Expose stream parameter in public strings combine APIs

cpp/include/cudf/strings/combine.hpp

@@ -66,18 +66,20 @@ enum class output_if_empty_list {
  *
  * @throw cudf::logic_error if separator is not valid.
  *
- * @param strings Strings for this operation.
+ * @param input Strings for this operation
  * @param separator String that should inserted between each string.
  *        Default is an empty string.
- * @param narep String that should represent any null strings found.
+ * @param narep String to replace any null strings found.
  *        Default of invalid-scalar will ignore any null entries.
+ * @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 column containing one string.
  */
 std::unique_ptr<column> join_strings(
-  strings_column_view const& strings,
+  strings_column_view const& input,
   string_scalar const& separator      = string_scalar(""),
   string_scalar const& narep          = string_scalar("", false),
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -127,25 +129,25 @@ std::unique_ptr<column> join_strings(
  * @throw cudf::logic_error if the number of rows from @p separators and @p strings_columns
  *                          do not match
  *
- * @param strings_columns List of strings columns to concatenate.
+ * @param strings_columns List of strings columns to concatenate
  * @param separators Strings column that provides the separator for a given row
- * @param separator_narep String that should be used in place of a null separator for a given
- *        row. Default of invalid-scalar means no row separator value replacements.
- *        Default is an invalid string.
- * @param col_narep String that should be used in place of any null strings
- *        found in any column. Default of invalid-scalar means no null column value replacements.
- *        Default is an invalid string.
+ * @param separator_narep String to replace a null separator for a given row.
+ *        Default of invalid-scalar means no row separator value replacements.
+ * @param col_narep String that should be used in place of any null strings found in any column.
+ *        Default of invalid-scalar means no null column value replacements.
  * @param separate_nulls If YES, then the separator is included for null rows
  *        if `col_narep` is valid.
- * @param mr Resource for allocating device memory.
- * @return New column with concatenated results.
+ * @param stream CUDA stream used for device memory operations and kernel launches
+ * @param mr Resource for allocating device memory
+ * @return New column with concatenated results
  */
 std::unique_ptr<column> concatenate(
   table_view const& strings_columns,
   strings_column_view const& separators,
   string_scalar const& separator_narep = string_scalar("", false),
   string_scalar const& col_narep       = string_scalar("", false),
   separator_on_nulls separate_nulls    = separator_on_nulls::YES,
+  rmm::cuda_stream_view stream         = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr  = rmm::mr::get_current_device_resource());
 
 /**
@@ -184,21 +186,23 @@ std::unique_ptr<column> concatenate(
  * @throw cudf::logic_error if separator is not valid.
  * @throw cudf::logic_error if only one column is specified
  *
- * @param strings_columns List of string columns to concatenate.
+ * @param strings_columns List of string columns to concatenate
  * @param separator String that should inserted between each string from each row.
  *        Default is an empty string.
- * @param narep String that should be used in place of any null strings
- *        found in any column. Default of invalid-scalar means any null entry in any column will
+ * @param narep String to replace any null strings found in any column.
+ *        Default of invalid-scalar means any null entry in any column will
  *        produces a null result for that row.
- * @param separate_nulls If YES, then the separator is included for null rows if `narep` is valid.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New column with concatenated results.
+ * @param separate_nulls If YES, then the separator is included for null rows if `narep` is valid
+ * @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 column with concatenated results
  */
 std::unique_ptr<column> concatenate(
   table_view const& strings_columns,
   string_scalar const& separator      = string_scalar(""),
   string_scalar const& narep          = string_scalar("", false),
   separator_on_nulls separate_nulls   = separator_on_nulls::YES,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -243,19 +247,20 @@ std::unique_ptr<column> concatenate(
  * @throw cudf::logic_error if the number of rows from `separators` and `lists_strings_column` do
  *        not match
  *
- * @param lists_strings_column Column containing lists of strings to concatenate.
- * @param separators Strings column that provides separators for concatenation.
- * @param separator_narep String that should be used to replace null separator, default is an
- *        invalid-scalar denoting that rows containing null separator will result in null string in
- *        the corresponding output rows.
- * @param string_narep String that should be used to replace null strings in any non-null list row,
- *        default is an invalid-scalar denoting that list rows containing null strings will result
- *        in null string in the corresponding output rows.
- * @param separate_nulls If YES, then the separator is included for null rows if `narep` is valid.
- * @param empty_list_policy if set to EMPTY_STRING, any input row that is an empty list will
+ * @param lists_strings_column Column containing lists of strings to concatenate
+ * @param separators Strings column that provides separators for concatenation
+ * @param separator_narep String that should be used to replace a null separator.
+ *        Default is an invalid-scalar denoting that rows containing null separator will result in
+ *        a null string in the corresponding output rows.
+ * @param string_narep String to replace null strings in any non-null list row.
+ *        Default is an invalid-scalar denoting that list rows containing null strings will result
+ *        in a null string in the corresponding output rows.
+ * @param separate_nulls If YES, then the separator is included for null rows if `narep` is valid
+ * @param empty_list_policy If set to EMPTY_STRING, any input row that is an empty list will
  *        result in an empty string. Otherwise, it will result in a null.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New strings column with concatenated results.
+ * @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 strings column with concatenated results
  */
 std::unique_ptr<column> join_list_elements(
   lists_column_view const& lists_strings_column,
@@ -264,6 +269,7 @@ std::unique_ptr<column> join_list_elements(
   string_scalar const& string_narep      = string_scalar("", false),
   separator_on_nulls separate_nulls      = separator_on_nulls::YES,
   output_if_empty_list empty_list_policy = output_if_empty_list::EMPTY_STRING,
+  rmm::cuda_stream_view stream           = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr    = rmm::mr::get_current_device_resource());
 
