felipeblazing commented on a change in pull request #10994:
URL: https://github.com/apache/arrow/pull/10994#discussion_r695975190
##########
File path: cpp/src/arrow/compute/kernels/hash_aggregate.cc
##########
@@ -933,31 +967,36 @@ struct GroupedReducingAggregator : public
GroupedAggregator {
}
Status Consume(const ExecBatch& batch) override {
- c_type* reduced = reduced_.mutable_data();
+ CType* reduced = reduced_.mutable_data();
int64_t* counts = counts_.mutable_data();
uint8_t* no_nulls = no_nulls_.mutable_data();
- auto g = batch[1].array()->GetValues<uint32_t>(1);
-
- return Impl::Consume(*batch[0].array(), reduced, counts, no_nulls, g);
+ VisitGroupedValues<Type>(
+ batch,
+ [&](uint32_t g, InputCType value) {
Review comment:
can we also capture explicitly here?
##########
File path: cpp/src/arrow/compute/kernels/hash_aggregate.cc
##########
@@ -1833,30 +1853,45 @@ struct GroupedBooleanAggregator : public
GroupedAggregator {
uint8_t* seen = seen_.mutable_data();
uint8_t* no_nulls = no_nulls_.mutable_data();
int64_t* counts = counts_.mutable_data();
- const auto& input = *batch[0].array();
auto g = batch[1].array()->GetValues<uint32_t>(1);
- if (input.MayHaveNulls()) {
- const uint8_t* bitmap = input.buffers[1]->data();
- arrow::internal::VisitBitBlocksVoid(
- input.buffers[0], input.offset, input.length,
- [&](int64_t position) {
- counts[*g]++;
- Impl::UpdateGroupWith(seen, *g, BitUtil::GetBit(bitmap, position));
- g++;
- },
- [&] { BitUtil::SetBitTo(no_nulls, *g++, false); });
+ if (batch[0].is_array()) {
+ const auto& input = *batch[0].array();
+ if (input.MayHaveNulls()) {
+ const uint8_t* bitmap = input.buffers[1]->data();
+ arrow::internal::VisitBitBlocksVoid(
+ input.buffers[0], input.offset, input.length,
+ [&](int64_t position) {
+ counts[*g]++;
+ Impl::UpdateGroupWith(seen, *g, BitUtil::GetBit(bitmap,
position));
+ g++;
+ },
+ [&] { BitUtil::SetBitTo(no_nulls, *g++, false); });
+ } else {
+ arrow::internal::VisitBitBlocksVoid(
+ input.buffers[1], input.offset, input.length,
+ [&](int64_t) {
+ Impl::UpdateGroupWith(seen, *g, true);
Review comment:
Just a question but could you explain briefly what seen's purpose is
here?
##########
File path: cpp/src/arrow/compute/kernels/hash_aggregate.cc
##########
@@ -824,6 +824,36 @@ Status AddHashAggKernels(
return Status::OK();
}
+template <typename Type, typename ConsumeValue, typename ConsumeNull>
+void VisitGroupedValues(const ExecBatch& batch, ConsumeValue&& valid_func,
+ ConsumeNull&& null_func) {
+ auto g = batch[1].array()->GetValues<uint32_t>(1);
+ if (batch[0].is_array()) {
+ VisitArrayValuesInline<Type>(
+ *batch[0].array(),
+ [&](typename TypeTraits<Type>::CType val) { valid_func(*g++, val); },
Review comment:
Is there a reason we don't want to be explicit in our capture here?
--
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]
