jfsii commented on code in PR #3628:
URL: https://github.com/apache/hive/pull/3628#discussion_r983634554
##########
ql/src/java/org/apache/hadoop/hive/ql/io/parquet/serde/ParquetHiveSerDe.java:
##########
@@ -91,11 +93,61 @@ public class ParquetHiveSerDe extends AbstractSerDe
implements SchemaInference {
private ObjectInspector objInspector;
private ParquetHiveRecord parquetRow;
+ private ObjectInspectorConverters.Converter converter;
public ParquetHiveSerDe() {
parquetRow = new ParquetHiveRecord();
}
+ // Recursively check if CHAR or VARCHAR types are used
+ private boolean needsConversion(TypeInfo type) {
Review Comment:
Yes - I agree it could be a perf hit here. Though it wouldn't surprise me if
the perf difference is small since these paths already do many allocations per
row.
BUT - I did change my approach with what I think is a much more appropriate
solution that hopefully avoids the extra allocation.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
