test(python): Set up TPC-H benchmark tests

.gitignore

@@ -8,7 +8,6 @@
 .yarn/
 coverage.lcov
 coverage.xml
-data/
 polars/vendor
 
 # OS
@@ -32,6 +31,12 @@ __pycache__/
 .cargo/
 target/
 
+# Data
+*.csv
+*.parquet
+*.feather
+*.tbl
+
 # Project
 /docs/data/
 /docs/images/

py-polars/Makefile

@@ -109,6 +109,7 @@ clean:  ## Clean up caches and build artifacts
 	@rm -rf .mypy_cache/
 	@rm -rf .pytest_cache/
 	@$(VENV_BIN)/ruff clean
+	@rm -rf tests/data/tpch/sf*
 	@rm -f .coverage
 	@rm -f coverage.xml
 	@rm -f polars/polars.abi3.so

py-polars/tests/benchmark/conftest.py

@@ -1,7 +1,7 @@
 import pytest
 
 import polars as pl
-from tests.benchmark.datagen_groupby import generate_group_by_data
+from tests.benchmark.data import generate_group_by_data
 
 
 @pytest.fixture(scope="session")

py-polars/tests/benchmark/data/__init__.py

@@ -0,0 +1,6 @@
+"""Data generation functionality for use in the benchmarking suite."""
+
+from tests.benchmark.data.h2oai import generate_group_by_data
+from tests.benchmark.data.tpch import load_tpch_table
+
+__all__ = ["load_tpch_table", "generate_group_by_data"]

py-polars/tests/benchmark/data/h2oai/__init__.py

@@ -0,0 +1,5 @@
+"""Data generation functionality for use in the test suite."""
+
+from tests.benchmark.data.h2oai.datagen_groupby import generate_group_by_data
+
+__all__ = ["generate_group_by_data"]

py-polars/tests/benchmark/data/tpch/__init__.py

@@ -0,0 +1,5 @@
+"""Generate data for the TPC-H benchmark tests."""
+
+from tests.benchmark.data.tpch.generate_data import load_tpch_table
+
+__all__ = ["load_tpch_table"]