Github user dongjoon-hyun commented on the issue:
https://github.com/apache/spark/pull/18460
Hi, @cloud-fan . I moved it to `findTightestCommonType`.
For the comment `Cast doesn't know it and we will end up executing
unnecessary casts.`, the `canCast` of the current Spark already ignores names
[here](https://github.com/apache/spark/blob/master/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/Cast.scala#L76-L84),
doesn't it?
```scala
scala> spark.version
res2: String = 2.2.0
scala> sql("select cast(struct(1) as struct<a:int>)").explain(true)
== Parsed Logical Plan ==
'Project [unresolvedalias(cast(named_struct(col1, 1) as struct<a:int>),
None)]
+- OneRowRelation$
== Analyzed Logical Plan ==
named_struct(col1, 1): struct<a:int>
Project [cast(named_struct(col1, 1) as struct<a:int>) AS named_struct(col1,
1)#15]
+- OneRowRelation$
== Optimized Logical Plan ==
Project [[1] AS named_struct(col1, 1)#15]
+- OneRowRelation$
== Physical Plan ==
*Project [[1] AS named_struct(col1, 1)#15]
+- Scan OneRowRelation[]
```
---
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]
