datafusion_csv_registration_and_sql_aggregation_query.py

1import datafusion
2from datafusion import SessionContext
3
4# create a context
5ctx = SessionContext()
6
7# create a temporary table from a CSV file
8ctx.register_csv("example", "tests/data/example.csv")
9
10# execute a SQL query
11df = ctx.sql("SELECT a, MIN(b) FROM example GROUP BY a")
12
13# collect the results as a list of record batches
14results = df.collect()
15
16# print the results
17for batch in results:
18    print(batch)