fix polars pyproject.toml merge conflict

Author: davidwendt

.github/workflows/build.yaml

@@ -90,7 +90,7 @@ jobs:
       package-name: libcudf
       package-type: cpp
   wheel-build-pylibcudf:
-    needs: [wheel-publish-libcudf]
+    needs: [wheel-build-libcudf]
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]
     with:
@@ -111,7 +111,7 @@ jobs:
       package-name: pylibcudf
       package-type: python
   wheel-build-cudf:
-    needs: wheel-publish-pylibcudf
+    needs: wheel-build-pylibcudf
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]
     with:
@@ -132,7 +132,7 @@ jobs:
       package-name: cudf
       package-type: python
   wheel-build-dask-cudf:
-    needs: wheel-publish-cudf
+    needs: wheel-build-cudf
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]
     with:
@@ -155,7 +155,7 @@ jobs:
       package-name: dask_cudf
       package-type: python
   wheel-build-cudf-polars:
-    needs: wheel-publish-pylibcudf
+    needs: wheel-build-pylibcudf
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]
     with:

.pre-commit-config.yaml

@@ -41,13 +41,6 @@ repos:
                "python/cudf_polars/cudf_polars",
                "python/dask_cudf/dask_cudf"]
         pass_filenames: false
-  - repo: https://github.com/nbQA-dev/nbQA
-    rev: 1.9.1
-    hooks:
-      - id: nbqa-isort
-        # Use the cudf_kafka isort orderings in notebooks so that dask
-        # and RAPIDS packages have their own sections.
-        args: ["--settings-file=python/cudf_kafka/pyproject.toml"]
   - repo: https://github.com/pre-commit/mirrors-clang-format
     rev: v16.0.6
     hooks:
@@ -153,13 +146,11 @@ repos:
             ^CHANGELOG.md$
           )
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.8.0
+    rev: v0.9.3
     hooks:
       - id: ruff
         args: ["--fix"]
-        files: python/.*$
       - id: ruff-format
-        files: python/.*$
   - repo: https://github.com/rapidsai/pre-commit-hooks
     rev: v0.4.0
     hooks:
@@ -173,7 +164,7 @@ repos:
           )
       - id: verify-alpha-spec
   - repo: https://github.com/rapidsai/dependency-file-generator
-    rev: v1.16.0
+    rev: v1.17.0
     hooks:
       - id: rapids-dependency-file-generator
         args: ["--clean"]

ci/cudf_pandas_scripts/fetch_pandas_versions.py

@@ -1,24 +1,34 @@
-# Copyright (c) 2024, NVIDIA CORPORATION.
+# Copyright (c) 2024-2025, NVIDIA CORPORATION.
+
+import argparse
 
 import requests
-from packaging.version import Version
 from packaging.specifiers import SpecifierSet
-import argparse
+from packaging.version import Version
+
 
 def get_pandas_versions(pandas_range):
     url = "https://pypi.org/pypi/pandas/json"
     response = requests.get(url)
     data = response.json()
-    versions = [Version(v) for v in data['releases']]
+    versions = [Version(v) for v in data["releases"]]
     specifier = SpecifierSet(pandas_range.lstrip("pandas"))
     matching_versions = [v for v in versions if v in specifier]
-    matching_minors = sorted(set(".".join((str(v.major), str(v.minor))) for v in matching_versions), key=Version)
+    matching_minors = sorted(
+        set(".".join((str(v.major), str(v.minor))) for v in matching_versions),
+        key=Version,
+    )
     return matching_minors
 
+
 if __name__ == "__main__":
-    parser = argparse.ArgumentParser(description="Filter pandas versions by prefix.")
-    parser.add_argument("pandas_range", type=str, help="The version prefix to filter by.")
+    parser = argparse.ArgumentParser(
+        description="Filter pandas versions by prefix."
+    )
+    parser.add_argument(
+        "pandas_range", type=str, help="The version prefix to filter by."
+    )
     args = parser.parse_args()
 
     versions = get_pandas_versions(args.pandas_range)
-    print(','.join(versions))
+    print(",".join(versions))

