curcur commented on a change in pull request #14799:
URL: https://github.com/apache/flink/pull/14799#discussion_r568316262



##########
File path: 
flink-runtime/src/main/java/org/apache/flink/runtime/state/proxy/ProxyKeyedStateBackend.java
##########
@@ -0,0 +1,306 @@
+/*
+ * 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.runtime.state.proxy;
+
+import org.apache.flink.api.common.state.AggregatingStateDescriptor;
+import org.apache.flink.api.common.state.ListStateDescriptor;
+import org.apache.flink.api.common.state.MapStateDescriptor;
+import org.apache.flink.api.common.state.ReducingStateDescriptor;
+import org.apache.flink.api.common.state.State;
+import org.apache.flink.api.common.state.StateDescriptor;
+import org.apache.flink.api.common.state.ValueStateDescriptor;
+import org.apache.flink.api.common.typeutils.TypeSerializer;
+import org.apache.flink.api.java.tuple.Tuple2;
+import org.apache.flink.runtime.checkpoint.CheckpointOptions;
+import org.apache.flink.runtime.state.AbstractKeyedStateBackend;
+import org.apache.flink.runtime.state.CheckpointStreamFactory;
+import org.apache.flink.runtime.state.KeyGroupedInternalPriorityQueue;
+import org.apache.flink.runtime.state.Keyed;
+import org.apache.flink.runtime.state.KeyedStateFunction;
+import org.apache.flink.runtime.state.KeyedStateHandle;
+import org.apache.flink.runtime.state.PriorityComparable;
+import org.apache.flink.runtime.state.SnapshotResult;
+import org.apache.flink.runtime.state.StateSnapshotTransformer;
+import org.apache.flink.runtime.state.heap.HeapKeyedStateBackend;
+import org.apache.flink.runtime.state.heap.HeapPriorityQueueElement;
+import org.apache.flink.runtime.state.internal.InternalKvState;
+import org.apache.flink.util.FlinkRuntimeException;
+import org.apache.flink.util.IOUtils;
+
+import javax.annotation.Nonnull;
+
+import java.util.List;
+import java.util.Map;
+import java.util.concurrent.RunnableFuture;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
+
+/** */
+public class ProxyKeyedStateBackend<K> extends AbstractKeyedStateBackend<K> {
+    // wrapped keyed state backend, either HeapKeyedStateBackend or 
RocksDBKeyedStateBackend
+    AbstractKeyedStateBackend<K> keyedStateBackend;
+
+    private static final Map<Class<? extends StateDescriptor>, StateFactory> 
STATE_FACTORIES =
+            Stream.of(
+                            Tuple2.of(
+                                    ValueStateDescriptor.class,
+                                    (StateFactory) ProxyValueState::create),
+                            Tuple2.of(
+                                    ListStateDescriptor.class,

Review comment:
       > 1. Yes, because of unnecessary complexity (plus `instanceof` is more 
flexible than keys of `class` type).
   > 2. Can you explain why is it more flexible and extensible than just 
`if-else`?. And why the same level of flexibility and extensibility is needed 
in wrapping and wrapped backends?
   > 3. I'd prefer such a refactoring (if any) to be performed before adding a 
new backend
   
   **Why more flexible and extensible:**
   as you said, the mappings can be updated dynamically or programmatically, 
   like `mapper.put (new type, new factory)`; 
   and also easy for check `mapper.get(a type)` is existed or not. 
   
   That's something if-else can not simply and cleanly do
   
   **why the same level of flexibility and extensibility**
   If the underlying can support dynamically add a new type, as what is now; I 
do want the wrapper to support "dynamically add a new type" as well.




----------------------------------------------------------------
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


Reply via email to