Snowflake and database field

[ppo-corp]

Feature Description

In Snowflake, tables fully qualified names are database_name.schema_name.table_name.

I would like to be able to use the same Snowflake connection in my env.yaml file to load tables into different databases.

Currently, I am forced to duplicate my Snowflake connections because the database parameter is mandatory and --tgt-object does not support the database_name.schema_name.table_name format, but only schema_name.table_name (in fact, if we use the database_name.schema_name.table_name format, the database name is ignored)

Desired changes:

• Make the database parameter for Snowflake connections optional.
• Manage a target object in the form database_name.schema_name.table_name

[martinburch]

What is the use case for targeting different Snowflake databases with a single connection? Do you have a single source system with a three-tiered architecture you want to Sling over to Snowflake in one go?

[ppo-corp]

Depending on the sources, I may need to load the data into database A or database B, and the same service user may have access to both databases. So I could have one replication for Snowflake that targets database A and another replication that targets database B.

Currently, even though it's not a big deal, I have to duplicate the connection in the env.yaml file with the same credentials. Most SQL clients allow connection without forcing the database name, and overall, we tend to use the full name of the tables in all of our flows (database_name.schema_name.table_name).

However, I understand that adding the database name in the target object could have an impact on the overall configuration of Sling. This is just a suggestion, and if the impact is too significant, I understand if you wish to close this suggestion.

[martinburch]

Thanks for the additional detail. I'll leave this with @flarco as I'm merely an interested user of Sling and Snowflake. I don't have any power to close suggestions 😉 Please, do not give me this power. 😂