alex-plekhanov commented on a change in pull request #8960: URL: https://github.com/apache/ignite/pull/8960#discussion_r607007843
########## File path: modules/core/src/main/java/org/apache/ignite/internal/client/thin/ClientCacheEntryListenersRegistry.java ########## @@ -0,0 +1,70 @@ +/* + * 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.ignite.internal.client.thin; + +import java.util.Map; +import java.util.concurrent.ConcurrentHashMap; +import javax.cache.configuration.CacheEntryListenerConfiguration; + +/** + * Per-cache cache entry listeners registry. Listeners can't be stored inside ClientCache instance, since there can be + * several such instances per one cache. + */ +public class ClientCacheEntryListenersRegistry { Review comment: > As I understand, this limitation already exists - all messages have an ID in any case. There is no requirement for the uniqueness of this ID. We can use 0 for some events (for example, we have topology change event implemented as special flag of every response, but it can be implemented with notification and no additional resource ID required here). Also, perhaps in the future events initiated by server-side for some resource will be needed, in this case, client should be able to listen some type of notification with any resource ID. > why have Collection<NotificationListener>, then separate maps in Compute and Continuous Query We can replace the first collection with `EnumMap` (Listener per ClientOperation), it is very lightweight and fast, and there will be no iterations. There will be at most one listener per one operation type, and each listener can implement its own strategy for resource ID mapping. This change does not require to refactor compute a lot, so, I think, it can be implemented in the current ticket. Or we can even provide two API on client channel level: listener for client operation and any resource id (refactor compute to use this API) and listener for client operation and exactly one resource Id. WDYT? -- 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: [email protected]
