David Wendt

Refactor pad_side and strip_type enums into side_type enum

1

## Description Refactors the `cudf::strings::pad_side` and `cudf::strings::strip_type` to a single enum `cudf::strings::side_type`. These have the same values as used by `cudf::strings::pad` and `cudf::strings::strip` Moving these into a single header helps...

Add regex ASCII flag support for matching builtin character classes

1

## Description Adds ASCII flag to the libcudf `regex_flags` for support with builtin character classes: `\w, \W, \s, \S, \d, \D`. Somewhat equivalent to https://docs.python.org/3/library/re.html#re.ASCII But strictly the flag modifies...

Excessive loop unrolling and forced inlining causes performance issues in sort with custom comparators

3

In the [libcudf](https://github.com/rapidsai/cudf) component of RAPIDS we have a sort API that calls `thrust::sort` and `thrust::stable_sort` using a custom comparator for columns of data. Reference libcudf calling sort/stable_sort: https://github.com/rapidsai/cudf/blob/branch-0.17/cpp/src/sort/sort_impl.cuh As...

inclusive_scan passing invalid data to AssociativeOperator

4

The `thrust::inclusive_scan` passes random data not included in the input vector/iterators to the AssociateOperator functor parameter. The following code illustrates the issue. ``` #include #include #include #include #include int main(int...

CUDA thrust::lower_bound fails when give a custom output iterator and compiled with -G option [NVBug 3322776]

4

I believe this is a compiler issue since the problem only appears when using the -G option on nvcc. Unfortunately I'm not able to follow the thrust source code here...

The `thrust::reduce_by_key` function passes random data not included in the input vector/iterators to the BinaryFunction parameter as defined here: https://thrust.github.io/doc/group__reductions_gaafd6f34b72f1ea30d58ba916d53ee754.html#gaafd6f34b72f1ea30d58ba916d53ee754 The result from the binary-operator when given the invalid data...

## Description Updates the instruction to build the libcudf documentation files in DOCUMENTATION.md. The `cmake --build . --target docs_cudf` will invoke the appropriate make tool as setup when cmake was...

Fix cudf::lists::sort_lists for NaN and Infinity values

3

## Description Fixes logic in `cudf::lists::sort_lists` handling of sorting floating-point values containing `NaN`, `-NaN`, `Infinity` and `-Infinity`. For large lists (elements >100) of any numeric type, the `cub::DeviceSegmentedRadixSort` is used...

Remove using using `namespace cudf;` and `using namespace cudf::test;` from gtests. Likewise remove unit test source coded inside `namespace cudf {}` or `namespace cudf::test {}` The unit tests should not...

There are some inconsistencies in the sort logic used by `sort_column.cu` and `segmented_sort.cu` both use a radix-sort optimization but apply different logic for handling nulls and fixed-width types. Overall the...