before there are some problems with execution semantics
> of INSTEAD OF triggers. It especially applies to the search operation. If
> there are multiple triggers defined on an entry for a search operation,
> which of them will be executed? This decision is important because search
> operation returns a response to the client and this will be provided by the
> stored procedure invoked by the trigger.
>
> Alex
> >
>
> I can make a list of problems I have seen about implementing those hooks
> so we can discuss on them. WDYT?
>
> --
> Ersin Er
> http://www.ersin-er.name
Re: [ApacheDS] Completing all hooks for triggers
Posted by Ersin Er <er...@gmail.com>.
Hi,
On 9/20/07, Alex Karasulu <ak...@apache.org> wrote:
>
> Ersin,
>
> I was wondering what the plan is for completing all the trigger hooks.
> While looking into
> implementing a custom authenticator for delegating authentication to
> external LDAP servers
> (namely AD) it occurred to me that I could use a trigger and stored
> procedure instead. I
> think Emmanuel had this idea a little while ago.
Basically I consider triggers as a user-side tool. A trigger is a
replacement for code developed at client side to keep things in sync. If we
want to use triggers for server extensions we need to go beyond the
standardization effort. I don't think that we can propose as standard those
all constructs needed by our server extension process.
I can implement a SP and fire it via an INSTEAD OF trigger on BIND rather
> implementing a
> custom authenticator. However we don't have INSTEAD OF triggers nor do we
> have a hook
> on the BIND operation. You planning on completing this stuff soon or is
> this something that
> you recommend I start looking into?
As we've discussed before there are some problems with execution semantics
of INSTEAD OF triggers. It especially applies to the search operation. If
there are multiple triggers defined on an entry for a search operation,
which of them will be executed? This decision is important because search
operation returns a response to the client and this will be provided by the
stored procedure invoked by the trigger.
Alex
>
I can make a list of problems I have seen about implementing those hooks so
we can discuss on them. WDYT?
--
Ersin Er
http://www.ersin-er.name