[
https://issues.apache.org/jira/browse/FLINK-27238?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
hehuiyuan updated FLINK-27238:
------------------------------
Attachment: image-2022-04-14-15-43-07-544.png
> Whether the HiveGenericUDTF and HiveScalarFunction should support primitive
> array and element type may be not null ，for example Array<Double>
> Array<String> ...
> ---------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: FLINK-27238
> URL: https://issues.apache.org/jira/browse/FLINK-27238
> Project: Flink
> Issue Type: Bug
> Reporter: hehuiyuan
> Priority: Minor
> Attachments: image-2022-04-14-10-27-50-340.png,
> image-2022-04-14-15-43-07-544.png
>
>
> !image-2022-04-14-10-27-50-340.png|width=381,height=260!
>
> If argTypes[0] is Array<Double> ,it will throw exception.
>
> {code:java}
> Caused by: org.apache.flink.table.functions.hive.FlinkHiveUDFException: Flink
> doesn't support primitive array for Hive functions yet.
> at
> org.apache.flink.table.functions.hive.util.HiveFunctionUtil.isSingleBoxedArray(HiveFunctionUtil.java:34)
> {code}
>
>
>
> I think we should determine whether the argTypes is Array , replace primitive
> Array type.`
> If argsType is Array , inputs should be converted to an object[].
> If argsType is not Array, we can keep origin code logic.
> HiveFunctionUtil.isSingleBoxedArray(argTypes); throws exception when argTypes
> is Array and the element type is BOOLEAN | TINYINT | SMALLINT | INTEGER |
> BIGINT | FLOAT | DOUBLE.
> Correct me if I am wrong， thanks
>
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
