A new ADBC driver for Apache DataFusion is now available via dbc. Run dbc install datafusion to try it out today.

The driver supports query execution, bulk ingestion (create, append, and replace modes), catalog metadata retrieval, and prepared statements. Documentation can be found at docs.adbc-drivers.org. The driver was developed in Rust by the ADBC Driver Foundry based on the driver originally available from apache/arrow-adbc, but has been extended with new features and bug fixes.

As DataFusion is embedded, you can use the driver without any need for a connection string. For example, load it in Python with adbc-driver-manager:

from adbc_driver_manager import dbapi

with (
    dbapi.connect("datafusion") as con,
    con.cursor() as cursor,
):
    cursor.execute("SELECT 42")
    table = cursor.fetch_arrow_table()

Bug reports and feature requests are welcome through GitHub Issues in the datafusion repository in the ADBC Driver Foundry. You can also start a Discussion on GitHub or join the Columnar Community Slack.