feat: Add q6 of tpch (#894)

tpch/execute/q6.py

@@ -0,0 +1,31 @@
+from pathlib import Path
+
+import pandas as pd
+import polars as pl
+from queries import q6
+
+lineitem = Path("data") / "lineitem.parquet"
+IO_FUNCS = {
+    "pandas": lambda x: pd.read_parquet(x, engine="pyarrow"),
+    "pandas[pyarrow]": lambda x: pd.read_parquet(
+        x, engine="pyarrow", dtype_backend="pyarrow"
+    ),
+    "polars[eager]": lambda x: pl.read_parquet(x),
+    "polars[lazy]": lambda x: pl.scan_parquet(x),
+}
+
+tool = "pandas"
+fn = IO_FUNCS[tool]
+print(q6.query(fn(lineitem)))
+
+tool = "pandas[pyarrow]"
+fn = IO_FUNCS[tool]
+print(q6.query(fn(lineitem)))
+
+tool = "polars[eager]"
+fn = IO_FUNCS[tool]
+print(q6.query(fn(lineitem)))
+
+tool = "polars[lazy]"
+fn = IO_FUNCS[tool]
+print(q6.query(fn(lineitem)).collect())

tpch/queries/q6.py

@@ -0,0 +1,25 @@
+from datetime import datetime
+
+import pandas as pd
+
+import narwhals as nw
+from narwhals.typing import FrameT
+
+pd.options.mode.copy_on_write = True
+
+
+@nw.narwhalify
+def query(line_item_ds: FrameT) -> FrameT:
+    var_1 = datetime(1994, 1, 1)
+    var_2 = datetime(1995, 1, 1)
+    var_3 = 24
+
+    return (
+        line_item_ds.filter(
+            nw.col("l_shipdate").is_between(var_1, var_2, closed="left"),
+            nw.col("l_discount").is_between(0.05, 0.07),
+            nw.col("l_quantity") < var_3,
+        )
+        .with_columns((nw.col("l_extendedprice") * nw.col("l_discount")).alias("revenue"))
+        .select(nw.sum("revenue"))
+    )