 /**
@@ -303,24 +309,26 @@ std::unique_ptr<column> join_list_elements(
  * @throw cudf::logic_error if input column is not lists of strings column.
  * @throw cudf::logic_error if separator is not valid.
  *
- * @param lists_strings_column Column containing lists of strings to concatenate.
- * @param separator String that should inserted between strings of each list row, default is an
- *        empty string.
- * @param narep String that should be used to replace null strings in any non-null list row, default
- *        is an invalid-scalar denoting that list rows containing null strings will result in null
- *        string in the corresponding output rows.
- * @param separate_nulls If YES, then the separator is included for null rows if `narep` is valid.
- * @param empty_list_policy if set to EMPTY_STRING, any input row that is an empty list will result
+ * @param lists_strings_column Column containing lists of strings to concatenate
+ * @param separator String to insert between strings of each list row.
+ *        Default is an empty string.
+ * @param narep String to replace null strings in any non-null list row.
+ *        Default is an invalid-scalar denoting that list rows containing null strings will result
+ *        in a null string in the corresponding output rows.
+ * @param separate_nulls If YES, then the separator is included for null rows if `narep` is valid
+ * @param empty_list_policy If set to EMPTY_STRING, any input row that is an empty list will result
  *        in an empty string. Otherwise, it will result in a null.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return New strings column with concatenated results.
+ * @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 strings column with concatenated results
  */
 std::unique_ptr<column> join_list_elements(
   lists_column_view const& lists_strings_column,
   string_scalar const& separator         = string_scalar(""),
   string_scalar const& narep             = string_scalar("", false),
   separator_on_nulls separate_nulls      = separator_on_nulls::YES,
   output_if_empty_list empty_list_policy = output_if_empty_list::EMPTY_STRING,
+  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/repeat_strings.hpp

@@ -52,12 +52,14 @@ namespace strings {
  *
  * @param input The scalar containing the string to repeat
  * @param repeat_times The number of times the input string is repeated
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr Device memory resource used to allocate the returned string scalar
  * @return New string scalar in which the input string is repeated
  */
 std::unique_ptr<string_scalar> repeat_string(
   string_scalar const& input,
   size_type repeat_times,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -81,12 +83,14 @@ std::unique_ptr<string_scalar> repeat_string(
  *
  * @param input The column containing strings to repeat
  * @param repeat_times The number of times each input string is repeated
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr Device memory resource used to allocate the returned strings column
  * @return New column containing the repeated strings
  */
 std::unique_ptr<column> repeat_strings(
   strings_column_view const& input,
   size_type repeat_times,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -115,13 +119,15 @@ std::unique_ptr<column> repeat_strings(
  *
  * @param input The column containing strings to repeat
  * @param repeat_times The column containing numbers of times that the corresponding input strings
- *                     are repeated
+ *                     for each row are repeated
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr Device memory resource used to allocate the returned strings column
  * @return New column containing the repeated strings.
  */
 std::unique_ptr<column> repeat_strings(
   strings_column_view const& input,
   column_view const& repeat_times,
+  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/combine/concatenate.cu

