Query Firetiger with Python

ADBC is an API for interacting with databases using columnar data structures end-to-end. ADBC “drivers” implement the ADBC API for specific databases.

An official Python FlightSQL ADBC driver exists. Generally speaking, it is configured with a URL such as jdbc:arrow-flight-sql://{host}:{port}

Install with a command like pip install adbc-driver-flightsql.

Usage

Create a FlightSQL client:

import adbc_driver_flightsql.dbapi
import adbc_driver_manager

host = 'query.<your firetiger deployment name>.firetigerapi.com'
port = 443
tls = True # False to disable TLS
username = '<find this on the Firetiger Integrations page>'
password = '<find this on the Firetiger Integrations page>'

scheme = 'grpc+tls' if tls else 'grpc'
uri = f"{scheme}://{host}:{port}"

conn = adbc_driver_flightsql.dbapi.connect(
    uri,
    db_kwargs={
        adbc_driver_manager.DatabaseOptions.USERNAME.value: username,
        adbc_driver_manager.DatabaseOptions.PASSWORD.value: password,
    }
)

Now execute a query:

with conn.cursor() as cur:
    cur.execute('SELECT COUNT(*) FROM logs')

    table = cur.fetch_arrow_table()

The table object is a pyarrow.Table, which is a columnar data structure. If you would rather work with a Pandas DataFrame, you can convert it with table.to_pandas():

    df = table.to_pandas()

Clean up:

conn.close()