Github user fhueske commented on a diff in the pull request:
https://github.com/apache/flink/pull/4625#discussion_r139407531
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/runtime/join/TimeBoundedStreamInnerJoin.scala
---
@@ -55,18 +49,18 @@ import org.apache.flink.util.Collector
* @param timeIndicator indicate whether joining on proctime or rowtime
*
*/
-class TimeBoundedStreamInnerJoin(
- private val leftLowerBound: Long,
- private val leftUpperBound: Long,
- private val allowedLateness: Long,
- private val leftType: TypeInformation[Row],
- private val rightType: TypeInformation[Row],
- private val genJoinFuncName: String,
- private val genJoinFuncCode: String,
- private val leftTimeIdx: Int,
- private val rightTimeIdx: Int,
- private val timeIndicator: JoinTimeIndicator)
- extends CoProcessFunction[CRow, CRow, CRow]
+abstract class TimeBoundedStreamInnerJoin(
+ private val leftLowerBound: Long,
+ private val leftUpperBound: Long,
+ private val allowedLateness: Long,
+ private val leftType: TypeInformation[Row],
+ private val rightType: TypeInformation[Row],
+ private val genJoinFuncName: String,
+ private val genJoinFuncCode: String,
+ private val leftTimeIdx: Int,
+ private val rightTimeIdx: Int,
+ private val timeIndicator: JoinTimeIndicator)
--- End diff --
We can remove `timeIndicator` and `JoinTimeIndicator`. They are only used
for the state names which do not need to distinguish row and processing time.
---