ci/cudf_pandas_scripts/pandas-tests/job-summary.py

@@ -1,4 +1,4 @@
-# SPDX-FileCopyrightText: Copyright (c) 2023-2024, NVIDIA CORPORATION & AFFILIATES.
+# SPDX-FileCopyrightText: Copyright (c) 2023-2025, NVIDIA CORPORATION & AFFILIATES.
 # All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 
@@ -68,17 +68,27 @@ def emoji_failed(x):
 pr_df = pd.DataFrame.from_dict(pr_results, orient="index").sort_index()
 main_df = pd.DataFrame.from_dict(main_results, orient="index").sort_index()
 total_usage = main_df["_slow_function_call"] + main_df["_fast_function_call"]
-main_df["CPU Usage"] = ((main_df["_slow_function_call"] / total_usage) * 100.0).round(1)
-main_df["GPU Usage"] = ((main_df["_fast_function_call"] / total_usage) * 100.0).round(1)
+main_df["CPU Usage"] = (
+    (main_df["_slow_function_call"] / total_usage) * 100.0
+).round(1)
+main_df["GPU Usage"] = (
+    (main_df["_fast_function_call"] / total_usage) * 100.0
+).round(1)
 
 total_usage = pr_df["_slow_function_call"] + pr_df["_fast_function_call"]
-pr_df["CPU Usage"] = ((pr_df["_slow_function_call"] / total_usage) * 100.0).round(1)
-pr_df["GPU Usage"] = ((pr_df["_fast_function_call"] / total_usage) * 100.0).round(1)
+pr_df["CPU Usage"] = (
+    (pr_df["_slow_function_call"] / total_usage) * 100.0
+).round(1)
+pr_df["GPU Usage"] = (
+    (pr_df["_fast_function_call"] / total_usage) * 100.0
+).round(1)
 
 cpu_usage_mean = pr_df["CPU Usage"].mean().round(2)
 gpu_usage_mean = pr_df["GPU Usage"].mean().round(2)
 
-gpu_usage_rate_change = abs(pr_df["GPU Usage"].mean() - main_df["GPU Usage"].mean())
+gpu_usage_rate_change = abs(
+    pr_df["GPU Usage"].mean() - main_df["GPU Usage"].mean()
+)
 pr_df["CPU Usage"] = pr_df["CPU Usage"].fillna(0)
 pr_df["GPU Usage"] = pr_df["GPU Usage"].fillna(0)
 main_df["CPU Usage"] = main_df["CPU Usage"].fillna(0)
@@ -92,8 +102,12 @@ def emoji_failed(x):
 pr_df["CPU Usage"] = pr_df["CPU Usage"].astype(str) + "%"
 pr_df["GPU Usage"] = pr_df["GPU Usage"].astype(str) + "%"
 
-pr_df = pr_df[["total", "passed", "failed", "skipped", "CPU Usage", "GPU Usage"]]
-diff_df = diff_df[["total", "passed", "failed", "skipped", "CPU Usage", "GPU Usage"]]
+pr_df = pr_df[
+    ["total", "passed", "failed", "skipped", "CPU Usage", "GPU Usage"]
+]
+diff_df = diff_df[
+    ["total", "passed", "failed", "skipped", "CPU Usage", "GPU Usage"]
+]
 diff_df.columns = diff_df.columns + "_diff"
 diff_df["passed_diff"] = diff_df["passed_diff"].map(emoji_passed)
 diff_df["failed_diff"] = diff_df["failed_diff"].map(emoji_failed)

ci/utils/nbtestlog2junitxml.py

@@ -1,15 +1,16 @@
-# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+# Copyright (c) 2020-2025, NVIDIA CORPORATION.
 # Generate a junit-xml file from parsing a nbtest log
 
 import re
-from xml.etree.ElementTree import Element, ElementTree
-from os import path
 import string
 from enum import Enum
-
+from os import path
+from xml.etree.ElementTree import Element, ElementTree
 
 startingPatt = re.compile(r"^STARTING: ([\w\.\-]+)$")
-skippingPatt = re.compile(r"^SKIPPING: ([\w\.\-]+)\s*(\(([\w\.\-\ \,]+)\))?\s*$")
+skippingPatt = re.compile(
+    r"^SKIPPING: ([\w\.\-]+)\s*(\(([\w\.\-\ \,]+)\))?\s*$"
+)
 exitCodePatt = re.compile(r"^EXIT CODE: (\d+)$")
 folderPatt = re.compile(r"^FOLDER: ([\w\.\-]+)$")
 timePatt = re.compile(r"^real\s+([\d\.ms]+)$")
@@ -37,12 +38,8 @@ def makeFailureElement(outputLines):
 
 
 def setFileNameAttr(attrDict, fileName):
-    attrDict.update(file=fileName,
-                    classname="",
-                    line="",
-                    name="",
-                    time=""
-                   )
+    attrDict.update(file=fileName, classname="", line="", name="", time="")
+
 
 def setClassNameAttr(attrDict, className):
     attrDict["classname"] = className
@@ -76,11 +73,12 @@ def parseLog(logFile, testSuiteElement):
         testSuiteElement.attrib["timestamp"] = ""
 
         attrDict = {}
-        #setFileNameAttr(attrDict, logFile)
+        # setFileNameAttr(attrDict, logFile)
         setFileNameAttr(attrDict, "nbtest")
 
