PDA

View Full Version : [Mondrian] descendants of a member



Anton Nikitin
05-29-2007, 11:50 AM
We really need a way to optimize DESCENDANTS MDX function in our custom
environment.

Current contract for mondrian.olap.SchemaReader doesn't include any method
to get member descendants (I'm sure it did in some earlier version). Instead
there is mondrian.olap.fun.DescendantsFunDef which calculates descendants by
requesting member children recursively.

Is it possible to move this functionality into SchemaReader interface in
order to let us overriding and optimizing it?

Anton

_______________________________________________
Mondrian mailing list
Mondrian (AT) pentaho (DOT) org
http://lists.pentaho.org/mailman/listinfo/mondrian

Julian Hyde
05-29-2007, 03:11 PM
I've no objection to that as long as you provide a default
implementation that various schema readers can use. Also please use
typed collections (e.g. List<Member>) rather than arrays - we're moving
away from arrays.

Note also that we have the ability to cache expression results. <a
member>.DESCENDANTS is being called many times in the same query, you
could generate a $Cache(<a member>.DESCENDANTS)

Be sure to test this functionality with NONEMPTY=true in the context.
That always breaks everything.

Julian

> -----Original Message-----
> From: mondrian-bounces (AT) pentaho (DOT) org
> [mailto:mondrian-bounces (AT) pentaho (DOT) org] On Behalf Of Anton Nikitin
> Sent: Tuesday, May 29, 2007 8:45 AM
> To: mondrian (AT) pentaho (DOT) org
> Subject: [Mondrian] descendants of a member
>
>
> We really need a way to optimize DESCENDANTS MDX function in
> our custom
> environment.
>
> Current contract for mondrian.olap.SchemaReader doesn't
> include any method
> to get member descendants (I'm sure it did in some earlier
> version). Instead
> there is mondrian.olap.fun.DescendantsFunDef which calculates
> descendants by
> requesting member children recursively.
>
> Is it possible to move this functionality into SchemaReader
> interface in
> order to let us overriding and optimizing it?
>
> Anton
>
> _______________________________________________
> Mondrian mailing list
> Mondrian (AT) pentaho (DOT) org
> http://lists.pentaho.org/mailman/listinfo/mondrian
>

_______________________________________________
Mondrian mailing list
Mondrian (AT) pentaho (DOT) org
http://lists.pentaho.org/mailman/listinfo/mondrian

Anton Nikitin
05-30-2007, 07:10 AM
Ok. Thanks.

Anton

-----Original Message-----
From: mondrian-bounces (AT) pentaho (DOT) org [mailto:mondrian-bounces (AT) pentaho (DOT) org] On
Behalf Of Julian Hyde
Sent: Tuesday, May 29, 2007 11:10 PM
To: 'Mondrian developer mailing list'
Subject: RE: [Mondrian] descendants of a member

I've no objection to that as long as you provide a default
implementation that various schema readers can use. Also please use
typed collections (e.g. List<Member>) rather than arrays - we're moving
away from arrays.

Note also that we have the ability to cache expression results. <a
member>.DESCENDANTS is being called many times in the same query, you
could generate a $Cache(<a member>.DESCENDANTS)

Be sure to test this functionality with NONEMPTY=true in the context.
That always breaks everything.

Julian

> -----Original Message-----
> From: mondrian-bounces (AT) pentaho (DOT) org
> [mailto:mondrian-bounces (AT) pentaho (DOT) org] On Behalf Of Anton Nikitin
> Sent: Tuesday, May 29, 2007 8:45 AM
> To: mondrian (AT) pentaho (DOT) org
> Subject: [Mondrian] descendants of a member
>
>
> We really need a way to optimize DESCENDANTS MDX function in
> our custom
> environment.
>
> Current contract for mondrian.olap.SchemaReader doesn't
> include any method
> to get member descendants (I'm sure it did in some earlier
> version). Instead
> there is mondrian.olap.fun.DescendantsFunDef which calculates
> descendants by
> requesting member children recursively.
>
> Is it possible to move this functionality into SchemaReader
> interface in
> order to let us overriding and optimizing it?
>
> Anton
>
> _______________________________________________
> Mondrian mailing list
> Mondrian (AT) pentaho (DOT) org
> http://lists.pentaho.org/mailman/listinfo/mondrian
>

_______________________________________________
Mondrian mailing list
Mondrian (AT) pentaho (DOT) org
http://lists.pentaho.org/mailman/listinfo/mondrian

_______________________________________________
Mondrian mailing list
Mondrian (AT) pentaho (DOT) org
http://lists.pentaho.org/mailman/listinfo/mondrian