KurtYoung commented on a change in pull request #8051: [FLINK-12018]
[table-planner-blink] Add support for generating optimized logical plan for
Sort and Rank
URL: https://github.com/apache/flink/pull/8051#discussion_r270618704
##########
File path:
flink-table/flink-table-planner-blink/src/main/scala/org/apache/flink/table/plan/nodes/calcite/Rank.scala
##########
@@ -91,10 +89,13 @@ abstract class Rank(
}
override def deriveRowType(): RelDataType = {
+ if (!outputRankNumber) {
+ return input.getRowType
+ }
val typeFactory = cluster.getRexBuilder.getTypeFactory
val typeBuilder = typeFactory.builder()
input.getRowType.getFieldList.foreach(typeBuilder.add)
- // rank function column is always the last column, and its type is BIGINT
NOT NULL
+ // rank number column is always the last column, and its type is BIGINT
NOT NULL
val allFieldNames = new util.HashSet[String]()
allFieldNames.addAll(input.getRowType.getFieldNames)
val rankFieldName = FlinkRelOptUtil.buildUniqueFieldName(allFieldNames,
"rk")
Review comment:
Why we can force the rank field name to "rk"? What if the user specified the
name for it? Like:
SELECT * FROM (
SELECT a, b, RANK() OVER (PARTITION BY b ORDER BY c) `myRank` FROM MyTable
) t
WHERE `myRank` < 10 ?
----------------------------------------------------------------
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:
us...@infra.apache.org
With regards,
Apache Git Services
