copying.hpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2018-2024, NVIDIA CORPORATION.
3  *
4  * Licensed under the Apache License, Version 2.0 (the "License");
5  * you may not use this file except in compliance with the License.
6  * You may obtain a copy of the License at
7  *
8  * http://www.apache.org/licenses/LICENSE-2.0
9  *
10  * Unless required by applicable law or agreed to in writing, software
11  * distributed under the License is distributed on an "AS IS" BASIS,
12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13  * See the License for the specific language governing permissions and
14  * limitations under the License.
15  */
16 
17 #pragma once
18 
19 #include <cudf/column/column_view.hpp>
20 #include <cudf/lists/lists_column_view.hpp>
21 #include <cudf/scalar/scalar.hpp>
22 #include <cudf/strings/strings_column_view.hpp>
23 #include <cudf/structs/structs_column_view.hpp>
24 #include <cudf/table/table.hpp>
25 #include <cudf/types.hpp>
26 #include <cudf/utilities/export.hpp>
27 #include <cudf/utilities/memory_resource.hpp>
28 
29 #include <memory>
30 #include <vector>
31 
32 namespace CUDF_EXPORT cudf {
33 
48 enum class out_of_bounds_policy : bool {
49  NULLIFY,
50  DONT_CHECK
51 };
52 
84 std::unique_ptr<table> gather(
85  table_view const& source_table,
86  column_view const& gather_map,
87  out_of_bounds_policy bounds_policy = out_of_bounds_policy::DONT_CHECK,
88  rmm::cuda_stream_view stream = cudf::get_default_stream(),
89  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
90 
106 std::unique_ptr<table> reverse(
107  table_view const& source_table,
108  rmm::cuda_stream_view stream = cudf::get_default_stream(),
109  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
110 
126 std::unique_ptr<column> reverse(
127  column_view const& source_column,
128  rmm::cuda_stream_view stream = cudf::get_default_stream(),
129  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
130 
173 std::unique_ptr<table> scatter(
174  table_view const& source,
175  column_view const& scatter_map,
176  table_view const& target,
177  rmm::cuda_stream_view stream = cudf::get_default_stream(),
178  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
179 
216 std::unique_ptr<table> scatter(
217  std::vector<std::reference_wrapper<scalar const>> const& source,
218  column_view const& indices,
219  table_view const& target,
220  rmm::cuda_stream_view stream = cudf::get_default_stream(),
221  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
222 
226 enum class mask_allocation_policy : int32_t {
227  NEVER,
228  RETAIN,
229  ALWAYS
230 };
231 
238 std::unique_ptr<column> empty_like(column_view const& input);
239 
246 std::unique_ptr<column> empty_like(scalar const& input);
247 
265 std::unique_ptr<column> allocate_like(
266  column_view const& input,
267  mask_allocation_policy mask_alloc = mask_allocation_policy::RETAIN,
268  rmm::cuda_stream_view stream = cudf::get_default_stream(),
269  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
270 
287 std::unique_ptr<column> allocate_like(
288  column_view const& input,
289  size_type size,
290  mask_allocation_policy mask_alloc = mask_allocation_policy::RETAIN,
291  rmm::cuda_stream_view stream = cudf::get_default_stream(),
292  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
293 
304 std::unique_ptr<table> empty_like(table_view const& input_table);
305 
338 void copy_range_in_place(column_view const& source,
339  mutable_column_view& target,
340  size_type source_begin,
341  size_type source_end,
342  size_type target_begin,
343  rmm::cuda_stream_view stream = cudf::get_default_stream());
344 
377 std::unique_ptr<column> copy_range(
378  column_view const& source,
379  column_view const& target,
380  size_type source_begin,
381  size_type source_end,
382  size_type target_begin,
383  rmm::cuda_stream_view stream = cudf::get_default_stream(),
384  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
385 
423 std::unique_ptr<column> shift(
424  column_view const& input,
425  size_type offset,
426  scalar const& fill_value,
427  rmm::cuda_stream_view stream = cudf::get_default_stream(),
428  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
429 
461 std::vector<column_view> slice(column_view const& input,
462  host_span<size_type const> indices,
463  rmm::cuda_stream_view stream = cudf::get_default_stream());
468 std::vector<column_view> slice(column_view const& input,
469  std::initializer_list<size_type> indices,
470  rmm::cuda_stream_view stream = cudf::get_default_stream());
471 
505 std::vector<table_view> slice(table_view const& input,
506  host_span<size_type const> indices,
507  rmm::cuda_stream_view stream = cudf::get_default_stream());
512 std::vector<table_view> slice(table_view const& input,
513  std::initializer_list<size_type> indices,
514  rmm::cuda_stream_view stream = cudf::get_default_stream());
515 
549 std::vector<column_view> split(column_view const& input,
550  host_span<size_type const> splits,
551  rmm::cuda_stream_view stream = cudf::get_default_stream());
556 std::vector<column_view> split(column_view const& input,
