New Source Connector: DuckDB 🦆

[aaronsteers]

Overview

We do not yet have a DuckDB source connector. Normally, DuckDB database are local files and not very useful as sources, but now they can also be remote (e.g. MotherDuck) and they can be a pass-through for other datasource (e.g. #30 and the Hugging Face Datasets).

Technical spec

You would write a new source connector which can connect to a (remote) DuckDB dataset or database, and emit records from DuckDB, allowing Airbyte users to send these to any Airbyte destination.

Notes:

• We do have a DuckDB Destination and a PyAirbyte Cache and SQLProcessor.
• It is not obvious how (or if) incremental processing should be handled for DuckDB sources. Whoever pick this task should plan to propose a path forward for this during development.

Definition of Done

• You would build a new "DuckDB" source in Python (reusing code if helpful).
• If primary keys exist, they should be registered in the catalog.
• You should use the CDK as much as possible.
• The connector should pass integration tests and acceptance tests.

[aaronsteers]

Assigning myself in order to reserve/hold for @ombhardwajj, who has the related #30.

[ombhardwajj]

Hey @aaronsteers ,you can assign it to me now!

[marcosmarxm]

@ombhardwajj what is the status of this issue?

[ombhardwajj]

@marcosmarxm Its already been a week since I am working on the Hugging Face Datasets connector. Given the time constraint of this hackathon, I don't think i'll be able to build this DuckDB connector. Hence I am un-assigning myself.

[bala-ceg]

issue #30 is related to issue #31, can you please assign this to me as well?

[bala-ceg]

@marcosmarxm @aaronsteers can you please let me know which connector development method i should follow - python cdk or lowcode cdk

[marcosmarxm]

@bala-ceg Probably you'll need to use Python CDK as the stream are going to be dynamically created