smallzhongfeng commented on code in PR #39558: URL: https://github.com/apache/spark/pull/39558#discussion_r1071334085
########## docs/sql-migration-guide.md: ########## @@ -36,6 +36,7 @@ license: | - Since Spark 3.4, Spark throws only `PartitionsAlreadyExistException` when it creates partitions but some of them exist already. In Spark 3.3 or earlier, Spark can throw either `PartitionsAlreadyExistException` or `PartitionAlreadyExistsException`. - Since Spark 3.4, Spark will do validation for partition spec in ALTER PARTITION to follow the behavior of `spark.sql.storeAssignmentPolicy` which may cause an exception if type conversion fails, e.g. `ALTER TABLE .. ADD PARTITION(p='a')` if column `p` is int type. To restore the legacy behavior, set `spark.sql.legacy.skipTypeValidationOnAlterPartition` to `true`. - Since Spark 3.4, vectorized readers are enabled by default for the nested data types (array, map and struct). To restore the legacy behavior, set `spark.sql.orc.enableNestedColumnVectorizedReader` and `spark.sql.parquet.enableNestedColumnVectorizedReader` to `false`. + - Since Spark 3.4, When inserting(adding/renaming/dropping) a partition, the partition field is of `String` and the partition value is not quoted, it will be treated as a field of `Numeric` by default. If you want to keep the partition field of `String`, we can set `spark.sql.legacy.keepPartitionSpecAsStringLiteral` to `true`. Review Comment: Okay, I think I just misunderstood. -- 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]