-        parserStateEnum = Enum("parserStateEnum",
-                               "newTest startingLine finishLine exitCode")
+        parserStateEnum = Enum(
+            "parserStateEnum", "newTest startingLine finishLine exitCode"
+        )
         parserState = parserStateEnum.newTest
 
         testOutput = ""
@@ -98,7 +96,9 @@ def parseLog(logFile, testSuiteElement):
                     setTimeAttr(attrDict, "0m0s")
                     skippedElement = makeTestCaseElement(attrDict)
                     message = m.group(3) or ""
-                    skippedElement.append(Element("skipped", message=message, type=""))
+                    skippedElement.append(
+                        Element("skipped", message=message, type="")
+                    )
                     testSuiteElement.append(skippedElement)
                     incrNumAttr(testSuiteElement, "skipped")
                     incrNumAttr(testSuiteElement, "tests")
@@ -160,4 +160,6 @@ def parseLog(logFile, testSuiteElement):
     testSuiteElement = Element("testsuite", name="nbtest", hostname="")
     parseLog(sys.argv[1], testSuiteElement)
     testSuitesElement.append(testSuiteElement)
-    ElementTree(testSuitesElement).write(sys.argv[1]+".xml", xml_declaration=True)
+    ElementTree(testSuitesElement).write(
+        sys.argv[1] + ".xml", xml_declaration=True
+    )

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

@@ -67,7 +67,7 @@ dependencies:
 - pandas
 - pandas>=2.0,<2.2.4dev0
 - pandoc
-- polars>=1.11,<1.18
+- polars>=1.20,<1.22
 - pre-commit
 - ptxcompiler
 - pyarrow>=14.0.0,<20.0.0a0

conda/environments/all_cuda-125_arch-x86_64.yaml

@@ -65,7 +65,7 @@ dependencies:
 - pandas
 - pandas>=2.0,<2.2.4dev0
 - pandoc
-- polars>=1.11,<1.18
+- polars>=1.20,<1.22
 - 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 >=1.11,<1.18
+    - polars >=1.20,<1.22
     - {{ pin_compatible('cuda-version', max_pin='x', min_pin='x') }}
 
 test:

cpp/benchmarks/CMakeLists.txt

@@ -428,7 +428,10 @@ ConfigureNVBench(RESHAPE_NVBENCH reshape/interleave.cpp)
 # ##################################################################################################
 # * rolling benchmark
 # ---------------------------------------------------------------------------------
-ConfigureNVBench(ROLLING_NVBENCH rolling/grouped_rolling_sum.cpp rolling/rolling_sum.cpp)
+ConfigureNVBench(
+  ROLLING_NVBENCH rolling/grouped_range_rolling_sum.cu rolling/grouped_rolling_sum.cpp
+  rolling/range_rolling_sum.cu rolling/rolling_sum.cpp
+)
 
 add_custom_target(
   run_benchmarks

cpp/benchmarks/io/parquet/parquet_reader_input.cpp

@@ -121,6 +121,10 @@ void BM_parquet_read_long_strings(nvbench::state& state)
       cycle_dtypes(d_type, num_cols), table_size_bytes{data_size}, profile);  // THIS
     auto const view = tbl->view();
 
+    // set smaller threshold to reduce file size and execution time
+    auto const threshold = 1;
+    setenv("LIBCUDF_LARGE_STRINGS_THRESHOLD", std::to_string(threshold).c_str(), 1);
+
     cudf::io::parquet_writer_options write_opts =
       cudf::io::parquet_writer_options::builder(source_sink.make_sink_info(), view)
         .compression(compression);
@@ -129,6 +133,7 @@ void BM_parquet_read_long_strings(nvbench::state& state)
   }();
 
   parquet_read_common(num_rows_written, num_cols, source_sink, state);
+  unsetenv("LIBCUDF_LARGE_STRINGS_THRESHOLD");
 }
 
 template <data_type DataType>