557  std::initializer_list<size_type> splits,
558  rmm::cuda_stream_view stream = cudf::get_default_stream());
559 
595 std::vector<table_view> split(table_view const& input,
596  host_span<size_type const> splits,
597  rmm::cuda_stream_view stream = cudf::get_default_stream());
602 std::vector<table_view> split(table_view const& input,
603  std::initializer_list<size_type> splits,
604  rmm::cuda_stream_view stream = cudf::get_default_stream());
605 
626 std::unique_ptr<column> copy_if_else(
627  column_view const& lhs,
628  column_view const& rhs,
629  column_view const& boolean_mask,
630  rmm::cuda_stream_view stream = cudf::get_default_stream(),
631  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
632 
652 std::unique_ptr<column> copy_if_else(
653  scalar const& lhs,
654  column_view const& rhs,
655  column_view const& boolean_mask,
656  rmm::cuda_stream_view stream = cudf::get_default_stream(),
657  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
658 
678 std::unique_ptr<column> copy_if_else(
679  column_view const& lhs,
680  scalar const& rhs,
681  column_view const& boolean_mask,
682  rmm::cuda_stream_view stream = cudf::get_default_stream(),
683  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
684 
702 std::unique_ptr<column> copy_if_else(
703  scalar const& lhs,
704  scalar const& rhs,
705  column_view const& boolean_mask,
706  rmm::cuda_stream_view stream = cudf::get_default_stream(),
707  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
708 
746 std::unique_ptr<table> boolean_mask_scatter(
747  table_view const& input,
748  table_view const& target,
749  column_view const& boolean_mask,
750  rmm::cuda_stream_view stream = cudf::get_default_stream(),
751  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
752 
785 std::unique_ptr<table> boolean_mask_scatter(
786  std::vector<std::reference_wrapper<scalar const>> const& input,
787  table_view const& target,
788  column_view const& boolean_mask,
789  rmm::cuda_stream_view stream = cudf::get_default_stream(),
790  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
791 
806 std::unique_ptr<scalar> get_element(
807  column_view const& input,
808  size_type index,
809  rmm::cuda_stream_view stream = cudf::get_default_stream(),
810  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
811 
815 enum class sample_with_replacement : bool {
816  FALSE,
817  TRUE
818 };
819 
848 std::unique_ptr<table> sample(
849  table_view const& input,
850  size_type const n,
851  sample_with_replacement replacement = sample_with_replacement::FALSE,
852  int64_t const seed = 0,
853  rmm::cuda_stream_view stream = cudf::get_default_stream(),
854  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
855 
873 bool has_nonempty_nulls(column_view const& input,
874  rmm::cuda_stream_view stream = cudf::get_default_stream());
875 
898 bool may_have_nonempty_nulls(column_view const& input);
899 
968 std::unique_ptr<column> purge_nonempty_nulls(
969  column_view const& input,
970  rmm::cuda_stream_view stream = cudf::get_default_stream(),
971  rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
972 
974 } // namespace CUDF_EXPORT cudf
cudf::column_view
A non-owning, immutable view of device data as a column of elements, some of which may be null as ind...
Definition: column_view.hpp:321
cudf::mutable_column_view
A non-owning, mutable view of device data as a column of elements, some of which may be null as indic...
Definition: column_view.hpp:497
cudf::scalar
An owning class to represent a singular value.
Definition: scalar.hpp:48
cudf::table_view
A set of cudf::column_view's of the same size.
Definition: table_view.hpp:200
rmm::cuda_stream_view
column_view.hpp
column view class definitions
cudf::mask_allocation_policy
mask_allocation_policy
Indicates when to allocate a mask, based on an existing mask.
Definition: copying.hpp:226
cudf::reverse
std::unique_ptr< column > reverse(column_view const &source_column, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Reverses the elements of a column.
cudf::copy_if_else
std::unique_ptr< column > copy_if_else(scalar const &lhs, scalar const &rhs, column_view const &boolean_mask, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Returns a new column, where each element is selected from either lhs or rhs based on the value of the...
cudf::allocate_like
std::unique_ptr< column > allocate_like(column_view const &input, size_type size, mask_allocation_policy mask_alloc=mask_allocation_policy::RETAIN, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Creates an uninitialized new column of the specified size and same type as the input.
cudf::may_have_nonempty_nulls
bool may_have_nonempty_nulls(column_view const &input)
Approximates if a column or its descendants may have non-empty null elements.
cudf::copy_range_in_place
void copy_range_in_place(column_view const &source, mutable_column_view &target, size_type source_begin, size_type source_end, size_type target_begin, rmm::cuda_stream_view stream=cudf::get_default_stream())
Copies a range of elements in-place from one column to another.
