Merge branch 'branch-24.08' of github.com:rapidsai/cudf into fea-json…

…_spark_validation

.devcontainer/cuda12.2-conda/devcontainer.json → .devcontainer/cuda12.5-conda/devcontainer.json

@@ -3,15 +3,15 @@
     "context": "${localWorkspaceFolder}/.devcontainer",
     "dockerfile": "${localWorkspaceFolder}/.devcontainer/Dockerfile",
     "args": {
-      "CUDA": "12.2",
+      "CUDA": "12.5",
       "PYTHON_PACKAGE_MANAGER": "conda",
       "BASE": "rapidsai/devcontainers:24.08-cpp-mambaforge-ubuntu22.04"
     }
   },
   "runArgs": [
     "--rm",
     "--name",
-    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.2-conda"
+    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.5-conda"
   ],
   "hostRequirements": {"gpu": "optional"},
   "features": {
@@ -20,7 +20,7 @@
   "overrideFeatureInstallOrder": [
     "ghcr.io/rapidsai/devcontainers/features/rapids-build-utils"
   ],
-  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config,conda/pkgs,conda/${localWorkspaceFolderBasename}-cuda12.2-envs}"],
+  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config,conda/pkgs,conda/${localWorkspaceFolderBasename}-cuda12.5-envs}"],
   "postAttachCommand": ["/bin/bash", "-c", "if [ ${CODESPACES:-false} = 'true' ]; then . devcontainer-utils-post-attach-command; . rapids-post-attach-command; fi"],
   "workspaceFolder": "/home/coder",
   "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/cudf,type=bind,consistency=consistent",
@@ -29,7 +29,7 @@
     "source=${localWorkspaceFolder}/../.cache,target=/home/coder/.cache,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.config,target=/home/coder/.config,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.conda/pkgs,target=/home/coder/.conda/pkgs,type=bind,consistency=consistent",
-    "source=${localWorkspaceFolder}/../.conda/${localWorkspaceFolderBasename}-cuda12.2-envs,target=/home/coder/.conda/envs,type=bind,consistency=consistent"
+    "source=${localWorkspaceFolder}/../.conda/${localWorkspaceFolderBasename}-cuda12.5-envs,target=/home/coder/.conda/envs,type=bind,consistency=consistent"
   ],
   "customizations": {
     "vscode": {

.devcontainer/cuda12.2-pip/devcontainer.json → .devcontainer/cuda12.5-pip/devcontainer.json

@@ -3,15 +3,15 @@
     "context": "${localWorkspaceFolder}/.devcontainer",
     "dockerfile": "${localWorkspaceFolder}/.devcontainer/Dockerfile",
     "args": {
-      "CUDA": "12.2",
+      "CUDA": "12.5",
       "PYTHON_PACKAGE_MANAGER": "pip",
-      "BASE": "rapidsai/devcontainers:24.08-cpp-cuda12.2-ubuntu22.04"
+      "BASE": "rapidsai/devcontainers:24.08-cpp-cuda12.5-ubuntu22.04"
     }
   },
   "runArgs": [
     "--rm",
     "--name",
-    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.2-pip"
+    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-24.08-cuda12.5-pip"
   ],
   "hostRequirements": {"gpu": "optional"},
   "features": {
@@ -20,15 +20,15 @@
   "overrideFeatureInstallOrder": [
     "ghcr.io/rapidsai/devcontainers/features/rapids-build-utils"
   ],
-  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config/pip,local/share/${localWorkspaceFolderBasename}-cuda12.2-venvs}"],
+  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config/pip,local/share/${localWorkspaceFolderBasename}-cuda12.5-venvs}"],
   "postAttachCommand": ["/bin/bash", "-c", "if [ ${CODESPACES:-false} = 'true' ]; then . devcontainer-utils-post-attach-command; . rapids-post-attach-command; fi"],
   "workspaceFolder": "/home/coder",
   "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/cudf,type=bind,consistency=consistent",
   "mounts": [
     "source=${localWorkspaceFolder}/../.aws,target=/home/coder/.aws,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.cache,target=/home/coder/.cache,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.config,target=/home/coder/.config,type=bind,consistency=consistent",
-    "source=${localWorkspaceFolder}/../.local/share/${localWorkspaceFolderBasename}-cuda12.2-venvs,target=/home/coder/.local/share/venvs,type=bind,consistency=consistent"
+    "source=${localWorkspaceFolder}/../.local/share/${localWorkspaceFolderBasename}-cuda12.5-venvs,target=/home/coder/.local/share/venvs,type=bind,consistency=consistent"
   ],
   "customizations": {
     "vscode": {

