Install

pip install duckdq 

Import

from duckdq import client

Libraries

from supabase import create_client, Client # type: ignore
from dotenv import load_dotenv # type: ignore
import duckdb # type: ignore
import os

# env
load_dotenv()
URL = os.environ.get("API_URL")
KEY = os.environ.get("API_KEY")
BUCKET_KEY_ID = os.environ.get("BUCKET_KEY_ID")
BUCKET_SECRET = os.environ.get("BUCKET_SECRET")

# Session
session: Client = create_client(URL, KEY)

# Engine
engine = duckdb.connect(':memory:')

# Session
dq = client.APIClient(session, engine)

Job

# Name
dataset = "test" 
run_id = "2024-09-24"

# Data
engine.sql(f""" create or replace table {dataset} as select * from read_csv_auto('.csv/fake_customers.csv') limit 10000 """)

Register (initial one-time only)

dq.register(dataset)

Run

dq.run(dataset, run_id)

Env

.env File

API_KEY=xyz
API_URL=xyz

Or Export variables

export API_KEY=xyz
export API_URL=xyz

Database User

GRANT USAGE on schema "validation" to anon;
GRANT SELECT, INSERT, UPDATE, DELETE ON ALL TABLES IN SCHEMA "validation" TO anon;
GRANT ALL ON SEQUENCE validation.rule_output_rule_output_id_seq TO anon;
GRANT ALL ON SEQUENCE validation. connections_connection_id_seq TO anon;
GRANT ALL ON SEQUENCE validation.owlcheck_q_job_id_seq TO anon;
GRANT ALL ON SEQUENCE validation.job_log_log_id_seq TO anon;
GRANT ALL ON SEQUENCE validation.assignment_q_id_seq TO anon;