Github user sunjincheng121 commented on a diff in the pull request:
https://github.com/apache/flink/pull/3386#discussion_r102927451
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/plan/nodes/datastream/DataStreamSlideEventTimeRowAgg.scala
---
@@ -0,0 +1,179 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.flink.table.plan.nodes.datastream
+
+import org.apache.calcite.plan.{RelOptCluster, RelTraitSet}
+import org.apache.calcite.rel.`type`.RelDataType
+import org.apache.calcite.rel.core.AggregateCall
+import org.apache.calcite.rel.{RelNode, RelWriter, SingleRel}
+import org.apache.flink.api.java.tuple.Tuple
+import org.apache.flink.types.Row
+import org.apache.flink.table.calcite.{FlinkRelBuilder, FlinkTypeFactory}
+import FlinkRelBuilder.NamedWindowProperty
+import org.apache.flink.table.runtime.aggregate.AggregateUtil._
+import org.apache.flink.table.runtime.aggregate._
+import org.apache.flink.streaming.api.datastream.{AllWindowedStream,
DataStream, WindowedStream}
+import org.apache.flink.streaming.api.windowing.assigners._
+import org.apache.flink.streaming.api.windowing.windows.{Window =>
DataStreamWindow}
+import org.apache.flink.table.api.StreamTableEnvironment
+import org.apache.flink.table.plan.nodes.CommonAggregate
+
+class DataStreamSlideEventTimeRowAgg(
+ namedProperties: Seq[NamedWindowProperty],
+ cluster: RelOptCluster,
+ traitSet: RelTraitSet,
+ inputNode: RelNode,
+ namedAggregates: Seq[CalcitePair[AggregateCall, String]],
+ rowRelDataType: RelDataType,
+ inputType: RelDataType,
+ grouping: Array[Int])
+ extends SingleRel(cluster, traitSet, inputNode)
+ with CommonAggregate
+ with DataStreamRel {
+
+ override def deriveRowType(): RelDataType = rowRelDataType
+
+ override def copy(traitSet: RelTraitSet, inputs:
java.util.List[RelNode]): RelNode = {
+ new DataStreamSlideEventTimeRowAgg(
+ namedProperties,
+ cluster,
+ traitSet,
+ inputs.get(0),
+ namedAggregates,
+ getRowType,
+ inputType,
+ grouping)
--- End diff --
I think `I check whether the current data is out of order in WindowOperator
isLate function, and now just discard if islate.` will not work well. Because
the logic in this method is:
```
if (windowAssigner instanceof GlobalEventTimeRowWindowAssigner) {
return
windowAssignerContext.getCurrentElementTime() <
windowAssignerContext.getCurrentMaxTime();
}
```
e.g. Test Data:
```
1, 1L, "Hi", 1400000L
2, 2L, "Hello", 1400005L
3, 2L, "Hello w", 1300000
4, 3L, "Hello world", 1400010L
```
You do not know which element first comes, so you will get different
results every time you run itï¼Just likeï¼
`SELECT` d, SUM(a) over (order by rowtime() range between unbounded
preceding and current row) from T1`
You can get the following resultsï¼
The first time:
```
1400005,2
1400010,6
```
The second time
```
1400000,1
1400010,5
```
The third time
```
1300000,3
1400005,5
1400010,9
```
Soï¼IMHO. Event-time over must handle the situation above. How do you
think?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
