[ https://issues.apache.org/jira/browse/CURATOR-549?focusedWorklogId=411717&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-411717 ]
ASF GitHub Bot logged work on CURATOR-549: ------------------------------------------ Author: ASF GitHub Bot Created on: 28/Mar/20 13:51 Start Date: 28/Mar/20 13:51 Worklog Time Spent: 10m Work Description: Randgalt commented on pull request #335: [CURATOR-549] Recipes based on Persistent Recursive Watchers URL: https://github.com/apache/curator/pull/335#discussion_r399664766 ########## File path: curator-recipes/src/main/java/org/apache/curator/framework/recipes/cache/CuratorCacheImpl.java ########## @@ -0,0 +1,301 @@ +/** + * 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.curator.framework.recipes.cache; + +import com.google.common.annotations.VisibleForTesting; +import com.google.common.base.Preconditions; +import com.google.common.collect.Sets; +import org.apache.curator.framework.CuratorFramework; +import org.apache.curator.framework.api.BackgroundCallback; +import org.apache.curator.framework.api.CuratorEvent; +import org.apache.curator.framework.listen.Listenable; +import org.apache.curator.framework.listen.StandardListenerManager; +import org.apache.curator.framework.recipes.watch.PersistentWatcher; +import org.apache.curator.utils.ThreadUtils; +import org.apache.curator.utils.ZKPaths; +import org.apache.zookeeper.KeeperException; +import org.apache.zookeeper.WatchedEvent; +import org.apache.zookeeper.data.Stat; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; +import java.util.Collections; +import java.util.Optional; +import java.util.Set; +import java.util.concurrent.atomic.AtomicReference; +import java.util.function.Consumer; +import java.util.stream.Stream; + +import static org.apache.curator.framework.recipes.cache.CuratorCacheListener.Type.*; +import static org.apache.zookeeper.KeeperException.Code.NONODE; +import static org.apache.zookeeper.KeeperException.Code.OK; + +class CuratorCacheImpl implements CuratorCache +{ + private final Logger log = LoggerFactory.getLogger(getClass()); + private final AtomicReference<State> state = new AtomicReference<>(State.LATENT); + private final PersistentWatcher persistentWatcher; + private final CuratorFramework client; + private final CuratorCacheStorage storage; + private final String path; + private final boolean recursive; + private final boolean compressedData; + private final boolean clearOnClose; + private final StandardListenerManager<CuratorCacheListener> listenerManager = StandardListenerManager.standard(); + private final Consumer<Exception> exceptionHandler; + private final OutstandingOps outstandingOps = new OutstandingOps(() -> callListeners(CuratorCacheListener::initialized)); + + private enum State + { + LATENT, + STARTED, + CLOSED + } + + CuratorCacheImpl(CuratorFramework client, CuratorCacheStorage storage, String path, Options[] optionsArg, Consumer<Exception> exceptionHandler) + { + Set<Options> options = (optionsArg != null) ? Sets.newHashSet(optionsArg) : Collections.emptySet(); + this.client = client; + this.storage = (storage != null) ? storage : CuratorCacheStorage.standard(); + this.path = path; + recursive = !options.contains(Options.SINGLE_NODE_CACHE); + compressedData = options.contains(Options.COMPRESSED_DATA); + clearOnClose = !options.contains(Options.DO_NOT_CLEAR_ON_CLOSE); + persistentWatcher = new PersistentWatcher(client, path, recursive); + persistentWatcher.getListenable().addListener(this::processEvent); + persistentWatcher.getResetListenable().addListener(this::rebuild); + this.exceptionHandler = (exceptionHandler != null) ? exceptionHandler : e -> log.error("CuratorCache error", e); + } + + @Override + public void start() + { + Preconditions.checkState(state.compareAndSet(State.LATENT, State.STARTED), "Already started"); + persistentWatcher.start(); + } + + @Override + public void close() + { + if ( state.compareAndSet(State.STARTED, State.CLOSED) ) + { + persistentWatcher.close(); + if ( clearOnClose ) + { + storage.clear(); + } + } + } + + @Override + public Listenable<CuratorCacheListener> listenable() + { + return listenerManager; + } + + @Override + public Optional<ChildData> get(String path) + { + return storage.get(path); + } + + @Override + public int size() + { + return storage.size(); + } + + @Override + public Stream<ChildData> stream() + { + return storage.stream(); + } + + @VisibleForTesting + CuratorCacheStorage storage() + { + return storage; + } + + private void rebuild() + { + if ( state.get() != State.STARTED ) + { + return; + } + + // rebuild from the root first + nodeChanged(path); + + // rebuild remaining nodes - note: this may cause some nodes to be queried twice + // (though versions checks will minimize that). If someone can think of a better + // way let us know + storage.stream() + .map(ChildData::getPath) + .filter(p -> !p.equals(path)) + .forEach(this::nodeChanged); + } + + private void processEvent(WatchedEvent event) + { + if ( state.get() != State.STARTED ) + { + return; + } + + // NOTE: Persistent/Recursive watchers never trigger NodeChildrenChanged + + switch ( event.getType() ) + { + case NodeDataChanged: + case NodeCreated: + { + nodeChanged(event.getPath()); + break; + } + + case NodeDeleted: + { + removeStorage(event.getPath()); + break; + } + } + } + + private void checkChildrenChanged(String fromPath, Stat oldStat, Stat newStat) + { + if ( (state.get() != State.STARTED) || !recursive ) + { + return; + } + + if ( (oldStat != null) && (oldStat.getCversion() == newStat.getCversion()) ) + { + return; // children haven't changed + } + + try + { + BackgroundCallback callback = (__, event) -> { + if ( event.getResultCode() == OK.intValue() ) + { + event.getChildren().forEach(child -> nodeChanged(ZKPaths.makePath(fromPath, child))); + } + else if ( event.getResultCode() == NONODE.intValue() ) + { + removeStorage(event.getPath()); + } + else + { + handleException(event); + } + outstandingOps.decrement(); + }; + + outstandingOps.increment(); + client.getChildren().inBackground(callback).forPath(fromPath); + } + catch ( Exception e ) + { + handleException(e); + } + } + + private void nodeChanged(String fromPath) + { + if ( state.get() != State.STARTED ) + { + return; + } + + try + { + BackgroundCallback callback = (__, event) -> { + if ( event.getResultCode() == OK.intValue() ) + { + Optional<ChildData> childData = putStorage(new ChildData(event.getPath(), event.getStat(), event.getData())); + checkChildrenChanged(event.getPath(), childData.map(ChildData::getStat).orElse(null), event.getStat()); + } + else if ( event.getResultCode() == NONODE.intValue() ) + { + removeStorage(event.getPath()); + } + else + { + handleException(event); + } + outstandingOps.decrement(); + }; + + outstandingOps.increment(); + if ( compressedData ) + { + client.getData().decompressed().inBackground(callback).forPath(fromPath); Review comment: Yes, background callbacks are called from the single ZK event thread. Is there a concern? ---------------------------------------------------------------- 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 Issue Time Tracking ------------------- Worklog Id: (was: 411717) Time Spent: 14h (was: 13h 50m) > ZooKeeper 3.6 will add support for Persistent Recursive Watchers - Add > Curator support > -------------------------------------------------------------------------------------- > > Key: CURATOR-549 > URL: https://issues.apache.org/jira/browse/CURATOR-549 > Project: Apache Curator > Issue Type: Improvement > Components: Client, Documentation, Framework, Recipes, Tests > Affects Versions: 4.2.0 > Reporter: Jordan Zimmerman > Assignee: Jordan Zimmerman > Priority: Major > Fix For: 5.0.0 > > Time Spent: 14h > Remaining Estimate: 0h > > ZOOKEEPER-1416 will add support for Persistent Recursive Watchers. This > feature allows Curator to re-think/re-write the various cache recipes so that > they're simpler, use less resources and more stable. > * Change Curator to depend on ZooKeeper 3.6.x > * Add a new module so that backward compatibility with 3.5 is ensured > * Add new methods corresponding to {{ZooKeeper.addWatch()}} > * Add a new recipe that enables a fully managed Persistent watch > * Create new Cache recipes that use the new Persistent watch recipe > * Replace internal usages of TreeCache/PathChildrenCache/NodeCache with new > cache recipe > * Deprecate TreeCache/PathChildrenCache/NodeCache > * Updated docs/tests/etc -- This message was sent by Atlassian Jira (v8.3.4#803005)