Polars

All integration guides. Use an ADBC Flight SQL connection for SQL reads, or pl.from_arrow() for downloaded Arrow data.

Install and query

Follow the Python environment and sample table setup, which includes Polars. Then run:

import polars as pl
import adbc_driver_flightsql.dbapi as sql

with sql.connect("grpc://localhost:8815", autocommit=True) as conn:
    query = "SELECT * FROM flight_tool_sample ORDER BY id"
    frame = pl.read_database(query, conn)
    print(frame)
    assert frame["id"].to_list() == [1, 2, 3]
    assert frame["label"].null_count() == 1
    for batch in pl.read_database(query, conn, iter_batches=True, batch_size=2):
        print(batch)

Keep the connection open while consuming the batch iterator. Batch sizes depend on driver behavior; do not assume each batch has exactly two rows.

Arrow results and writes

For a native dataset, follow the Python download example and use pl.from_arrow(received). SQL tables and named datasets are separate. For parameterized SQL, bind with an ADBC cursor before converting its Arrow result. For ingestion, pass frame.to_arrow() to cur.adbc_ingest with a new table name and mode="create"; Polars write APIs have not been validated here.

Validation and cleanup

Run make clients-smoke against a disposable server. The recorded Polars checks cover ordered reads, NULLs, empty results and batch iteration. They do not establish arbitrary Polars API compatibility. Use the shared fixture cleanup when done.

Write and verify a SQL table

Run in the environment installed above. This example writes the tool's Arrow output through ADBC, reads the persisted values back and drops its unique table. The live recipe run passed with the documented client requirements. It does not establish tool-native writer API support.

import uuid
import pyarrow as pa
import adbc_driver_flightsql.dbapi as sql

import polars as pl
output = pl.DataFrame({"id": [1, 2], "amount": [2.0, 4.0]}).to_arrow()
name = "guide_" + uuid.uuid4().hex
with sql.connect("grpc://localhost:8815", autocommit=True) as conn, conn.cursor() as cur:
    try:
        cur.adbc_ingest(name, output, mode="create")
        cur.execute(f'SELECT * FROM "{name}" ORDER BY id')
        assert cur.fetch_arrow_table().equals(output)
    finally:
        cur.execute(f'DROP TABLE IF EXISTS "{name}"')
        cur.fetch_arrow_table()