.github/workflows/build.yaml

@@ -108,6 +108,28 @@ jobs:
       sha: ${{ inputs.sha }}
       date: ${{ inputs.date }}
       package-name: dask_cudf
+  wheel-build-cudf-polars:
+    needs: wheel-publish-cudf
+    secrets: inherit
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.08
+    with:
+      # This selects "ARCH=amd64 + the latest supported Python + CUDA".
+      matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
+      build_type: ${{ inputs.build_type || 'branch' }}
+      branch: ${{ inputs.branch }}
+      sha: ${{ inputs.sha }}
+      date: ${{ inputs.date }}
+      script: ci/build_wheel_cudf_polars.sh
+  wheel-publish-cudf-polars:
+    needs: wheel-build-cudf-polars
+    secrets: inherit
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.08
+    with:
+      build_type: ${{ inputs.build_type || 'branch' }}
+      branch: ${{ inputs.branch }}
+      sha: ${{ inputs.sha }}
+      date: ${{ inputs.date }}
+      package-name: cudf_polars
   trigger-pandas-tests:
     if: inputs.build_type == 'nightly'
     needs: wheel-build-cudf

.github/workflows/pandas-tests.yaml

@@ -19,7 +19,7 @@ jobs:
       secrets: inherit
       uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.08
       with:
-        matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.9" and .CUDA_VER == "12.2.2" ))
+        matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.9" and (.CUDA_VER | startswith("12.5.")) ))
         build_type: nightly
         branch: ${{ inputs.branch }}
         date: ${{ inputs.date }}

.github/workflows/pr.yaml

@@ -25,7 +25,8 @@ jobs:
       - docs-build
       - wheel-build-cudf
       - wheel-tests-cudf
-      - test-cudf-polars
+      - wheel-build-cudf-polars
+      - wheel-tests-cudf-polars
       - wheel-build-dask-cudf
       - wheel-tests-dask-cudf
       - devcontainer
@@ -133,17 +134,26 @@ jobs:
     with:
       build_type: pull-request
       script: ci/test_wheel_cudf.sh
-  test-cudf-polars:
+  wheel-build-cudf-polars:
     needs: wheel-build-cudf
     secrets: inherit
+    uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.08
+    with:
+      # This selects "ARCH=amd64 + the latest supported Python + CUDA".
+      matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
+      build_type: pull-request
+      script: "ci/build_wheel_cudf_polars.sh"
+  wheel-tests-cudf-polars:
+    needs: wheel-build-cudf-polars
+    secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.08
     with:
       # This selects "ARCH=amd64 + the latest supported Python + CUDA".
       matrix_filter: map(select(.ARCH == "amd64")) | group_by(.CUDA_VER|split(".")|map(tonumber)|.[0]) | map(max_by([(.PY_VER|split(".")|map(tonumber)), (.CUDA_VER|split(".")|map(tonumber))]))
       build_type: pull-request
       # This always runs, but only fails if this PR touches code in
       # pylibcudf or cudf_polars
-      script: "ci/test_cudf_polars.sh"
+      script: "ci/test_wheel_cudf_polars.sh"
   wheel-build-dask-cudf:
     needs: wheel-build-cudf
     secrets: inherit
@@ -167,7 +177,7 @@ jobs:
     uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@branch-24.08
     with:
       arch: '["amd64"]'
-      cuda: '["12.2"]'
+      cuda: '["12.5"]'
       build_command: |
         sccache -z;
         build-all -DBUILD_BENCHMARKS=ON --verbose;
@@ -186,7 +196,7 @@ jobs:
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.08
     with:
-      matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.9" and .CUDA_VER == "12.2.2" ))
+      matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.9" and (.CUDA_VER | startswith("12.5.")) ))
       build_type: pull-request
       script: ci/cudf_pandas_scripts/pandas-tests/run.sh pr
       # Hide test failures because they exceed the GITHUB_STEP_SUMMARY output limit.

