Can not create hierarchy because the fields can not be found, though their levels exist

[cmariscalvo]

I am experiencing a problem while setting up a multilevel hierarchy for my cube.

First in my code, I define the hierarchies that are necessary for later defining the multilevel hierarchy:

Later in the process, I define the following multilevel hierarchy:

As you can see first the self._hierarchy_economic is executed and the first image hierarchies and levels are computed.
However when it reaches the multilevel hierarchy self.h['PnL'], I get the following error:

And these fields are actually created, as I can access them under the self.h atribute of my cube (where my hierarchies are):

Do you have a hint why this is happening? I does not make sense to me as I can access the actual hierarchies and levels that supposedly are not created.

thank you!

[OPeyrusse]

This is indeed intriguing. Can you provide a small notebook reproducing the issue?
My guess is that this is coming from the way you retrieve the definition of the levels. When it succeeds, you directly use the table. When it fails, you are reusing levels from another hierarchies.

[cmariscalvo]

I am adding another error that I found. I deleted the multilevel hierarchy as a sideroad for continuing with my project. I added the following measure:

I believe the syntax is correct, and the table column DTD PnL Native, as well as the hierarchy's level 'PlSubComponent' exist:

However I obtain the above error: Issue while updating schem Levelling. I search for information on this error on the internet as well as Atoti pages but I could not find anything.

Is this error normal? What triggers it?

I am using atoti version 0.7.1 btw.

Thank you!

[cmariscalvo]

Adding new info. I made a much simpler measure, just a single value:
self.m['DTD PnL Native.valor'] = tt.agg.single_value(self.data_store.get_table(PL_STORE)['DTD PnL Native'])

When I set up a debug pause before arriving there, and I write the measure in the debug console, first time I do it raises an error, second time, it works fine:

Maybe it is a problem while updating internal measures in Java API? Why should it work fine the second time, being the same line of code?

[OPeyrusse]

Hello @cmariscalvo ,

Thanks for reporting all those issues. However, I am a bit concerned because you are posting multiple different unrelated issues under the same GitHub issue. It makes it very hard to reply in an efficient manner.
And most of all, you are using an 18-month old version. For the community, only the latest version is supported. So, please, migrate to 0.8.11 and tell us if all your issues still exist.

[cmariscalvo]

I am sorry, you're right. I'll open a new issue with the last problem on latest version. Thank you @OPeyrusse