rtpsw commented on code in PR #34311:
URL: https://github.com/apache/arrow/pull/34311#discussion_r1117884396
##########
cpp/src/arrow/compute/kernels/hash_aggregate_test.cc:
##########
@@ -135,22 +141,84 @@ Result<Datum> NaiveGroupBy(std::vector<Datum> arguments,
std::vector<Datum> keys
return Take(struct_arr, sorted_indices);
}
+Result<Datum> MakeGroupByOutput(const std::vector<ExecBatch>& output_batches,
+ const std::shared_ptr<Schema> output_schema,
+ size_t num_aggregates, size_t num_keys, bool
naive) {
+ ArrayVector out_arrays(num_aggregates + num_keys);
+ for (size_t i = 0; i < out_arrays.size(); ++i) {
+ std::vector<std::shared_ptr<Array>> arrays(output_batches.size());
+ for (size_t j = 0; j < output_batches.size(); ++j) {
+ arrays[j] = output_batches[j].values[i].make_array();
+ }
+ if (arrays.empty()) {
+ ARROW_ASSIGN_OR_RAISE(
+ out_arrays[i],
+ MakeArrayOfNull(output_schema->field(static_cast<int>(i))->type(),
+ /*length=*/0));
+ } else {
+ ARROW_ASSIGN_OR_RAISE(out_arrays[i], Concatenate(arrays));
+ }
+ }
+
+ ARROW_ASSIGN_OR_RAISE(
+ std::shared_ptr<Array> struct_arr,
+ StructArray::Make(std::move(out_arrays), output_schema->fields()));
+
+ bool need_sort = !naive;
+ for (size_t i = num_aggregates; need_sort && i < out_arrays.size(); i++) {
+ if (output_schema->field(i)->type()->id() == Type::DICTIONARY) {
Review Comment:
Because the API is `Schema::field(int i)`. Around here, `i` is used in both
`int` (here) and `size_t` (at the for-loop) contexts, so a cast is needed.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