.pre-commit-config.yaml

@@ -136,11 +136,6 @@ repos:
             .*test.*|
             ^CHANGELOG.md$
           )
-  - repo: https://github.com/rapidsai/dependency-file-generator
-    rev: v1.13.11
-    hooks:
-      - id: rapids-dependency-file-generator
-        args: ["--clean"]
   - repo: https://github.com/astral-sh/ruff-pre-commit
     rev: v0.4.8
     hooks:
@@ -149,7 +144,7 @@ repos:
       - id: ruff-format
         files: python/.*$
   - repo: https://github.com/rapidsai/pre-commit-hooks
-    rev: v0.0.3
+    rev: v0.2.0
     hooks:
       - id: verify-copyright
         exclude: |
@@ -158,6 +153,12 @@ repos:
             cpp/src/io/parquet/ipc/Message_generated[.]h$|
             cpp/src/io/parquet/ipc/Schema_generated[.]h$
           )
+      - id: verify-alpha-spec
+  - repo: https://github.com/rapidsai/dependency-file-generator
+    rev: v1.13.11
+    hooks:
+      - id: rapids-dependency-file-generator
+        args: ["--clean"]
 
 default_language_version:
       python: python3

CONTRIBUTING.md

@@ -71,15 +71,14 @@ for a minimal build of libcudf without using conda are also listed below.
 
 Compilers:
 
-* `gcc` version 9.3+
-* `nvcc` version 11.5+
-* `cmake` version 3.26.4+
+* `gcc` version 11.4+
+* `nvcc` version 11.8+
+* `cmake` version 3.29.6+
 
-CUDA/GPU:
+CUDA/GPU Runtime:
 
-* CUDA 11.5+
-* NVIDIA driver 450.80.02+
-* Volta architecture or better (Compute Capability >=7.0)
+* CUDA 11.4+
+* Volta architecture or better ([Compute Capability](https://docs.nvidia.com/deploy/cuda-compatibility/) >=7.0)
 
 You can obtain CUDA from
 [https://developer.nvidia.com/cuda-downloads](https://developer.nvidia.com/cuda-downloads).
@@ -105,7 +104,7 @@ Instructions for a minimal build environment without conda are included below.
 # create the conda environment (assuming in base `cudf` directory)
 # note: RAPIDS currently doesn't support `channel_priority: strict`;
 # use `channel_priority: flexible` instead
-conda env create --name cudf_dev --file conda/environments/all_cuda-122_arch-x86_64.yaml
+conda env create --name cudf_dev --file conda/environments/all_cuda-125_arch-x86_64.yaml
 # activate the environment
 conda activate cudf_dev
 ```

README.md

@@ -83,7 +83,7 @@ cuDF can be installed with conda (via [miniconda](https://docs.conda.io/projects
 
