Expose stream parameter in public strings find APIs

cpp/include/cudf/strings/find.hpp

@@ -43,19 +43,21 @@ namespace strings {
  *
  * @throw cudf::logic_error if start position is greater than stop position.
  *
- * @param strings Strings instance for this operation.
- * @param target UTF-8 encoded string to search for in each string.
- * @param start First character position to include in the search.
+ * @param input Strings instance for this operation
+ * @param target UTF-8 encoded string to search for in each string
+ * @param start First character position to include in the search
  * @param stop Last position (exclusive) to include in the search.
  *             Default of -1 will search to the end of the string.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New integer column with character position values.
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return New integer column with character position values
  */
 std::unique_ptr<column> find(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   string_scalar const& target,
   size_type start                     = 0,
   size_type stop                      = -1,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -72,19 +74,21 @@ std::unique_ptr<column> find(
  *
  * @throw cudf::logic_error if start position is greater than stop position.
  *
- * @param strings Strings instance for this operation.
- * @param target UTF-8 encoded string to search for in each string.
- * @param start First position to include in the search.
+ * @param input Strings instance for this operation
+ * @param target UTF-8 encoded string to search for in each string
+ * @param start First position to include in the search
  * @param stop Last position (exclusive) to include in the search.
  *             Default of -1 will search starting at the end of the string.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New integer column with character position values.
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return New integer column with character position values
  */
 std::unique_ptr<column> rfind(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   string_scalar const& target,
   size_type start                     = 0,
   size_type stop                      = -1,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -123,37 +127,41 @@ std::unique_ptr<column> find(
  *
  * Any null string entries return corresponding null entries in the output columns.
  *
- * @param strings Strings instance for this operation.
- * @param target UTF-8 encoded string to search for in each string.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New type_id::BOOL8 column.
+ * @param input Strings instance for this operation
+ * @param target UTF-8 encoded string to search for in each string
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return New BOOL8 column
  */
 std::unique_ptr<column> contains(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   string_scalar const& target,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
  * @brief Returns a column of boolean values for each string where true indicates
  * the corresponding target string was found within that string in the provided column.
  *
- * The 'output[i] = true` if string `targets[i]` is found inside `strings[i]` otherwise
+ * The 'output[i] = true` if string `targets[i]` is found inside `input[i]` otherwise
  * `output[i] = false`.
  * If `target[i]` is an empty string, true is returned for `output[i]`.
  * If `target[i]` is null, false is returned for `output[i]`.
  *
- * Any null `strings[i]` row results in a null `output[i]` row.
+ * Any null string entries return corresponding null entries in the output columns.
  *
  * @throw cudf::logic_error if `strings.size() != targets.size()`.
  *
- * @param strings Strings instance for this operation.
- * @param targets Strings column of targets to check row-wise in `strings`.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New type_id::BOOL8 column.
+ * @param input Strings instance for this operation
+ * @param targets Strings column of targets to check row-wise in `strings`
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return New BOOL8 column
  */
 std::unique_ptr<column> contains(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   strings_column_view const& targets,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -166,14 +174,16 @@ std::unique_ptr<column> contains(
  *
  * Any null string entries return corresponding null entries in the output columns.
  *
- * @param strings Strings instance for this operation.
- * @param target UTF-8 encoded string to search for in each string.
+ * @param input Strings instance for this operation
+ * @param target UTF-8 encoded string to search for in each string
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr Device memory resource used to allocate the returned column's device memory.
  * @return New type_id::BOOL8 column.
  */
 std::unique_ptr<column> starts_with(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   string_scalar const& target,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -190,14 +200,16 @@ std::unique_ptr<column> starts_with(
  *
  * @throw cudf::logic_error if `strings.size() != targets.size()`.
  *
- * @param strings Strings instance for this operation.
- * @param targets Strings instance for this operation.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New type_id::BOOL8 column.
+ * @param input Strings instance for this operation
+ * @param targets Strings instance for this operation
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return New BOOL8 column
  */
 std::unique_ptr<column> starts_with(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   strings_column_view const& targets,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -210,14 +222,16 @@ std::unique_ptr<column> starts_with(
  *
  * Any null string entries return corresponding null entries in the output columns.
  *
- * @param strings Strings instance for this operation.
- * @param target UTF-8 encoded string to search for in each string.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New type_id::BOOL8 column.
+ * @param input Strings instance for this operation
+ * @param target UTF-8 encoded string to search for in each string
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return New BOOL8 column
  */
 std::unique_ptr<column> ends_with(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   string_scalar const& target,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -234,14 +248,16 @@ std::unique_ptr<column> ends_with(
  *
  * @throw cudf::logic_error if `strings.size() != targets.size()`.
  *
- * @param strings Strings instance for this operation.
- * @param targets Strings instance for this operation.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New type_id::BOOL8 column.
+ * @param input Strings instance for this operation
+ * @param targets Strings instance for this operation
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return New BOOL8 column
  */
 std::unique_ptr<column> ends_with(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   strings_column_view const& targets,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 /** @} */  // end of doxygen group
 }  // namespace strings

cpp/include/cudf/strings/find_multiple.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -48,14 +48,16 @@ namespace strings {
  *
  * @throw cudf::logic_error if `targets` is empty or contains nulls
  *
- * @param input Strings instance for this operation.
- * @param targets Strings to search for in each string.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return Lists column with character position values.
+ * @param input Strings instance for this operation
+ * @param targets Strings to search for in each string
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Device memory resource used to allocate the returned column's device memory
+ * @return Lists column with character position values
  */
 std::unique_ptr<column> find_multiple(
   strings_column_view const& input,
   strings_column_view const& targets,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /** @} */  // end of doxygen group

