When I try to create a slicer for an already filtered dimension I'm getting "XMLA MDX parse failed" error. See attached figure and log file "pentaho.log_problem1".
No problem occurs if no filters are applyed to the dimension.
A similar error occurs when I have a dimension slicer and I try to apply a dynamic table report filter to the same dimension. See log file "pentaho.log_problem2".
Indeed version 58 corrected some problems, but we still have some issues.
Now we can create a slicer for an already filtered dimension but the slicer and filter doesn´t behave correctly when they are "sharing" same dimension. I verified two error cases:
Case A:
For regular dimensions (string dimensions), It´s possible to change applyied filter using slicer component, but not by Excel filtering tool (see attached "case_a" pictures and log).
Case B:
For time dimensions (year, for instance), Excel filtering tool works just fine, but with slicer I can´t slice the dimension to a single value (single year, for instance). In this case I got a "Mondrian XML: java.lang.NullPointerException" (see attached "case_b" log file).
This looks like an error in Mondrian: "Mondrian XML: java.lang.NullPointerException". We usually get this with malformed schemes. Are you able to define a test case like this on the FoodMart schema? We test it working there (check the attach).
When creating a slicer for a filtered dimension, an error can occur due to various reasons, such as incorrect data model relationships, missing field references, or conflicts with existing filters. This issue often arises when the slicer is based on a dimension that has already been filtered in a report or dashboard, leading to ambiguity in data selection arche Rechner . To resolve this, ensure that the dimension is correctly linked to the fact table, remove any conflicting filters, and verify that the slicer field is compatible with the applied filters.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The error occurs due to conflicts between the slicer and existing filters on the same dimension. Try removing the filters before applying the slicer, check the MDX query for syntax issues using InchesPad: Convert Units Easily, or restructure the report logic. If the issue persists, review the logs for more details.
Last edit: jack austin 2025-02-04
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
When encountering an error while creating a slicer for a filtered dimension in a data visualization tool like Power BI or Tableau, it typically stems from conflicting filters or dependencies within the data model. A slicer is designed to allow users to interactively filter data based on a particular field. However, if the dimension being used in the slicer has already been pre-filtered—either through report-level filters, page-level filters of Schedule 1 APK, or calculated measures—it may result in ambiguous or invalid references. This can lead to issues such as blank slicers, incorrect data display, or outright errors during report execution.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
When I try to create a slicer for an already filtered dimension I'm getting "XMLA MDX parse failed" error. See attached figure and log file "pentaho.log_problem1".
No problem occurs if no filters are applyed to the dimension.
A similar error occurs when I have a dimension slicer and I try to apply a dynamic table report filter to the same dimension. See log file "pentaho.log_problem2".
Could you check this problem?
Thanks in advance,
Felipe Trevisan
Last edit: Felipe Trevisan 2015-01-14
Can you check with version 58? Should behave better.
Hi Alexandra,
Indeed version 58 corrected some problems, but we still have some issues.
Now we can create a slicer for an already filtered dimension but the slicer and filter doesn´t behave correctly when they are "sharing" same dimension. I verified two error cases:
Case A:
For regular dimensions (string dimensions), It´s possible to change applyied filter using slicer component, but not by Excel filtering tool (see attached "case_a" pictures and log).
Case B:
For time dimensions (year, for instance), Excel filtering tool works just fine, but with slicer I can´t slice the dimension to a single value (single year, for instance). In this case I got a "Mondrian XML: java.lang.NullPointerException" (see attached "case_b" log file).
Thanks in advance,
Felipe Trevisan
Last edit: Felipe Trevisan 2015-03-16
For case A:
Can you copy the attached file in:
(the best is to copy in both folders)
Then, restart Excel and it should work.
Please confirm.
For case B:
This looks like an error in Mondrian: "Mondrian XML: java.lang.NullPointerException". We usually get this with malformed schemes. Are you able to define a test case like this on the FoodMart schema? We test it working there (check the attach).
When creating a slicer for a filtered dimension, an error can occur due to various reasons, such as incorrect data model relationships, missing field references, or conflicts with existing filters. This issue often arises when the slicer is based on a dimension that has already been filtered in a report or dashboard, leading to ambiguity in data selection arche Rechner . To resolve this, ensure that the dimension is correctly linked to the fact table, remove any conflicting filters, and verify that the slicer field is compatible with the applied filters.
The error occurs due to conflicts between the slicer and existing filters on the same dimension. Try removing the filters before applying the slicer, check the MDX query for syntax issues using InchesPad: Convert Units Easily, or restructure the report logic. If the issue persists, review the logs for more details.
Last edit: jack austin 2025-02-04
When encountering an error while creating a slicer for a filtered dimension in a data visualization tool like Power BI or Tableau, it typically stems from conflicting filters or dependencies within the data model. A slicer is designed to allow users to interactively filter data based on a particular field. However, if the dimension being used in the slicer has already been pre-filtered—either through report-level filters, page-level filters of Schedule 1 APK, or calculated measures—it may result in ambiguous or invalid references. This can lead to issues such as blank slicers, incorrect data display, or outright errors during report execution.