[ https://issues.apache.org/jira/browse/TINKERPOP-2245?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17320073#comment-17320073 ]
ASF GitHub Bot commented on TINKERPOP-2245: ------------------------------------------- spmallette commented on a change in pull request #1414: URL: https://github.com/apache/tinkerpop/pull/1414#discussion_r612336793 ########## File path: gremlin-server/src/main/java/org/apache/tinkerpop/gremlin/server/handler/MultiTaskSession.java ########## @@ -0,0 +1,283 @@ +/* + * 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.tinkerpop.gremlin.server.handler; + +import org.apache.tinkerpop.gremlin.driver.message.ResponseMessage; +import org.apache.tinkerpop.gremlin.driver.message.ResponseStatusCode; +import org.apache.tinkerpop.gremlin.groovy.engine.GremlinExecutor; +import org.apache.tinkerpop.gremlin.groovy.jsr223.GroovyCompilerGremlinPlugin; +import org.apache.tinkerpop.gremlin.jsr223.GremlinScriptEngine; +import org.apache.tinkerpop.gremlin.jsr223.GremlinScriptEngineManager; +import org.apache.tinkerpop.gremlin.server.Settings; +import org.apache.tinkerpop.gremlin.structure.Transaction; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import javax.script.Bindings; +import java.util.HashMap; +import java.util.Map; +import java.util.concurrent.BlockingQueue; +import java.util.concurrent.ConcurrentMap; +import java.util.concurrent.ExecutorService; +import java.util.concurrent.LinkedBlockingQueue; +import java.util.concurrent.RejectedExecutionException; +import java.util.concurrent.ScheduledExecutorService; +import java.util.concurrent.ScheduledFuture; +import java.util.concurrent.TimeUnit; +import java.util.concurrent.atomic.AtomicBoolean; + +import static org.apache.tinkerpop.gremlin.server.op.session.SessionOpProcessor.CONFIG_GLOBAL_FUNCTION_CACHE_ENABLED; + +/** + * A {@link Session} implementation that queues tasks given to it and executes them in a serial fashion within the + * same thread which thus allows multiple tasks to be executed in the same transaction. The first {@link SessionTask} + * to execute is supplied on the constructor and additional ones may be added as they arrive with + * {@link #submitTask(SessionTask)} where they will be added to a queue where they will await execution in the thread + * bound to this session. + */ +public class MultiTaskSession extends AbstractSession { + private static final Logger logger = LoggerFactory.getLogger(MultiTaskSession.class); + protected final BlockingQueue<SessionTask> queue; + private final AtomicBoolean ending = new AtomicBoolean(false); + private final ScheduledExecutorService scheduledExecutorService; + private final GremlinScriptEngineManager scriptEngineManager; + private ScheduledFuture<?> requestCancelFuture; + private Bindings bindings; + + /** + * Creates a new {@code MultiTaskSession} object providing the initial starting {@link SessionTask} that gets + * executed by the session when it starts. + * + * @param initialSessionTask The tasks that starts the session. + * @param sessionId The id of the session + * @param sessions The session id to {@link Session} instances mapping + */ + public MultiTaskSession(final SessionTask initialSessionTask, final String sessionId, + final ConcurrentMap<String, Session> sessions) { + super(initialSessionTask, sessionId, false, sessions); + + queue = new LinkedBlockingQueue<>(initialSessionTask.getSettings().maxSessionTaskQueueSize); + + // using a global function cache is cheaper than creating a new on per session especially if you have to + // create a lot of sessions. it will generate a ton of throw-away objects. mostly keeping the option open + // to not use it to preserve the ability to use the old functionality if wanted or if there is some specific + // use case with sessions that needs it. if we wanted this could eventually become a per-request option + // so that the client could control it as necessary and get scriptengine isolation if they need it. + if (initialSessionTask.getSettings().useCommonEngineForSessions) + scriptEngineManager = initialSessionTask.getGremlinExecutor().getScriptEngineManager(); + else + scriptEngineManager = initializeGremlinExecutor(initialSessionTask).getScriptEngineManager(); + + scheduledExecutorService = initialSessionTask.getScheduledExecutorService(); + submitTask(initialSessionTask); + } + + /** + * Gets the script engine specific to this session which is dependent on the + * {@link Settings#useCommonEngineForSessions} configuration. + */ + @Override + public GremlinScriptEngine getScriptEngine(final SessionTask sessionTask, final String language) { + return scriptEngineManager.getEngineByName(language); + } + + @Override + public boolean isAcceptingTasks() { + return !ending.get(); + } + + @Override + public boolean submitTask(final SessionTask sessionTask) throws RejectedExecutionException { + try { + return isAcceptingTasks() && queue.add(sessionTask); + } catch (IllegalStateException ise) { + final String msg = String.format("Task %s rejected from session %s", + sessionTask.getRequestMessage().getRequestId(), getSessionId()); + throw new RejectedExecutionException(msg); + } + } + + @Override + public void run() { + // allow the Session to know about the thread that is running it - the thread really only has relevance + // once the session has started. + this.sessionThread = Thread.currentThread(); Review comment: so, the basic problem is that the TinkerPop API doesn't really let us move a transaction context across threads. There is a notion of the "threaded transactions" which does allow this but no one really implements that besides JanusGraph. will think about this some more but i think that ultimately it will expand scope beyond what this pull request should get into. there is already some documentation on the potential impact to performance that you describe but i will review again. -- 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 > Consolidate the executor for bytecode & string based client > ----------------------------------------------------------- > > Key: TINKERPOP-2245 > URL: https://issues.apache.org/jira/browse/TINKERPOP-2245 > Project: TinkerPop > Issue Type: Improvement > Components: server > Affects Versions: 3.4.2 > Reporter: Divij Vaidya > Assignee: Stephen Mallette > Priority: Minor > > We have two code paths in the server which perform (more or less) the same > functions. One is the executor for string based queries and other is the > executor for bytecode. This code can be refactored together so that the logic > to handle timeout, handle exception during execution, handle exception before > execution and others can consolidated. > [https://github.com/apache/tinkerpop/blob/master/gremlin-groovy/src/main/java/org/apache/tinkerpop/gremlin/groovy/engine/GremlinExecutor.java#L246] > and > [https://github.com/apache/tinkerpop/blob/master/gremlin-server/src/main/java/org/apache/tinkerpop/gremlin/server/op/traversal/TraversalOpProcessor.java#L333] -- This message was sent by Atlassian Jira (v8.3.4#803005)