Bump polars version to <1.24

ci/test_cudf_polars_polars_tests.sh

@@ -26,6 +26,8 @@ git clone https://github.com/pola-rs/polars.git --branch "${TAG}" --depth 1
 
 # Install requirements for running polars tests
 rapids-logger "Install polars test requirements"
+# TODO: Remove sed command when polars-colud supports 1.23
+sed -i '/^polars-cloud$/d' polars/py-polars/requirements-dev.txt
 rapids-pip-retry install -r polars/py-polars/requirements-dev.txt -r polars/py-polars/requirements-ci.txt
 
 # shellcheck disable=SC2317

conda/environments/all_cuda-118_arch-x86_64.yaml

@@ -66,7 +66,7 @@ dependencies:
 - pandas
 - pandas>=2.0,<2.2.4dev0
 - pandoc
-- polars>=1.20,<1.23
+- polars>=1.20,<1.24
 - pre-commit
 - ptxcompiler
 - pyarrow>=14.0.0,<20.0.0a0

conda/environments/all_cuda-128_arch-x86_64.yaml

@@ -64,7 +64,7 @@ dependencies:
 - pandas
 - pandas>=2.0,<2.2.4dev0
 - pandoc
-- polars>=1.20,<1.23
+- polars>=1.20,<1.24
 - pre-commit
 - pyarrow>=14.0.0,<20.0.0a0
 - pydata-sphinx-theme>=0.15.4

conda/recipes/cudf-polars/meta.yaml

@@ -43,7 +43,7 @@ requirements:
   run:
     - python
     - pylibcudf ={{ version }}
-    - polars >=1.20,<1.23
+    - polars >=1.20,<1.24
     - {{ pin_compatible('cuda-version', max_pin='x', min_pin='x') }}
 
 test:

dependencies.yaml

@@ -813,7 +813,7 @@ dependencies:
     common:
       - output_types: [conda, requirements, pyproject]
         packages:
-          - polars>=1.20,<1.23
+          - polars>=1.20,<1.24
   run_cudf_polars_experimental:
     common:
       - output_types: [conda, requirements, pyproject]

python/cudf_polars/cudf_polars/testing/plugin.py

@@ -197,7 +197,6 @@ def pytest_configure(config: pytest.Config) -> None:
     "tests/unit/io/test_multiscan.py::test_include_file_paths[scan_csv-write_csv]": "Need to expose include_file_paths xref: cudf#18012",
     "tests/unit/streaming/test_streaming_io.py::test_parquet_eq_statistics[False]": "Debug output on stderr doesn't match",
     # Maybe flaky, order-dependent?
-    "tests/unit/test_projections.py::test_schema_full_outer_join_projection_pd_13287": "Order-specific result check, query is correct but in different order",
     "tests/unit/test_queries.py::test_group_by_agg_equals_zero_3535": "libcudf sums all nulls to null, not zero",
 }
 

python/cudf_polars/cudf_polars/utils/dtypes.py

@@ -1,4 +1,4 @@
-# SPDX-FileCopyrightText: Copyright (c) 2024 NVIDIA CORPORATION & AFFILIATES.
+# SPDX-FileCopyrightText: Copyright (c) 2024-2025, NVIDIA CORPORATION & AFFILIATES.
 # SPDX-License-Identifier: Apache-2.0
 
 """Datatype utilities."""
@@ -71,7 +71,9 @@ def can_cast(from_: plc.DataType, to: plc.DataType) -> bool:
     -------
     True if casting is supported, False otherwise
     """
-    has_empty = from_.id() == plc.TypeId.EMPTY or to.id() == plc.TypeId.EMPTY
+    to_is_empty = to.id() == plc.TypeId.EMPTY
+    from_is_empty = from_.id() == plc.TypeId.EMPTY
+    has_empty = to_is_empty or from_is_empty
     return (
         (
             from_ == to
@@ -84,8 +86,16 @@ def can_cast(from_: plc.DataType, to: plc.DataType) -> bool:
                 )
             )
         )
-        or (from_.id() == plc.TypeId.STRING and is_numeric_not_bool(to))
-        or (to.id() == plc.TypeId.STRING and is_numeric_not_bool(from_))
+        or (
+            from_.id() == plc.TypeId.STRING
+            and not to_is_empty
+            and is_numeric_not_bool(to)
+        )
+        or (
+            to.id() == plc.TypeId.STRING
+            and not from_is_empty
+            and is_numeric_not_bool(from_)
+        )
     )
 
 

python/cudf_polars/pyproject.toml

@@ -19,7 +19,7 @@ authors = [
 license = { text = "Apache 2.0" }
 requires-python = ">=3.10"
 dependencies = [
-    "polars>=1.20,<1.23",
+    "polars>=1.20,<1.24",
     "pylibcudf==25.4.*,>=0.0.0a0",
 ] # This list was generated by `rapids-dependency-file-generator`. To make changes, edit ../../dependencies.yaml and run `rapids-dependency-file-generator`.
 classifiers = [