[Mondrian] JPivot re-write generates Mondrian bug

[Will Gorman]

Richard,

Recently a sourceforge bug was created for this identical issue:

https://sourceforge.net/tracker/index.php?func=detail&aid=1839568&group_id=35302&atid=414613

A test case still needs to be created within Foodmart.  We plan on
fixing this before 3.0, any help is appreciated.

Thanks!

Will

On Thu, 2008-01-03 at 11:40 -0800, Richard Emberson wrote:
> Mondrian Version:
> Title: mondrian
> Version: 2.4.2.9831
> Vendor: Pentaho
> 
> 
> Via JPivot query re-writing, what seems to be valid MDX
> generates a Mondrian Error:
> 
> This is for a private schema and dataset.
> 
> Steps:
> 
> 1) In JPivot execute MDX.
> 2) Then add "Store" dimension to the “Rows” and click “OK”.
> 3) When the view re-populates, click on the “+” to expand the “All Stores”
> 
> Result:
> 
> Caused by: mondrian.olap.MondrianException: Mondrian Error:Internal error:
> Error while executing query
> [select {[Time].[All Time]} ON COLUMNS,
>    Hierarchize(Union(
>      Crossjoin(
>          {[Product].[All Products]},
>          {[Store].[All Stores]}),
>      Crossjoin(
>          {[Product].[All Products]},
>          [Store].[All Stores].Children)
>      )) ON ROWS
> from [Store Shipments]
> where [Measures].[Value]
>         at
> mondrian.resource.MondrianResource$_Def0.ex(MondrianResource.java:755)
>         at mondrian.olap.Util.newInternal(Util.java:1175)
>         at mondrian.olap.Util.newError(Util.java:1191)
>         at mondrian.rolap.RolapConnection.execute(RolapConnection.java:445)
>         at 
> com.tonbeller.jpivot.mondrian.MondrianModel.getResult(MondrianModel.java:280)
>         ... 37 more
> Caused by: mondrian.olap.MondrianException: Mondrian Error:Internal error:
> assert failed
>         at 
> mondrian.resource.MondrianResource$_Def0.ex(MondrianResource.java:752)
>         at mondrian.olap.Util.newInternal(Util.java:1168)
>         at mondrian.olap.Util.assertTrue(Util.java:1149)
>         at 
> mondrian.olap.fun.FunUtil$HierarchizeArrayComparator.compare(FunUtil.java:2155)
>         at 
> mondrian.olap.fun.FunUtil$ArrayComparator.compare(FunUtil.java:2014)
>         at java.util.Arrays.mergeSort(Arrays.java:1284)
>         at java.util.Arrays.mergeSort(Arrays.java:1295)
>         at java.util.Arrays.mergeSort(Arrays.java:1296)
>         at java.util.Arrays.mergeSort(Arrays.java:1296)
>         at java.util.Arrays.mergeSort(Arrays.java:1296)
>         at java.util.Arrays.mergeSort(Arrays.java:1296)
>         at java.util.Arrays.mergeSort(Arrays.java:1295)
>         at java.util.Arrays.sort(Arrays.java:1223)
>         at java.util.Collections.sort(Collections.java:159)
>         at mondrian.olap.fun.FunUtil.hierarchize(FunUtil.java:520)
>         at 
> mondrian.olap.fun.HierarchizeFunDef$1.evaluateList(HierarchizeFunDef.java:49)
>         at 
> mondrian.calc.impl.AbstractListCalc.evaluate(AbstractListCalc.java:67)
>         at mondrian.rolap.RolapResult.executeAxis(RolapResult.java:690)
>         at mondrian.rolap.RolapResult.evalLoad(RolapResult.java:559)
>         at mondrian.rolap.RolapResult.loadMembers(RolapResult.java:534)
>         at mondrian.rolap.RolapResult.<init>(RolapResult.java:256)
>         at mondrian.rolap.RolapConnection.execute(RolapConnection.java:417)
>         ... 38 more
> 
> 
> Line numbers might vary depending upon Mondrian version but
> we have seen this error when combining Hierarchize, Union
> and Crossjoin for various versions of Mondrian.
> 
> Has anyone else ever seen this kind of error, a stack with
> multiple Arrays.mergeSort method calls being called from
> FunUtil.hierarchize and ending in
> FunUtil$HierarchizeArrayComparator.compare
> where
> Util.assertTrue(m1.equals(m2));
> is not true?
> 
> 
> 
> 
> Richard
>