Github user windpiger commented on the issue:
https://github.com/apache/spark/pull/15918
this change of `schemaFor` will affect the default behavior.
For example:
[udf in
functions.scala](https://github.com/apache/spark/blob/master/sql/core/src/main/scala/org/apache/spark/sql/functions.scala#L3177)
if the input type is not supported and fall back to kryo with BinaryType,
here `inputTypes` before will be `Nil` which is ok for the following
expression's typecheck, but after fall back to kryo it will failed when
typecheck(BinaryType not equal to ReturnType), so I left logic to throw
exception in the `schemaFor` for `un-nested` unsupported type , while for
complex nested including `Map` etc, it will fall back to kryo.
even if I use `schemaForDefaultBinaryType ` not `schemaFor` in
`deserializerFor ` , and I think it is ok for this serde situation. But I am
not sure the change in `schmaFor` will affect other logics like `udf` described
above.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