cpp/include/cudf/strings/findall.hpp

@@ -57,12 +57,14 @@ struct regex_program;
  *
  * @param input Strings instance for this operation
  * @param prog Regex program instance
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr Device memory resource used to allocate the returned column's device memory
  * @return New lists column of strings
  */
 std::unique_ptr<column> findall(
   strings_column_view const& input,
   regex_program const& prog,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /** @} */  // end of doxygen group

cpp/src/strings/search/find.cu

@@ -305,20 +305,22 @@ std::unique_ptr<column> find(strings_column_view const& strings,
                              string_scalar const& target,
                              size_type start,
                              size_type stop,
+                             rmm::cuda_stream_view stream,
                              rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::find(strings, target, start, stop, cudf::get_default_stream(), mr);
+  return detail::find(strings, target, start, stop, stream, mr);
 }
 
 std::unique_ptr<column> rfind(strings_column_view const& strings,
                               string_scalar const& target,
                               size_type start,
                               size_type stop,
+                              rmm::cuda_stream_view stream,
                               rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::rfind(strings, target, start, stop, cudf::get_default_stream(), mr);
+  return detail::rfind(strings, target, start, stop, stream, mr);
 }
 
 std::unique_ptr<column> find(strings_column_view const& input,
@@ -618,50 +620,56 @@ std::unique_ptr<column> ends_with(strings_column_view const& strings,
 
 std::unique_ptr<column> contains(strings_column_view const& strings,
                                  string_scalar const& target,
+                                 rmm::cuda_stream_view stream,
                                  rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::contains(strings, target, cudf::get_default_stream(), mr);
+  return detail::contains(strings, target, stream, mr);
 }
 
 std::unique_ptr<column> contains(strings_column_view const& strings,
                                  strings_column_view const& targets,
+                                 rmm::cuda_stream_view stream,
                                  rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::contains(strings, targets, cudf::get_default_stream(), mr);
+  return detail::contains(strings, targets, stream, mr);
 }
 
 std::unique_ptr<column> starts_with(strings_column_view const& strings,
                                     string_scalar const& target,
+                                    rmm::cuda_stream_view stream,
                                     rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::starts_with(strings, target, cudf::get_default_stream(), mr);
+  return detail::starts_with(strings, target, stream, mr);
 }
 
 std::unique_ptr<column> starts_with(strings_column_view const& strings,
                                     strings_column_view const& targets,
+                                    rmm::cuda_stream_view stream,
                                     rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::starts_with(strings, targets, cudf::get_default_stream(), mr);
+  return detail::starts_with(strings, targets, stream, mr);
 }
 
 std::unique_ptr<column> ends_with(strings_column_view const& strings,
                                   string_scalar const& target,
+                                  rmm::cuda_stream_view stream,
                                   rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::ends_with(strings, target, cudf::get_default_stream(), mr);
+  return detail::ends_with(strings, target, stream, mr);
 }
 
 std::unique_ptr<column> ends_with(strings_column_view const& strings,
                                   strings_column_view const& targets,
+                                  rmm::cuda_stream_view stream,
                                   rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::ends_with(strings, targets, cudf::get_default_stream(), mr);
+  return detail::ends_with(strings, targets, stream, mr);
 }
 
 }  // namespace strings

cpp/src/strings/search/find_multiple.cu

@@ -70,8 +70,8 @@ std::unique_ptr<column> find_multiple(strings_column_view const& input,
   results->set_null_count(0);
 
   auto offsets = cudf::detail::sequence(strings_count + 1,
-                                        numeric_scalar<size_type>(0),
-                                        numeric_scalar<size_type>(targets_count),
+                                        numeric_scalar<size_type>(0, true, stream),
+                                        numeric_scalar<size_type>(targets_count, true, stream),
                                         stream,
                                         mr);
   return make_lists_column(strings_count,
@@ -88,10 +88,11 @@ std::unique_ptr<column> find_multiple(strings_column_view const& input,
 // external API
 std::unique_ptr<column> find_multiple(strings_column_view const& input,
                                       strings_column_view const& targets,
+                                      rmm::cuda_stream_view stream,
                                       rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::find_multiple(input, targets, cudf::get_default_stream(), mr);
+  return detail::find_multiple(input, targets, stream, mr);
 }
 
 }  // namespace strings

cpp/src/strings/search/findall.cu

@@ -134,10 +134,11 @@ std::unique_ptr<column> findall(strings_column_view const& input,
 
 std::unique_ptr<column> findall(strings_column_view const& input,
                                 regex_program const& prog,
+                                rmm::cuda_stream_view stream,
                                 rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::findall(input, prog, cudf::get_default_stream(), mr);
+  return detail::findall(input, prog, stream, mr);
 }
 
 }  // namespace strings

cpp/tests/CMakeLists.txt

@@ -627,6 +627,7 @@ ConfigureTest(STREAM_CONCATENATE_TEST streams/concatenate_test.cpp STREAM_MODE t
 ConfigureTest(STREAM_FILLING_TEST streams/filling_test.cpp STREAM_MODE testing)
 ConfigureTest(STREAM_REPLACE_TEST streams/replace_test.cpp STREAM_MODE testing)
 ConfigureTest(STREAM_SEARCH_TEST streams/search_test.cpp STREAM_MODE testing)
+ConfigureTest(STREAM_STRINGS_TEST streams/strings/find_test.cpp STREAM_MODE testing)
 
 # ##################################################################################################
 # Install tests ####################################################################################