mihailom-db commented on code in PR #46987:
URL: https://github.com/apache/spark/pull/46987#discussion_r1642267167
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/CollationTypeCasts.scala:
##########
@@ -78,6 +78,11 @@ object CollationTypeCasts extends TypeCoercionRule {
}
raiseError.withNewChildren(Seq(raiseError.errorClass, newErrorParams))
+ case framelessOffsetWindow @ (_: Lag | _: Lead) =>
Review Comment:
Apart from implicit casting for inputs no. But this change introduces only
implicit casting for collations. The problem is that we have
```
override def inputTypes: Seq[AbstractDataType] =
Seq(AnyDataType, IntegerType, TypeCollection(input.dataType, NullType))
```
Because of this we need to make sure we know what behaviour is expected. I
would say as last parameter is default value, we would expect this not to fail
when we pass a literal with default collation i.e. we would need to make sure
that literal is cast to the proper collation for the `inputTypeCheck` not to
fail.
--
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]
