cloud-fan commented on a change in pull request #27478: [SPARK-25829][SQL] Add
config `spark.sql.legacy.allowDuplicatedMapKeys` and change the default behavior
URL: https://github.com/apache/spark/pull/27478#discussion_r379831096
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/util/ArrayBasedMapBuilder.scala
##########
@@ -63,6 +68,11 @@ class ArrayBasedMapBuilder(keyType: DataType, valueType:
DataType) extends Seria
keys.append(key)
values.append(value)
} else {
+ if (!allowDuplicatedMapKey) {
+ throw new RuntimeException(s"Duplicate map key $key was founded,
please set " +
Review comment:
We shouldn't recommend users to set legacy config to fix a problem. We
should ask them to check the input data to see why there are duplicated keys.
If they want to remove duplicated keys, tell them that they can set the legacy
config which uses last wins policy.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
