Github user huawei-flink commented on a diff in the pull request:
https://github.com/apache/flink/pull/3547#discussion_r106366744
--- 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 --
Thanks a lot for the clarification. I am really willing to do it right, but
at the same time I need to understand. So, please be patient. :-)
When we started discussing the issue with @fhueske
(https://issues.apache.org/jira/browse/FLINK-5654?filter=-1) there was a
decision to use window, not process function.
Code consistency is pretty much the same, just extening a different
interface. I understand that ProcessFunction can manage its state, but window
checkpointing should replay all events in case of failure, so we would have
consistent processing even without managing this level of granularity in the
state. With procTime semantic, we can neglect retraction, and window can anyway
customize triggering function.
I don't understand the third point.
The main argument I see for this specific case is that ProcessFunction
supports granular state management. Besides the alleged code consistency.
---
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.
---
