huaxingao commented on code in PR #238:
URL:
https://github.com/apache/arrow-datafusion-comet/pull/238#discussion_r1550911908
##########
common/src/main/java/org/apache/comet/parquet/TypeUtil.java:
##########
@@ -196,7 +196,9 @@ && isUnsignedIntTypeMatched(logicalTypeAnnotation, 64)) {
|| canReadAsBinaryDecimal(descriptor, sparkType)
|| sparkType == DataTypes.BinaryType
// for uuid, since iceberg maps uuid to StringType
- || sparkType == DataTypes.StringType) {
+ || sparkType == DataTypes.StringType
+ && descriptor.getPrimitiveType().getLogicalTypeAnnotation()
+ instanceof
LogicalTypeAnnotation.UUIDLogicalTypeAnnotation) {
Review Comment:
@viirya If the `SparkType` is `StringType` and `LogicalTypeAnnotation` is
`UUID`, then this must be iceberg UUID column, because only iceberg maps UUID
to Spark `StringType`. I feel the change is safe. Or we can add an extra
parameter in
[getColumnReader](https://github.com/apache/arrow-datafusion-comet/blob/main/common/src/main/java/org/apache/comet/parquet/Utils.java#L39)
to indicate whether the ColumnReader is an Iceberg ColumnReader.
--
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]
