Skip to main content

Community

Connector Improvement: Connector SDK needs ability to set Destination Schema

Answered

Please sign in to leave a comment.

Comments

3 comments

  • Official comment

    HI Michael,

    Can you share more information about how you would use the ability to name your connector differently to the schema that connection populates in your destination?
    I'm looking to understand the use case

    Thank you 

    Alison

    First, let me state that I am new to Fivetran and my only experience with it has been with non-database sources. To that end, it was not making sense to me or my client as to why each connector would create its own schema matching the name of the connector.  I didn't realize from the documentation that with the database sources the schema name option was just a prefix. As I spent more time with Fivetran, the purpose of the schema name prefix dawned on me as it seems Fivetran was initially meant to for database replication, not just a table here or there. The Connector SDK does differ in that a schema is not being replicated, only a single dataset is being dumped into a single table. What I would like to see is the ability to create a single schema for landing all my datasets from a single source system. This would allow me to have several connectors, all pulling from different api points, dump into different tables in the same schema.

    Thank you for your response Michael,

    We recommend creating multiple tables within the schema for one connector SDK if you want to deliver data that way - it helps ensure you don't accidentally write multiple datasets into the same table.

    We have a few examples you can review for how to do this in Connector SDK:
    * User Profiles

    * Accelo source example

    Does that help?

    Alison