| File: System\Linq\Parallel\Scheduling\SpoolingTaskBase.cs | Web Access |
| Project: src\runtime\src\libraries\System.Linq.Parallel\src\System.Linq.Parallel.csproj (System.Linq.Parallel) |
// Licensed to the .NET Foundation under one or more agreements. // The .NET Foundation licenses this file to you under the MIT license. // =+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ // // SpoolingTaskBase.cs // // =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- using System.Collections.Generic; using System.Diagnostics; using System.Threading; namespace System.Linq.Parallel { /// <summary> /// A spooling task handles marshaling data from a producer to a consumer. It simply /// takes data from a producer and hands it off to a consumer. This class is the base /// class from which other concrete spooling tasks derive, encapsulating some common /// logic (such as capturing exceptions). /// </summary> internal abstract class SpoolingTaskBase : QueryTask { //----------------------------------------------------------------------------------- // Constructs a new spooling task. // // Arguments: // taskIndex - the unique index of this task // protected SpoolingTaskBase(int taskIndex, QueryTaskGroupState groupState) : base(taskIndex, groupState) { } //----------------------------------------------------------------------------------- // The implementation of the Work API just enumerates the producer's data, and // enqueues it into the consumer channel. Well, really, it just defers to extension // points (below) implemented by subclasses to do these things. // protected override void Work() { try { // Defer to the base class for the actual work. SpoolingWork(); } catch (Exception ex) { if (ex is OperationCanceledException oce && oce.CancellationToken == _groupState.CancellationState.MergedCancellationToken && _groupState.CancellationState.MergedCancellationToken.IsCancellationRequested) { //an expected internal cancellation has occurred. suppress this exception. } else { // TPL will catch and store the exception on the task object. We'll then later // turn around and wait on it, having the effect of propagating it. In the meantime, // we want to cooperative cancel all workers. Debug.Assert(_groupState.CancellationState.InternalCancellationTokenSource != null); _groupState.CancellationState.InternalCancellationTokenSource.Cancel(); // And then repropagate to let TPL catch it. throw; } } finally { SpoolingFinally(); //dispose resources etc. } } //----------------------------------------------------------------------------------- // This method is responsible for enumerating results and enqueueing them to // the output channel(s) as appropriate. Each base class implements its own. // protected abstract void SpoolingWork(); //----------------------------------------------------------------------------------- // If the subclass needs to do something in the finally block of the main work loop, // it should override this and do it. Purely optional. // protected virtual void SpoolingFinally() { // (Intentionally left blank.) } } }