Igor Lukanin

As a workaround, I can recommend to put that table name in a variable, then use it in both cubes. In the meantime, we should definitely make `{cube.sql()}` work.

@ZiggiZagga Supporting this is on the roadmap, however, there's currently no estimate as to when this might be done.

Hi @ZiggiZagga, thanks for the code example! Frankly, I don't see it as an improvement over the current situation. First, it doesn't simplify the definition of data model: instead of...

Hi @alokthakur142 👋 Thanks for a very elaborate feature request. Is there anything that prevents you from using `rollup` pre-aggregations instead of `original_sql` ones? Using `rollup` pre-aggregations would give you...

@alokthakur142 Sure! Please see https://cube.dev/docs/reference/data-model/pre-aggregations#rollup and also https://cube.dev/docs/reference/data-model/pre-aggregations#rollup_join

@wasd171 Please feel free to reopen this issue if it reproduces for you.

@paveltiunov It looks like this might be fixed at the MSSQL driver level—or at a more general level, so that `order` will be stripped from queries used to calculate totals....

Thanks for posting this, @giacomochiarella 👋 I think this is a known limitation.

@guyzilb Did you have success using it as `@knowitall/vertica-driver`? Could you please describe your experience?

Hi @rgrohitg 👋 As far as I see, the official [Databricks SQL Driver for Node.js](https://docs.databricks.com/en/dev-tools/nodejs-sql-driver.html) that Cube uses internally supports OAuth authentication, so that might in turn be supported by...