Github user fhueske commented on a diff in the pull request:
https://github.com/apache/flink/pull/3547#discussion_r106488388
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/runtime/aggregate/BoundedProcessingOverWindowFunction.scala
---
@@ -0,0 +1,97 @@
+/*
+ * 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.runtime.aggregate
+
+import org.apache.flink.api.java.tuple.Tuple
+import org.apache.flink.types.Row
+import org.apache.flink.configuration.Configuration
+import
org.apache.flink.streaming.api.functions.windowing.RichWindowFunction
+import org.apache.flink.streaming.api.windowing.windows.Window
+import org.apache.flink.util.Collector
+import org.apache.flink.table.functions.AggregateFunction
+import org.apache.flink.api.java.typeutils.RowTypeInfo
+import org.apache.flink.util.Preconditions
+import org.apache.flink.table.functions.Accumulator
+import java.lang.Iterable
+
+class BoundedProcessingOverWindowFunction[W <: Window](
--- End diff --
As @huawei-flink pointed out, I agreed to use a WindowFunction for the
bounded OVER ROW case in our previous discussions. Although it would not be
consistent with the other OVER windows, this would be an easier solution, IMO.
However, I think we have to use a `ProcessFunction` for correctness
reasons. An OVER aggregation must emit one row for each input row. If we have
an `OVER (PARTITION BY a ORDER BY procTime() ROWS BETWEEN 3 PRECEDING AND
CURRENT ROW)` and implement this with a `SlidingCountWindow(3, 1)` we will lose
the two rows because the window function is called the first time, when three
elements have arrived.
Regarding the aggregation of COUNT DISTINCT, we could implement a custom
aggregation function, similar to the retractable min or max aggregation
function.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
