1 Release published by 1 person

• v24.12.00

  published Dec 11, 2024

36 Pull requests merged by 15 people

• Replace direct cudaMemcpyAsync calls with utility functions (within /include)

  #17557 merged Dec 11, 2024

• [RELEASE] cudf v24.12

  #17406 merged Dec 11, 2024

• Specify a version for rapids_logger dependency

  #17573 merged Dec 11, 2024

• Rework minhash APIs for deprecation cycle

  #17421 merged Dec 10, 2024

• Fix Dask-cuDF clip APIs

  #17509 merged Dec 10, 2024

• Update version references in workflow

  #17568 merged Dec 10, 2024

• Fix typos, rename types, and add null_probability benchmark axis for distinct

  #17546 merged Dec 10, 2024

• Use rapids-logger to generate the cudf logger

  #17307 merged Dec 10, 2024

• Simplify serialization protocols

  #17552 merged Dec 10, 2024

• Replace cudf::detail::copy_if logic with thrust::copy_if and gather

  #17520 merged Dec 10, 2024

• gate telemetry dispatch calls on TELEMETRY_ENABLED env var

  #17551 merged Dec 10, 2024

• Remove Thrust patch in favor of CMake definition for Thrust 32-bit offset types.

  #17527 merged Dec 10, 2024

• Fix CMake format in cudf/_lib/CMakeLists.txt

  #17559 merged Dec 10, 2024

• Enable rounding for Decimal32 and Decimal64 in cuDF

  #17332 merged Dec 10, 2024

• Remove unused BufferArrayFromVector

  #17549 merged Dec 9, 2024

• Remove cudf._lib.transform in favor of inlining pylibcudf

  #17505 merged Dec 9, 2024

• [JNI] remove rmm argument to set rw access for fabric handles

  #17553 merged Dec 9, 2024

• Remove cudf._lib.string.convert/split in favor of inlining pylibcudf

  #17496 merged Dec 9, 2024

• [JNI] Enables fabric handles for CUDA async memory pools

  #17526 merged Dec 9, 2024

• Document undefined behavior in div_rounding_up_safe

  #17542 merged Dec 9, 2024

• Fix nvcc-imposed UB in constexpr functions

  #17534 merged Dec 9, 2024

• Remove unused IO utilities from cudf python

  #17374 merged Dec 9, 2024

• Use cooperative-groups instead of cub warp-reduce for strings contains

  #17540 merged Dec 9, 2024

• Update cuda-python lower bounds to 12.6.2 / 11.8.5

  #17547 merged Dec 7, 2024

• Fix all null list column with missing child column in JSON reader

  #17348 merged Dec 6, 2024

• Add Parquet Reader options classes to pylibcudf

  #17464 merged Dec 6, 2024

• Remove cudf._lib.orc in favor of inlining pylibcudf

  #17466 merged Dec 6, 2024

• Remove cudf._lib.round in favor of inlining pylibcudf

  #17430 merged Dec 6, 2024

• Replaces uses of cudf._lib.Column.from_unique_ptr with pylibcudf.Column.from_libcudf

  #17517 merged Dec 6, 2024

• Remove cudf._lib.text in favor of inlining pylibcudf

  #17408 merged Dec 6, 2024

• Improve strings contains/find performance for smaller strings

  #17330 merged Dec 6, 2024

• Plumb pylibcudf.io.parquet options classes through cudf python

  #17506 merged Dec 6, 2024

• Migrate copy_column and Column.from_scalar to pylibcudf

  #17513 merged Dec 6, 2024

• Move cudf._lib.aggregation to cudf.core._internals

  #17516 merged Dec 6, 2024

• Remove cudf._lib.merge in favor of inlining pylibcudf

  #17370 merged Dec 5, 2024

• Force Thrust to use 32-bit offset type.