 ```bash
 conda install -c rapidsai -c conda-forge -c nvidia \
-    cudf=24.08 python=3.11 cuda-version=12.2
+    cudf=24.08 python=3.11 cuda-version=12.5
 ```
 
 We also provide [nightly Conda packages](https://anaconda.org/rapidsai-nightly) built from the HEAD

build.sh

@@ -17,7 +17,7 @@ ARGS=$*
 # script, and that this script resides in the repo dir!
 REPODIR=$(cd $(dirname $0); pwd)
 
-VALIDARGS="clean libcudf cudf cudfjar dask_cudf benchmarks tests libcudf_kafka cudf_kafka custreamz -v -g -n --pydevelop -l --allgpuarch --disable_nvtx --opensource_nvcomp  --show_depr_warn --ptds -h --build_metrics --incl_cache_stats"
+VALIDARGS="clean libcudf cudf cudfjar dask_cudf benchmarks tests libcudf_kafka cudf_kafka custreamz -v -g -n --pydevelop -l --allgpuarch --disable_nvtx --opensource_nvcomp  --show_depr_warn --ptds -h --build_metrics --incl_cache_stats --disable_large_strings"
 HELP="$0 [clean] [libcudf] [cudf] [cudfjar] [dask_cudf] [benchmarks] [tests] [libcudf_kafka] [cudf_kafka] [custreamz] [-v] [-g] [-n] [-h] [--cmake-args=\\\"<args>\\\"]
    clean                         - remove all existing build artifacts and configuration (start
                                    over)
@@ -39,6 +39,7 @@ HELP="$0 [clean] [libcudf] [cudf] [cudfjar] [dask_cudf] [benchmarks] [tests] [li
    --opensource_nvcomp           - disable use of proprietary nvcomp extensions
    --show_depr_warn              - show cmake deprecation warnings
    --ptds                        - enable per-thread default stream
+   --disable_large_strings       - disable large strings support
    --build_metrics               - generate build metrics report for libcudf
    --incl_cache_stats            - include cache statistics in build metrics report
    --cmake-args=\\\"<args>\\\"   - pass arbitrary list of CMake configuration options (escape all quotes in argument)
@@ -69,6 +70,7 @@ BUILD_DISABLE_DEPRECATION_WARNINGS=ON
 BUILD_PER_THREAD_DEFAULT_STREAM=OFF
 BUILD_REPORT_METRICS=OFF
 BUILD_REPORT_INCL_CACHE_STATS=OFF
+BUILD_DISABLE_LARGE_STRINGS=OFF
 USE_PROPRIETARY_NVCOMP=ON
 PYTHON_ARGS_FOR_INSTALL="-m pip install --no-build-isolation --no-deps --config-settings rapidsai.disable-cuda=true"
 
@@ -153,6 +155,7 @@ function buildLibCudfJniInDocker {
                 -DCUDF_ENABLE_ARROW_S3=OFF \
                 -DBUILD_TESTS=OFF \
                 -DCUDF_USE_PER_THREAD_DEFAULT_STREAM=ON \
+                -DCUDF_LARGE_STRINGS_DISABLED=ON \
                 -DRMM_LOGGING_LEVEL=OFF \
                 -DBUILD_SHARED_LIBS=OFF && \
              cmake --build . --parallel ${PARALLEL_LEVEL} && \
@@ -239,6 +242,9 @@ if [[ "${EXTRA_CMAKE_ARGS}" != *"DFIND_CUDF_CPP"* ]]; then
     EXTRA_CMAKE_ARGS="${EXTRA_CMAKE_ARGS} -DFIND_CUDF_CPP=ON"
 fi
 
+if hasArg --disable_large_strings; then
+    BUILD_DISABLE_LARGE_STRINGS="ON"
+fi
 
 # If clean given, run it prior to any other steps
 if hasArg clean; then
@@ -292,6 +298,7 @@ if buildAll || hasArg libcudf; then
           -DBUILD_BENCHMARKS=${BUILD_BENCHMARKS} \
           -DDISABLE_DEPRECATION_WARNINGS=${BUILD_DISABLE_DEPRECATION_WARNINGS} \
           -DCUDF_USE_PER_THREAD_DEFAULT_STREAM=${BUILD_PER_THREAD_DEFAULT_STREAM} \
+          -DCUDF_LARGE_STRINGS_DISABLED=${BUILD_DISABLE_LARGE_STRINGS} \
           -DCMAKE_BUILD_TYPE=${BUILD_TYPE} \
           ${EXTRA_CMAKE_ARGS}
 

ci/build_wheel_cudf_polars.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+# Copyright (c) 2023-2024, NVIDIA CORPORATION.
+
+set -euo pipefail
+
+package_dir="python/cudf_polars"
+
+./ci/build_wheel.sh ${package_dir}
+
+RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
+RAPIDS_PY_WHEEL_NAME="cudf_polars_${RAPIDS_PY_CUDA_SUFFIX}" RAPIDS_PY_WHEEL_PURE="1" rapids-upload-wheels-to-s3 ${package_dir}/dist

