AlenkaF commented on code in PR #40359:
URL: https://github.com/apache/arrow/pull/40359#discussion_r1519691960
##########
cpp/src/arrow/record_batch.cc:
##########
@@ -248,70 +248,62 @@ Result<std::shared_ptr<StructArray>>
RecordBatch::ToStructArray() const {
/*offset=*/0);
}
+#define TYPE_CASE(type) \
+ case Type::type: { \
+ using T = typename TypeIdTraits<Type::type>::Type; \
+ using CType_in = typename TypeTraits<T>::CType; \
+ auto* in_values = data->GetValues<CType_in>(1); \
+ for (int64_t i = 0; i < arr.length(); ++i) { \
+ *out_values++ = static_cast<CType>(*in_values++); \
+ } \
+ break; \
+ }
+
template <typename DataType>
inline void ConvertColumnsToTensor(const RecordBatch& batch, uint8_t* out) {
using CType = typename arrow::TypeTraits<DataType>::CType;
auto* out_values = reinterpret_cast<CType*>(out);
- if (TypeTraits<DataType>::type_singleton() ==
- batch.column(0)->type()) { // If all columns are of same data type
- // Loop through all of the columns
- for (int i = 0; i < batch.num_columns(); ++i) {
- const auto* in_values = batch.column(i)->data()->GetValues<CType>(1);
-
- // Copy data of each column
- memcpy(out_values, in_values, sizeof(CType) * batch.num_rows());
- out_values += batch.num_rows();
- } // End loop through columns
-
- } else { // If columns have mixed data type
- // Loop through all of the columns
- for (int i = 0; i < batch.num_columns(); ++i) {
- const auto& arr = *batch.column(i);
- auto data = arr.data();
-
- // Copy data of each column
- for (int64_t i = 0; i < arr.length(); ++i) {
- switch (arr.type_id()) {
- case Type::UINT8:
- *out_values++ = static_cast<CType>(data->GetValues<uint8_t>(1)[i]);
- break;
- case Type::UINT16:
- case Type::HALF_FLOAT:
- *out_values++ =
static_cast<CType>(data->GetValues<uint16_t>(1)[i]);
- break;
- case Type::UINT32:
- *out_values++ =
static_cast<CType>(data->GetValues<uint32_t>(1)[i]);
- break;
- case Type::UINT64:
- *out_values++ =
static_cast<CType>(data->GetValues<uint64_t>(1)[i]);
- break;
- case Type::INT8:
- *out_values++ = static_cast<CType>(data->GetValues<int8_t>(1)[i]);
- break;
- case Type::INT16:
- *out_values++ = static_cast<CType>(data->GetValues<int16_t>(1)[i]);
- break;
- case Type::INT32:
- *out_values++ = static_cast<CType>(data->GetValues<int32_t>(1)[i]);
- break;
- case Type::INT64:
- *out_values++ = static_cast<CType>(data->GetValues<int64_t>(1)[i]);
- break;
- case Type::FLOAT:
- *out_values++ = static_cast<CType>(data->GetValues<float>(1)[i]);
- break;
- case Type::DOUBLE:
- *out_values++ = static_cast<CType>(data->GetValues<double>(1)[i]);
- break;
- default:
- break;
+ for (int i = 0; i < batch.num_columns(); ++i) {
+ const auto& arr = *batch.column(i);
+ auto data = arr.data();
+
+ // If the column is of the same type than resulting data type
+ if (TypeTraits<DataType>::type_singleton() == batch.column(0)->type()) {
Review Comment:
Oh, nevermind about `arr->type()`. There was another leftover from the
missed suggestion a bit up in the code.
--
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]