  #17523 merged Dec 5, 2024

28 Pull requests opened by 13 people

• Propagate failures in pandas integration tests

  #17521 opened Dec 4, 2024

• Upgrade nvcomp to 4.1.1.1

  #17522 opened Dec 5, 2024

• Add anonymous namespace to libcudf test source

  #17529 opened Dec 5, 2024

• Remove unused masked keyword in column_empty

  #17530 opened Dec 5, 2024

• Use XXHash_32 as the default hash

  #17533 opened Dec 5, 2024

• Remove cudf._lib.nvtext in favor of inlining pylibcudf

  #17535 opened Dec 6, 2024

• Create `pylibcudf.Column` from a `column_view` and an arbitrary owning object

  #17543 opened Dec 6, 2024

• remove certs and simplify telemetry summarize

  #17544 opened Dec 6, 2024

• Mark more constexpr functions as device-available

  #17545 opened Dec 6, 2024

• Move cudf._lib.copying to cudf.core._internals

  #17548 opened Dec 7, 2024

• Replace direct `cudaMemcpyAsync` calls with utility functions (within `/src`)

  #17550 opened Dec 9, 2024

• Relax inconsistent schema handling in `dask_cudf.read_parquet`

  #17554 opened Dec 9, 2024

• Remove cudf._lib.interop in favor of inlining pylibcudf

  #17555 opened Dec 9, 2024

• Remove "legacy" Dask DataFrame support from Dask cuDF

  #17558 opened Dec 9, 2024

• Add support for unary negation operator

  #17560 opened Dec 10, 2024

• Remove cudf._lib.parquet in favor of inlining pylibcudf

  #17562 opened Dec 10, 2024

• Enable all json reader options in pylibcudf read_json

  #17563 opened Dec 10, 2024

• Forward-merge branch-24.12 into branch-25.02

  #17567 opened Dec 10, 2024

• Branch 25.02 merge 24.12

  #17569 opened Dec 10, 2024

• Fix the ORC decoding bug for the timestamp data

  #17570 opened Dec 10, 2024

• Allow large strings in nvbench strings benchmarks

  #17571 opened Dec 10, 2024

• Use batched memcpy when writing ORC statistics

  #17572 opened Dec 10, 2024

• Remove cudf._lib.reduce in favor of inlining pylibcudf

  #17574 opened Dec 11, 2024

• json top level list scoping

  #17575 opened Dec 11, 2024

• Fix ctest fail running libcudf tests in a Debug build

  #17576 opened Dec 11, 2024

• Allow large strings in nvtext benchmarks

  #17579 opened Dec 11, 2024

• Expose Scalar's constructor and `Scalar#getScalarHandle()` to public

  #17580 opened Dec 11, 2024

• Remove unused code of json schema in JSON reader

  #17581 opened Dec 11, 2024

8 Issues closed by 4 people

• [BUG] `clip` not implemented for Dask-cuDF

  #17502 closed Dec 10, 2024

• [FEA][JNI] Add fabric handle support for CUDA async pools in cuDF JNI

  #17525 closed Dec 9, 2024

• [DOC] clarify safety of integer utils

  #17539 closed Dec 9, 2024

• [FEA] `read_json` should output all-nulls columns for the schema columns that do not match with the input JSON

  #17341 closed Dec 6, 2024

• [BUG] `read_json` does not output null list when the input is a list with different data type than the specified schema

  #17349 closed Dec 6, 2024

• [BUG] `SubwordTokenize` does not show up in the cuDF user guide section of the docs

  #17536 closed Dec 6, 2024

• Check performance of changing default hasher to XXHash32

  #17532 closed Dec 5, 2024

• [FEA] A Python wrapper for `to_arrow_device`.

  #17528 closed Dec 5, 2024

9 Issues opened by 7 people

• [FEA] Clean up rolling window interfaces

  #17578 opened Dec 11, 2024

• [FEA][JNI] ContiguousTable.getBuffer should increase ref count of buffer

  #17566 opened Dec 10, 2024

• [FEA] Expose IO options structs in pylibcudf

  #17565 opened Dec 10, 2024

• [FEA] Python-level support for writing compressed JSON format

  #17564 opened Dec 10, 2024

• [FEA] Add benchmarks for computing polynomials

  #17561 opened Dec 10, 2024

• [BUG] Fencepost error in unbounded rolling windows if column has `size_type::max()` entries

  #17541 opened Dec 6, 2024

• [FEA] Support Polars unary function `negate`.

