lidavidm commented on a change in pull request #10538:
URL: https://github.com/apache/arrow/pull/10538#discussion_r663193637
##########
File path: cpp/src/arrow/compute/kernels/scalar_if_else.cc
##########
@@ -443,16 +451,217 @@ struct IfElseFunctor<Type, enable_if_number<Type>> {
// selectively copy values from left data
T left_data = internal::UnboxScalar<Type>::Unbox(left);
- RunIfElseLoop(
- cond,
- [&](int64_t data_offset, int64_t num_elems) {
- std::fill(out_values + data_offset, out_values + data_offset +
num_elems,
- left_data);
+ RunIfElseLoop(cond, [&](int64_t data_offset, int64_t num_elems) {
+ std::fill(out_values + data_offset, out_values + data_offset + num_elems,
+ left_data);
+ });
+
+ return Status::OK();
+ }
+};
+
+template <typename Type>
+struct IfElseFunctor<Type, enable_if_base_binary<Type>> {
+ using OffsetType = typename TypeTraits<Type>::OffsetType::c_type;
+ using ArrayType = typename TypeTraits<Type>::ArrayType;
+ using BuilderType = typename TypeTraits<Type>::BuilderType;
+
+ // A - Array, S - Scalar, X = Array/Scalar
+
+ // SXX
+ static Status Call(KernelContext* ctx, const BooleanScalar& cond, const
Datum& left,
+ const Datum& right, Datum* out) {
+ if (left.is_scalar() && right.is_scalar()) {
+ if (cond.is_valid) {
+ *out = cond.value ? left.scalar() : right.scalar();
+ } else {
+ *out = MakeNullScalar(left.type());
+ }
+ return Status::OK();
+ }
+ // either left or right is an array. Output is always an array
+ int64_t out_arr_len = std::max(left.length(), right.length());
+ if (!cond.is_valid) {
+ // cond is null; just create a null array
+ ARROW_ASSIGN_OR_RAISE(*out,
+ MakeArrayOfNull(left.type(), out_arr_len,
ctx->memory_pool()))
+ return Status::OK();
+ }
+
+ const auto& valid_data = cond.value ? left : right;
+ if (valid_data.is_array()) {
+ *out = valid_data;
+ } else {
+ // valid data is a scalar that needs to be broadcasted
+ ARROW_ASSIGN_OR_RAISE(*out, MakeArrayFromScalar(*valid_data.scalar(),
out_arr_len,
+ ctx->memory_pool()));
+ }
+ return Status::OK();
+ }
+
+ // AAA
+ static Status Call(KernelContext* ctx, const ArrayData& cond, const
ArrayData& left,
+ const ArrayData& right, ArrayData* out) {
+ const uint8_t* cond_data = cond.buffers[1]->data();
+ BitBlockCounter bit_counter(cond_data, cond.offset, cond.length);
+
+ const auto* left_offsets = left.GetValues<OffsetType>(1);
+ const uint8_t* left_data = left.buffers[2]->data();
+ const auto* right_offsets = right.GetValues<OffsetType>(1);
+ const uint8_t* right_data = right.buffers[2]->data();
+
+ // allocate data buffer conservatively
+ int64_t data_buff_alloc = left_offsets[left.length] - left_offsets[0] +
+ right_offsets[right.length] - right_offsets[0];
+
+ BuilderType builder(ctx->memory_pool());
+ ARROW_RETURN_NOT_OK(builder.Reserve(cond.length + 1));
+ ARROW_RETURN_NOT_OK(builder.ReserveData(data_buff_alloc));
+
+ RunLoop(
+ cond, *out,
+ [&](int64_t i) {
+ builder.UnsafeAppend(left_data + left_offsets[i],
+ left_offsets[i + 1] - left_offsets[i]);
+ },
+ [&](int64_t i) {
+ builder.UnsafeAppend(right_data + right_offsets[i],
+ right_offsets[i + 1] - right_offsets[i]);
+ },
+ [&]() { builder.AppendNull(); });
+ ARROW_ASSIGN_OR_RAISE(auto out_arr, builder.Finish());
+
+ out->buffers[0] = std::move(out_arr->data()->buffers[0]);
+ out->buffers[1] = std::move(out_arr->data()->buffers[1]);
+ out->buffers[2] = std::move(out_arr->data()->buffers[2]);
Review comment:
nit: maybe also copy over null_count?
--
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]
