[Mondrian] Non existent members

Julian Hyde jhyde at pentaho.com
Wed Jun 17 20:15:15 EDT 2009

Rather than creating a member, would it be sufficient for your purposes if
it mapped non-existent members to the default member of that hierarchy? That
would seem to be a variation of
mondrian.rolap.ignoreInvalidMembersDuringQuery that would be easy to


From: mondrian-bounces at pentaho.org [mailto:mondrian-bounces at pentaho.org] On
Behalf Of Matt Campbell
Sent: Wednesday, June 17, 2009 1:22 PM
To: Mondrian developer mailing list
Subject: [Mondrian] Non existent members

We have an issue with non-existent members in our application and I'm
curious if anyone has any suggestions.  We use Cognos.  Immediately prior to
executing a report, Cognos will fire a series of queries to validate that
all members explicitly referenced really do exist.  For example, Cognos may
execute a query like the following to determine whether [Apr 2002- Mar 2003]
actually exists. 


MEMBER [Measures].[calculated] AS '1' 


{[Time Period: Incurred Rolling Year].[All Time Period: Incurred Rolling
Years].[Apr 2002 - Mar 2003]}  ON AXIS(0) 





Unfortunately, we have sets of reports that a run month after month, and
some specific members may drop out of the database.  Or users may create
reports in anticipation of members that will be available in the future.  We
would like to ignore the non-existent members (i.e. do something just like
what mondrian.rolap.ignoreInvalidMembersDuringQuery does).  Unfortunately,
the validation query will not return the member [Apr 2002 - Mar 2003] (it
returns the null member), which causes Cognos to fail validation and skip
running the report.

Our preference would be to get Cognos to skip the validation step, but there
is no way to do that.  As a fall back, we are wondering if it would make
sense to create a variation on how Mondrian handles non-existent members--
rather than map the non-existent members to the null member, create a new
member with the name specified.  Other suggestions would be gladly accepted.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.pentaho.org/pipermail/mondrian/attachments/20090617/d98b3139/attachment.html 

More information about the Mondrian mailing list