  #17538 opened Dec 6, 2024

• [FEA] Update the default hasher in libcudf

  #17531 opened Dec 5, 2024

• [FEA] Provide an escape hatch to libraries that operate `cudf` aware

  #17524 opened Dec 5, 2024

31 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Support reading bloom filters from Parquet files and filter row groups using them

  #17289 commented on Dec 11, 2024 • 25 new comments

• Avoid converting Decimal32/Decimal64 in `to_arrow` and `from_arrow` APIs

  #17422 commented on Dec 11, 2024 • 12 new comments

• Add partition-wise `Select` support to cuDF-Polars

  #17495 commented on Dec 6, 2024 • 3 new comments

• Remove cudf._lib.csv in favor in inlining pylibcudf

  #17485 commented on Dec 10, 2024 • 3 new comments

• Precompute AST arity

  #17234 commented on Dec 10, 2024 • 2 new comments

• Introduce utility function to calculate inputs to rolling_window

  #16305 commented on Dec 9, 2024 • 1 new comment

• [WIP] Add multi-partition `Join` support to cuDF-Polars

  #17518 commented on Dec 6, 2024 • 0 new comments

• Add basic multi-partition `GroupBy` support to cuDF-Polars

  #17503 commented on Dec 6, 2024 • 0 new comments

• Move cudf._lib.sort to cudf.core._internals

  #17488 commented on Dec 10, 2024 • 0 new comments

• Fix some possible thread-id overflow calculations

  #17473 commented on Dec 11, 2024 • 0 new comments

• Remove cudf._lib.string_casting in favor of inlining pylibcudf

  #17460 commented on Dec 11, 2024 • 0 new comments

• Move cudf._lib.stream_compaction to cudf.core._internals

  #17456 commented on Dec 10, 2024 • 0 new comments

• [DO NOT MERGE] ORC debug

  #17357 commented on Dec 11, 2024 • 0 new comments

• Mukernels strings

  #17286 commented on Dec 5, 2024 • 0 new comments

• Implement `HOST_UDF` aggregation for reduction and groupby

  #17249 commented on Dec 5, 2024 • 0 new comments

• [DO NOT MERGE/REVIEW] GDS debugging

  #17116 commented on Dec 11, 2024 • 0 new comments

• Add Python interface for to_arrow_device

  #15370 commented on Dec 11, 2024 • 0 new comments

• [FEA] Support left-semi and left-anti joins in `cudf::hash_join`

  #13700 commented on Dec 11, 2024 • 0 new comments

• [FEA] Add spark-rapids-jni builds to cudf CI

  #17337 commented on Dec 11, 2024 • 0 new comments

• [BUG] Decimal division loses precision when dividing by decimals with fractional parts

  #17448 commented on Dec 10, 2024 • 0 new comments

• [BUG] `DataFrame.columns` returns a `pd.Index` not a `cudf.Index`

  #17481 commented on Dec 10, 2024 • 0 new comments

• [FEA] Add public interop functions between pylibcudf and cudf classic

  #17191 commented on Dec 10, 2024 • 0 new comments

• [FEA] Support more flexible construction of nested columns in pylibcudf

  #17192 commented on Dec 10, 2024 • 0 new comments

• [FEA] Consider disabling `--expt-relaxed-constexpr`

  #7795 commented on Dec 6, 2024 • 0 new comments

• [QST] is there any way to compile from source in offline?

  #17416 commented on Dec 6, 2024 • 0 new comments

• [FEA] Create `pylibcudf.Table` from a `table_view` and an arbitrary owning object

  #17040 commented on Dec 5, 2024 • 0 new comments

• [BUG] cudf pivot extremely slow compared to pandas

  #17515 commented on Dec 5, 2024 • 0 new comments

• [BUG] Cannot use `cudf::test::iterators::nulls_at ()` in `.cu` test file

  #17432 commented on Dec 5, 2024 • 0 new comments

• [FEA] Run DuckDB benchmarks

  #17474 commented on Dec 5, 2024 • 0 new comments

• Prevent grid stride loop overflow in libcudf kernels

  #10368 commented on Dec 5, 2024 • 0 new comments

• [FEA] Implement cudf.pandas Profiler using sys.monitoring

  #17519 commented on Dec 4, 2024 • 0 new comments