Per the document there is a workaround for the limitation that PivotTable is not supported with MongoDB as database.
The workaround is to build a Question (or Model?) to get the raw data first and have that Question (as opposed to the MongoDB tables) as a data source for the Question you want to create Pivot Table. It's great to know there is a workaround for the MongoDB limitation but I am a bit confused here.
My questions:
-
In the document, it says MongoDB currently doesnt' support Join operation but in the example below Joins are used to create the raw-data-collection Question. If Join is not supported with MongoDB why could Joins are used here in the example?
-
If this workaround is effective, can we safetly conclude that Pivot Table can be supported indirectly with MongoDB, despite with a hassle that user has to build the raw-data-collection Questions first, which is totally fine from my perspective because the additional effort is not comparable to the huge benefit Metabase provides.