Get Free Trial

Question about filters behaviour

Javier González asked on May 27, 2020

We are using Flexmonster with OLAP cubes, and we would like to know the way filters work when several of them are added to the pivot table, in terms of priority, and also when one or more of those filters is of “Top” or “Bottom” type, because we’re not sure we understand how they work.
For example, let’s say we have one dimension on the rows, and another one on the columns. Then we apply a “normal” filter on the rows by selecting multiple members of that dimension and a “Top 5” filter on the column dimension. We think that, in thiscase, it works as follows:

  1. Applies filter on rows (non Top filter).
  2. Takes the results from 1. and then takes the Top 5 of those already filtered members.

Is this correct?
Now, what happens if there are several top/bottom filters? And if there are several dimensions on rows/columns? With more filters applied?
We would like to know if there is a general rule on how all this cases work. Or, the set of rules for each case.
We are developing a custom Drillthrough and we need to know exactly how the filters are being applied, so we can show the same data as in the pivot table.
Thank you for your help.

3 answers

Illia Yatsyshyn Illia Yatsyshyn Flexmonster May 28, 2020

Hello, Javier,
Thank you for reaching out to us.
Concerning the example with filtering by members together with the “Top” filter.
You are right about the fact that “Top” filtering will be performed based on members which left after filtering by labels. It is actual for any situation where both filtering by labels and by values are applied.
Several filters by values.
In case several filters by values are applied, they will be independent. Therefore, in case two “Top” filters are specified, the top will be chosen independently for each hierarchy, and the intersection of such tops will be displayed.
Also, our team is wondering about the reason the built-in drill-through feature does not fit your case. Please provide us with details about problems preventing you from the usage of our drill-through implementation.
We hope it helps.
Do not hesitate to contact us in case additional assistance is required.
Kind regards,

Javier González May 28, 2020

Thank you for the explanation, Illia.

We need to develope our own drillthrough because all the data we have on the MSAS is related to additional data we have on another data source. And on the drillthrough window we want to show the data from that second source, not the data from the MSAS.

So we need to know exactly what we are showing on the pivot table so we can replicate the query on the second source and obtain the corresponding data.

Also, our customers need different functionalities in that drillthrough window (such as hyperlinks, downloading the data in different formats, searching for particular data, etc.).

Illia Yatsyshyn Illia Yatsyshyn Flexmonster May 28, 2020

Hello, Javier,
Thank you for providing us with details.
Our team would like to provide additional comments that may appear useful when implementing the custom drill-through feature.
In case the non-cube data source is used as a second data provider, we recommend considering using the following approach.
Create a new instance of Flexmonster and pass the corresponding data to it when the custom drill-through needs to be opened. The flat form can be chosen so that the raw data is displayed.

  • Hyperlinks can be added to the created instance using the customizeCell API call. Please see an example we have prepared for you.
  • Exporting will be available by default using built-in Flexmonster functionality.
  • Searching can be performed using filters.

Please let us know in case further assistance is needed.
Kind regards,

Please login or Register to Submit Answer