11-13-2023 12:19 PM
So you've built yourself a snapshot table and things are going along swimmingly... and then BAM! you want to add another column to it.
You can't do a full reload because all that sweet history would just disappear and never by properly appreciated. Unfortunately altering the Incorta table structure *requires* a full load.
What I want though is the DB equivalent of altering the table and putting default values in for NULLs ( although really what I'll be doing is manufacturing approximate data from context specific history to populate those fields ).
I'm sure I can do this in a series of tables using the original snapshot as a source, but that means every time I make a change I have to create another, new, snapshot table.
I turn to you fellow Community members... how would you approach such a conundrum?
11-14-2023 02:53 AM
Hello RADSr,
We are nearing the completion of a new feature designed to support schema evolution, allowing for the seamless addition or removal of columns. This enhancement directly addresses your request to eliminate the need for a full load after making changes to the schema.
With this new functionality, actions such as adding or deleting columns will no longer necessitate a full load or rewriting of the parquet files.
11-14-2023 06:13 AM
Thank you @mhelmy - is this a "take it easy for the holidays and expect this as a New Year's present" or a "find your workaround now and it'll be easier later" thing? 😉
11-20-2023 10:02 AM - edited 11-21-2023 05:54 AM
Expected in Q1 2024.
05-03-2024 04:02 PM
Is this Feature added to the application @mhelmy