I have a query that has a NONEMPTYCROSSJOIN on the columns and rows and
yet it is returning rows that have no data against them.

If i build a pivot table in excel bringing in the same field then the
problem goes away.

Could using virtual dimensions cause this problem as this is the only
difference between my MDX query and the excel pivot table.


Is this a known bug and are there any fixes?

I don't think the virtual dimension would cause this (althought it's
hard to say without seeing the exact environment). But my guess is that
it is more likely that Excel is using the NON EMPTY clause in front of a
standard crossjoin.

eg. NON EMPTY CROSSJOIN(...)

instead of the NONEMPTYCROSSJOIN(...) function. NONEMPTYCROSSJOIN
determines empty status by scanning the fact rows and can incorrectly
include/exclude rows in certain circumstances (mainly around calculated
members)

--
Regards
Darren Gosbell [MCSD]
Blog: http://www.geekswithblogs.net/darrengosbell

In article <1147338150.452663.155440 (AT) u72g2000cwu (DOT) googlegroups.com>,
pbuzzby (AT) gmail (DOT) com says...

What Darren is saying is to preface your ROWS and COLUMNS clause with NON
EMPTY

SELECT NON EMPTY { ... } ON ROWS,
NON EMPTY {... } ON COLUMNS
FROM [cube]

<pbuzzby (AT) gmail (DOT) com> wrote

Thanks i will give this a try today