@@ -267,28 +267,24 @@ std::unique_ptr<column> concatenate(table_view const& strings_columns,
                                     string_scalar const& separator,
                                     string_scalar const& narep,
                                     separator_on_nulls separate_nulls,
+                                    rmm::cuda_stream_view stream,
                                     rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::concatenate(
-    strings_columns, separator, narep, separate_nulls, cudf::get_default_stream(), mr);
+  return detail::concatenate(strings_columns, separator, narep, separate_nulls, stream, mr);
 }
 
 std::unique_ptr<column> concatenate(table_view const& strings_columns,
                                     strings_column_view const& separators,
                                     string_scalar const& separator_narep,
                                     string_scalar const& col_narep,
                                     separator_on_nulls separate_nulls,
+                                    rmm::cuda_stream_view stream,
                                     rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::concatenate(strings_columns,
-                             separators,
-                             separator_narep,
-                             col_narep,
-                             separate_nulls,
-                             cudf::get_default_stream(),
-                             mr);
+  return detail::concatenate(
+    strings_columns, separators, separator_narep, col_narep, separate_nulls, stream, mr);
 }
 
 }  // namespace strings

cpp/src/strings/combine/join.cu

@@ -180,10 +180,11 @@ std::unique_ptr<column> join_strings(strings_column_view const& input,
 std::unique_ptr<column> join_strings(strings_column_view const& strings,
                                      string_scalar const& separator,
                                      string_scalar const& narep,
+                                     rmm::cuda_stream_view stream,
                                      rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::join_strings(strings, separator, narep, cudf::get_default_stream(), mr);
+  return detail::join_strings(strings, separator, narep, stream, mr);
 }
 
 }  // namespace strings

cpp/src/strings/combine/join_list_elements.cu

@@ -301,16 +301,12 @@ std::unique_ptr<column> join_list_elements(lists_column_view const& lists_string
                                            string_scalar const& narep,
                                            separator_on_nulls separate_nulls,
                                            output_if_empty_list empty_list_policy,
+                                           rmm::cuda_stream_view stream,
                                            rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::join_list_elements(lists_strings_column,
-                                    separator,
-                                    narep,
-                                    separate_nulls,
-                                    empty_list_policy,
-                                    cudf::get_default_stream(),
-                                    mr);
+  return detail::join_list_elements(
+    lists_strings_column, separator, narep, separate_nulls, empty_list_policy, stream, mr);
 }
 
 std::unique_ptr<column> join_list_elements(lists_column_view const& lists_strings_column,
@@ -319,6 +315,7 @@ std::unique_ptr<column> join_list_elements(lists_column_view const& lists_string
                                            string_scalar const& string_narep,
                                            separator_on_nulls separate_nulls,
                                            output_if_empty_list empty_list_policy,
+                                           rmm::cuda_stream_view stream,
                                            rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
@@ -328,7 +325,7 @@ std::unique_ptr<column> join_list_elements(lists_column_view const& lists_string
                                     string_narep,
                                     separate_nulls,
                                     empty_list_policy,
-                                    cudf::get_default_stream(),
+                                    stream,
                                     mr);
 }
 

cpp/src/strings/repeat_strings.cu

@@ -67,7 +67,7 @@ std::unique_ptr<string_scalar> repeat_string(string_scalar const& input,
                       return in_ptr[idx % str_size];
                     });
 
-  return std::make_unique<string_scalar>(std::move(buff));
+  return std::make_unique<string_scalar>(std::move(buff), true, stream, mr);
 }
 
 namespace {
@@ -260,26 +260,29 @@ std::unique_ptr<column> repeat_strings(strings_column_view const& input,
 
 std::unique_ptr<string_scalar> repeat_string(string_scalar const& input,
                                              size_type repeat_times,
+                                             rmm::cuda_stream_view stream,
                                              rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::repeat_string(input, repeat_times, cudf::get_default_stream(), mr);
+  return detail::repeat_string(input, repeat_times, stream, mr);
 }
 
 std::unique_ptr<column> repeat_strings(strings_column_view const& input,
                                        size_type repeat_times,
+                                       rmm::cuda_stream_view stream,
                                        rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::repeat_strings(input, repeat_times, cudf::get_default_stream(), mr);
+  return detail::repeat_strings(input, repeat_times, stream, mr);
 }
 
 std::unique_ptr<column> repeat_strings(strings_column_view const& input,
                                        column_view const& repeat_times,
+                                       rmm::cuda_stream_view stream,
                                        rmm::mr::device_memory_resource* mr)
 {
   CUDF_FUNC_RANGE();
-  return detail::repeat_strings(input, repeat_times, cudf::get_default_stream(), mr);
+  return detail::repeat_strings(input, repeat_times, stream, mr);
 }
 
 }  // namespace strings

cpp/tests/CMakeLists.txt

@@ -635,6 +635,7 @@ ConfigureTest(STREAM_DICTIONARY_TEST streams/dictionary_test.cpp STREAM_MODE tes
 ConfigureTest(
   STREAM_STRINGS_TEST
   streams/strings/case_test.cpp
+  streams/strings/combine_test.cpp
   streams/strings/convert_test.cpp
   streams/strings/find_test.cpp
   streams/strings/replace_test.cpp