Remove unneeded output size parameter from internal count_matches uti…

…lity

cpp/src/strings/contains.cu

@@ -112,7 +112,7 @@ std::unique_ptr<column> count_re(strings_column_view const& input,
 
   auto const d_strings = column_device_view::create(input.parent(), stream);
 
-  auto result = count_matches(*d_strings, *d_prog, input.size(), stream, mr);
+  auto result = count_matches(*d_strings, *d_prog, stream, mr);
   if (input.has_nulls()) {
     result->set_null_mask(cudf::detail::copy_bitmask(input.parent(), stream, mr),
                           input.null_count());

cpp/src/strings/count_matches.cu

@@ -60,18 +60,15 @@ struct count_fn {
 
 std::unique_ptr<column> count_matches(column_device_view const& d_strings,
                                       reprog_device& d_prog,
-                                      size_type output_size,
                                       rmm::cuda_stream_view stream,
                                       rmm::device_async_resource_ref mr)
 {
-  assert(output_size >= d_strings.size() and "Unexpected output size");
-
   auto results = make_numeric_column(
-    data_type{type_to_id<size_type>()}, output_size, mask_state::UNALLOCATED, stream, mr);
+    data_type{type_to_id<size_type>()}, d_strings.size(), mask_state::UNALLOCATED, stream, mr);
 
-  if (d_strings.size() == 0) return results;
+  if (d_strings.size() == 0) { return results; }
 
-  auto d_results = results->mutable_view().data<int32_t>();
+  auto d_results = results->mutable_view().data<cudf::size_type>();
 
   launch_transform_kernel(count_fn{d_strings}, d_prog, d_results, d_strings.size(), stream);
 

cpp/src/strings/count_matches.hpp

@@ -37,14 +37,12 @@ class reprog_device;
  *
  * @param d_strings Device view of the input strings column.
  * @param d_prog Regex instance to evaluate on each string.
- * @param output_size Number of rows for the output column.
  * @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 Integer column of match counts
  */
 std::unique_ptr<column> count_matches(column_device_view const& d_strings,
                                       reprog_device& d_prog,
-                                      size_type output_size,
                                       rmm::cuda_stream_view stream,
                                       rmm::device_async_resource_ref mr);
 

cpp/src/strings/extract/extract_all.cu

@@ -119,7 +119,7 @@ std::unique_ptr<column> extract_all_record(strings_column_view const& input,
 
   // Get the match counts for each string.
   // This column will become the output lists child offsets column.
-  auto counts   = count_matches(*d_strings, *d_prog, strings_count, stream, mr);
+  auto counts   = count_matches(*d_strings, *d_prog, stream, mr);
   auto d_counts = counts->mutable_view().data<size_type>();
 
   // Compute null output rows

cpp/src/strings/search/findall.cu

@@ -104,7 +104,7 @@ std::unique_ptr<column> findall(strings_column_view const& input,
   auto d_prog = regex_device_builder::create_prog_device(prog, stream);
 
   // Create lists offsets column
-  auto const sizes              = count_matches(*d_strings, *d_prog, strings_count, stream, mr);
+  auto const sizes              = count_matches(*d_strings, *d_prog, stream, mr);
   auto [offsets, total_matches] = cudf::detail::make_offsets_child_column(
     sizes->view().begin<size_type>(), sizes->view().end<size_type>(), stream, mr);
   auto const d_offsets = offsets->view().data<size_type>();

cpp/src/strings/split/split_re.cu

@@ -206,8 +206,8 @@ std::unique_ptr<table> split_re(strings_column_view const& input,
   auto d_strings = column_device_view::create(input.parent(), stream);
 
   // count the number of delimiters matched in each string
-  auto const counts = count_matches(
-    *d_strings, *d_prog, strings_count, stream, rmm::mr::get_current_device_resource());
+  auto const counts =
+    count_matches(*d_strings, *d_prog, stream, rmm::mr::get_current_device_resource());
 
   // get the split tokens from the input column; this also converts the counts into offsets
   auto [tokens, offsets] =
@@ -271,7 +271,7 @@ std::unique_ptr<column> split_record_re(strings_column_view const& input,
   auto d_strings = column_device_view::create(input.parent(), stream);
 
   // count the number of delimiters matched in each string
-  auto counts = count_matches(*d_strings, *d_prog, strings_count, stream, mr);
+  auto counts = count_matches(*d_strings, *d_prog, stream, mr);
 
   // get the split tokens from the input column; this also converts the counts into offsets
   auto [tokens, offsets] =