Github user dbtsai commented on a diff in the pull request:
https://github.com/apache/spark/pull/21235#discussion_r186240261
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/DataFrameWriter.scala ---
@@ -339,9 +339,16 @@ final class DataFrameWriter[T] private[sql](ds:
Dataset[T]) {
}
}
- private def assertNotBucketed(operation: String): Unit = {
- if (numBuckets.isDefined || sortColumnNames.isDefined) {
- throw new AnalysisException(s"'$operation' does not support
bucketing right now")
+ private def assertNotBucketedOrSorted(operation: String): Unit = {
+ (numBuckets.isDefined, sortColumnNames.isDefined) match {
+ case (true, true) =>
+ throw new AnalysisException(
+ s"'$operation' does not support bucketing and sorting right now")
+ case (true, false) =>
+ throw new AnalysisException(s"'$operation' does not support
bucketing right now")
+ case (false, true) =>
+ throw new AnalysisException(s"'$operation' does not support
sorting right now")
--- End diff --
I know this is the sorting in each bucket.
If a user just calls `writer.sortBy` without calling `bucketBy`, the user
will get `s"'$operation' does not support bucketing right now"` which is hard
to understand what's going on.
For the case of sortBy is enabled, and bucketBy is disabled, how about I
change the error message to `sortBy must be used together with bucketBy, and
'$operation' does not support bucketBy right now`
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
