Reading complex types with decimals inside with 'hive.parquet-batch-read-optimization-enabled=true' results in error

[denodo-research-labs]

Reading a Parquet file with complex types with decimals inside with hive.parquet-batch-read-optimization-enabled is true results in error.
It works when hive.parquet-batch-read-optimization-enabled is false.

Parquet definition of the complex type with decimals inside:

as_array_big_decimal: OPTIONAL F:1
.list:                REPEATED F:1
..element:            OPTIONAL INT32 L:DECIMAL(1,0) R:1 D:3

Your Environment

• Presto 0.289
• Storage S3
• Connector used: Hive connector
• Trace:

java.lang.UnsupportedOperationException: com.facebook.presto.common.block.IntArrayBlock at com.facebook.presto.common.block.Block.getLong(Block.java:81) at com.facebook.presto.common.type.ShortDecimalType.getObjectValue(ShortDecimalType.java:77) at com.facebook.presto.common.type.ArrayType.arrayBlockToObjectValues(ArrayType.java:160) at com.facebook.presto.common.type.ArrayType.lambda$getObjectValue$0(ArrayType.java:147) at com.facebook.presto.common.block.AbstractArrayBlock.apply(AbstractArrayBlock.java:298) at com.facebook.presto.common.type.ArrayType.getObjectValue(ArrayType.java:147) at com.facebook.presto.server.protocol.RowIterable$RowIterator.computeNext(RowIterable.java:77) at com.facebook.presto.server.protocol.RowIterable$RowIterator.computeNext(RowIterable.java:50) at com.google.common.collect.AbstractIterator.tryToComputeNext(AbstractIterator.java:145)

Expected Behavior

Vectorized reading should work.

Current Behavior

The following error is obtained:

[Error: java.lang.UnsupportedOperationException: com.facebook.presto.common.block.IntArrayBlock
    at com.facebook.presto.common.block.Block.getLong(Block.java:81)
    at com.facebook.presto.common.type.ShortDecimalType.getObjectValue(ShortDecimalType.java:77)...]

Steps to Reproduce

1. CREATE TABLE "hive"."default".hive_data_reader_array_primitives (
   as_array_int array(integer),
   as_array_long array(bigint),
   as_array_byte array(tinyint),
   as_array_short array(smallint),
   as_array_boolean array(boolean),
   as_array_float array(real),
   as_array_double array(double),
   as_array_string array(varchar),
   as_array_binary array(varbinary),
   as_array_big_decimal array(decimal(1,0))
   ) WITH (FORMAT = 'PARQUET', external_location = 's3a://path/to/parquet/')

2. Parquet file: https://github.com/prestodb/presto/blob/master/presto-delta/src/test/resources/delta_v3/data-reader-array-primitives/part-00000-13921577-6635-457e-9e3a-f32dc7fea982.c000.snappy.parquet

3. Hive catalog properties with hive.parquet-batch-read-optimization-enabled=true