[
https://issues.apache.org/jira/browse/HIVE-13957?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15317698#comment-15317698
]
Gopal V edited comment on HIVE-13957 at 6/7/16 2:36 AM:
--------------------------------------------------------
Would it be better to convert the implicit conversion (as done in GenericUDF)
into an explicit one if the type conversion is legal?
The tangential reason is that the type of the first argument is enforced for
the rest of the IN() expression - relaxing type safety and letting the UDF
handle it at runtime.
Similar type issues exist for concat() for instance, where the UDF internally
type-casts all args to String, but the actual plan doesn't.
was (Author: gopalv):
Would it be better to convert the implicit conversion (as done in GenericUDF)
into an explicit one if the type conversion is legal?
Similar type issues exist for concat() for instance, where the UDF internally
type-casts all args to String, but the actual plan doesn't.
> vectorized IN is inconsistent with non-vectorized (at least for decimal in
> (string))
> ------------------------------------------------------------------------------------
>
> Key: HIVE-13957
> URL: https://issues.apache.org/jira/browse/HIVE-13957
> Project: Hive
> Issue Type: Bug
> Reporter: Sergey Shelukhin
> Assignee: Sergey Shelukhin
> Attachments: HIVE-13957.patch, HIVE-13957.patch
>
>
> The cast is applied to the column in regular IN, but vectorized IN applies it
> to the IN() list
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
