pitrou commented on code in PR #46532:
URL: https://github.com/apache/arrow/pull/46532#discussion_r2101866165
##########
cpp/src/parquet/column_reader.cc:
##########
@@ -2069,11 +2070,35 @@ class ByteArrayChunkedRecordReader final : public
TypedRecordReader<ByteArrayTyp
virtual public BinaryRecordReader {
public:
ByteArrayChunkedRecordReader(const ColumnDescriptor* descr, LevelInfo
leaf_info,
- ::arrow::MemoryPool* pool, bool
read_dense_for_nullable)
+ ::arrow::MemoryPool* pool, bool
read_dense_for_nullable,
+ std::shared_ptr<::arrow::DataType> arrow_type)
: TypedRecordReader<ByteArrayType>(descr, leaf_info, pool,
read_dense_for_nullable) {
ARROW_DCHECK_EQ(descr_->physical_type(), Type::BYTE_ARRAY);
- accumulator_.builder = std::make_unique<::arrow::BinaryBuilder>(pool);
+ auto arrow_binary_type = arrow_type ? arrow_type->id() :
::arrow::Type::BINARY;
Review Comment:
Given that `arrow_type` is a new optional parameter, I suppose it could.
If we decide that `RecordReader` is only an internal API, we can probably
remove the fallback, but it doesn't complicate the code much in any case.
--
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]
