lirui-apache commented on issue #9927: [FLINK-14397][hive] Failed to run Hive UDTF with array arguments URL: https://github.com/apache/flink/pull/9927#issuecomment-543590413 > This change seems to assume that all elements in the list would be of the same type, while the original version doesn't. Wondering if the new assumption always hold for Hive array? Yeah I think it's safe to assume elements in a Hive array are of the same type. Hive [checks element type](https://github.com/apache/hive/blob/rel/release-2.3.4/serde/src/java/org/apache/hadoop/hive/serde2/typeinfo/ListTypeInfo.java#L80) in the `equals` method of `ListTypeInfo`, which is used to determine whether data insertion is allowed or conversion is needed. For example, it's not allowed to insert an `array<int>` into a column of `array<bigint>` type because the src and dest have different type info. I also want to clarify that this new assumption/limitation is imposed on us by Flink. While Hive UDTF may very well handle Object[], Flink assumes an array argument should contains elements of the same type. Therefore, even if Hive fails to satisfy this assumption (due to bug or behavior change), I think we have to cover it up in `toFlinkObject`.
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