cudf::sample
std::unique_ptr< table > sample(table_view const &input, size_type const n, sample_with_replacement replacement=sample_with_replacement::FALSE, int64_t const seed=0, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Gather n samples from given input randomly.
cudf::has_nonempty_nulls
bool has_nonempty_nulls(column_view const &input, rmm::cuda_stream_view stream=cudf::get_default_stream())
Checks if a column or its descendants have non-empty null rows.
cudf::purge_nonempty_nulls
std::unique_ptr< column > purge_nonempty_nulls(column_view const &input, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Copy input into output while purging any non-empty null rows in the column or its descendants.
cudf::copy_range
std::unique_ptr< column > copy_range(column_view const &source, column_view const &target, size_type source_begin, size_type source_end, size_type target_begin, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Copies a range of elements out-of-place from one column to another.
cudf::out_of_bounds_policy
out_of_bounds_policy
Policy to account for possible out-of-bounds indices.
Definition: copying.hpp:48
cudf::sample_with_replacement
sample_with_replacement
Indicates whether a row can be sampled more than once.
Definition: copying.hpp:815
cudf::get_element
std::unique_ptr< scalar > get_element(column_view const &input, size_type index, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Get the element at specified index from a column.
cudf::empty_like
std::unique_ptr< table > empty_like(table_view const &input_table)
Creates a table of empty columns with the same types as the input_table
cudf::mask_allocation_policy::RETAIN
@ RETAIN
Allocate a null mask if the input contains one.
cudf::out_of_bounds_policy::DONT_CHECK
@ DONT_CHECK
No bounds checking is performed, better performance.
cudf::out_of_bounds_policy::NULLIFY
@ NULLIFY
Output values corresponding to out-of-bounds indices are null.
cudf::sample_with_replacement::FALSE
@ FALSE
A row can be sampled only once.
cudf::sample_with_replacement::TRUE
@ TRUE
A row can be sampled more than once.
cudf::gather
std::unique_ptr< table > gather(table_view const &source_table, column_view const &gather_map, out_of_bounds_policy bounds_policy=out_of_bounds_policy::DONT_CHECK, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Gathers the specified rows (including null values) of a set of columns.
cudf::scatter
std::unique_ptr< table > scatter(std::vector< std::reference_wrapper< scalar const >> const &source, column_view const &indices, table_view const &target, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Scatters a row of scalar values into a copy of the target table according to a scatter map.
cudf::boolean_mask_scatter
std::unique_ptr< table > boolean_mask_scatter(std::vector< std::reference_wrapper< scalar const >> const &input, table_view const &target, column_view const &boolean_mask, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Scatters scalar values to rows of the output corresponding to true values in a boolean mask.
cudf::shift
std::unique_ptr< column > shift(column_view const &input, size_type offset, scalar const &fill_value, rmm::cuda_stream_view stream=cudf::get_default_stream(), rmm::device_async_resource_ref mr=cudf::get_current_device_resource_ref())
Creates a new column by shifting all values by an offset.
cudf::slice
std::vector< table_view > slice(table_view const &input, std::initializer_list< size_type > indices, rmm::cuda_stream_view stream=cudf::get_default_stream())
Slices a table_view into a set of table_views according to a set of indices.
cudf::split
std::vector< table_view > split(table_view const &input, std::initializer_list< size_type > splits, rmm::cuda_stream_view stream=cudf::get_default_stream())
Splits a table_view into a set of table_views according to a set of indices derived from expected spl...
cudf::get_default_stream
rmm::cuda_stream_view const get_default_stream()
Get the current default stream.
cudf::io::ALWAYS
@ ALWAYS
Use dictionary regardless of impact on compression.
Definition: io/types.hpp:228
cudf::io::NEVER
@ NEVER
Never use dictionary encoding.
Definition: io/types.hpp:226
cudf::get_current_device_resource_ref
rmm::device_async_resource_ref get_current_device_resource_ref()
Get the current device memory resource reference.
Definition: memory_resource.hpp:49
device_async_resource_ref
cuda::mr::async_resource_ref< cuda::mr::device_accessible > device_async_resource_ref
cudf::size_type
int32_t size_type
Row index type for columns and tables.
Definition: types.hpp:95
lists_column_view.hpp
Class definition for cudf::lists_column_view.
memory_resource.hpp
cudf
cuDF interfaces
Definition: aggregation.hpp:35
scalar.hpp
Class definitions for cudf::scalar.
strings_column_view.hpp
Class definition for cudf::strings_column_view.
cudf::host_span
C++20 std::span with reduced feature set.
Definition: span.hpp:231
structs_column_view.hpp
Class definition for cudf::structs_column_view.
table.hpp
Class definition for cudf::table.
types.hpp
Type declarations for libcudf.
Generated by doxygen 1.9.1