objectif

Interroger des fichiers Parquet et DataFrames en SQL.

code minimal

import duckdb, pandas as pd

con = duckdb.connect()
con.execute("CREATE TABLE t AS SELECT 1 AS id")
df = con.execute("SELECT * FROM t").df()
print(df.to_dict(orient="records"))

utilisation

import duckdb, pandas as pd
import numpy as np

df = pd.DataFrame({"id":[1,2,3], "x":[10,20,30]})
res = duckdb.query("SELECT id, x*2 AS x2 FROM df").df()
print(res.x2.tolist())

variante(s) utile(s)

# duckdb.query("SELECT * FROM 'data/*.parquet'").df()
print("ok")

notes

• Très pratique pour prototyper des requêtes analytiques.