ci/cudf_pandas_scripts/pandas-tests/run.sh

@@ -19,7 +19,7 @@ RAPIDS_TESTS_DIR=${RAPIDS_TESTS_DIR:-"${RESULTS_DIR}/test-results"}/
 mkdir -p "${RAPIDS_TESTS_DIR}"
 
 bash python/cudf/cudf/pandas/scripts/run-pandas-tests.sh \
-  -n 10 \
+  -n 5 \
   --tb=no \
   -m "not slow" \
   --max-worker-restart=3 \

ci/release/update-version.sh

@@ -68,15 +68,18 @@ done
 # README.md update
 sed_runner "s/version == ${CURRENT_SHORT_TAG}/version == ${NEXT_SHORT_TAG}/g" README.md
 sed_runner "s/cudf=${CURRENT_SHORT_TAG}/cudf=${NEXT_SHORT_TAG}/g" README.md
+sed_runner "s/cudf=${CURRENT_SHORT_TAG}/cudf=${NEXT_SHORT_TAG}/g" python/cudf_polars/docs/overview.md
+sed_runner "s/branch-${CURRENT_SHORT_TAG}/branch-${NEXT_SHORT_TAG}/g" python/cudf_polars/docs/overview.md
 
 # Libcudf examples update
 sed_runner "s/CUDF_TAG branch-${CURRENT_SHORT_TAG}/CUDF_TAG branch-${NEXT_SHORT_TAG}/" cpp/examples/versions.cmake
 
 # CI files
-for FILE in .github/workflows/*.yaml; do
+for FILE in .github/workflows/*.yaml .github/workflows/*.yml; do
   sed_runner "/shared-workflows/ s/@.*/@branch-${NEXT_SHORT_TAG}/g" "${FILE}"
   sed_runner "s/dask-cuda.git@branch-[^\"\s]\+/dask-cuda.git@branch-${NEXT_SHORT_TAG}/g" ${FILE};
 done
+sed_runner "s/branch-[0-9]+\.[0-9]+/branch-${NEXT_SHORT_TAG}/g" ci/test_wheel_cudf_polars.sh
 
 # Java files
 NEXT_FULL_JAVA_TAG="${NEXT_SHORT_TAG}.${PATCH_PEP440}-SNAPSHOT"

ci/run_cudf_polars_pytests.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+# Copyright (c) 2024, NVIDIA CORPORATION.
+
+set -euo pipefail
+
+# It is essential to cd into python/cudf_polars as `pytest-xdist` + `coverage` seem to work only at this directory level.
+
+# Support invoking run_cudf_polars_pytests.sh outside the script directory
+cd "$(dirname "$(realpath "${BASH_SOURCE[0]}")")"/../python/cudf_polars/
+
+python -m pytest --cache-clear "$@" tests

ci/test_java.sh

@@ -39,6 +39,9 @@ EXITCODE=0
 trap "EXITCODE=1" ERR
 set +e
 
+# disable large strings
+export LIBCUDF_LARGE_STRINGS_ENABLED=0
+
 rapids-logger "Run Java tests"
 pushd java
 mvn test -B -DCUDF_JNI_ENABLE_PROFILING=OFF

ci/test_cudf_polars.sh → ci/test_wheel_cudf_polars.sh

@@ -18,21 +18,14 @@ else
 fi
 
 RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
-RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
+RAPIDS_PY_WHEEL_NAME="cudf_polars_${RAPIDS_PY_CUDA_SUFFIX}" RAPIDS_PY_WHEEL_PURE="1" rapids-download-wheels-from-s3 ./dist
 
-RESULTS_DIR=${RAPIDS_TESTS_DIR:-"$(mktemp -d)"}
-RAPIDS_TESTS_DIR=${RAPIDS_TESTS_DIR:-"${RESULTS_DIR}/test-results"}/
-mkdir -p "${RAPIDS_TESTS_DIR}"
-
-rapids-logger "Install cudf wheel"
-# echo to expand wildcard before adding `[extra]` requires for pip
-python -m pip install $(echo ./dist/cudf*.whl)[test]
-
-rapids-logger "Install polars (allow pre-release versions)"
-python -m pip install 'polars>=1.0.0a0'
+# Download the cudf built in the previous step
+RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./local-cudf-dep
+python -m pip install ./local-cudf-dep/cudf*.whl
 
 rapids-logger "Install cudf_polars"
-python -m pip install --no-deps python/cudf_polars
+python -m pip install $(echo ./dist/cudf_polars*.whl)[test]
 
 rapids-logger "Run cudf_polars tests"
 
@@ -44,13 +37,11 @@ EXITCODE=0
 trap set_exitcode ERR
 set +e
 
-python -m pytest \
-       --cache-clear \
+./ci/run_cudf_polars_pytests.sh \
        --cov cudf_polars \
        --cov-fail-under=100 \
-       --cov-config=python/cudf_polars/pyproject.toml \
-       --junitxml="${RAPIDS_TESTS_DIR}/junit-cudf_polars.xml" \
-       python/cudf_polars/tests
+       --cov-config=./pyproject.toml \
+       --junitxml="${RAPIDS_TESTS_DIR}/junit-cudf-polars.xml"
 
 trap ERR
 set -e