Based on some questions on SO, mainly this one:
Throttling asynchronous tasks
I have implemented the SemaphoreSlim object to concurrently processes requests over a range of methods in my application. Most of these methods are taking in lists of IDs and getting single byte arrays back per ID in a concurrent fashion from the web. The implementation looks like this:
using (var semaphore = new SemaphoreSlim(MaxConcurrency))
{
var tasks = fileMetadata.GroupBy(x => x.StorageType).Select(async storageTypeFileMetadata=>
{
await semaphore.WaitAsync();
try
{
var fileManager = FileManagerFactory.CreateFileManager((StorageType)storageTypeFileMetadata.Key);
await fileManager.UpdateFilesAsync(storageTypeFileMetadata);
}
finally
{
semaphore.Release();
}
});
await Task.WhenAll(tasks);
}
Is there a way to abstract out a method or some reusable code snippet for the semaphore code, and pass in the work I need done, so it can be reused without re-writing the semaphore code each time? The only difference amongst multiple methods using this same semaphore pattern is the list I am iterating and the work it is doing in the try{}.
I am thinking something like pass list.select(x=> my task method with my work in it) to a semaphore method which is all the wrapper semaphore code.
So I'm guessing something like:
public static class Extension
{
public static async Task ExecuteAsync<T>(this IEnumerable<T> items, Func<T, Task> task, int concurrency)
{
var tasks = new List<Task>();
using (var semaphore = new SemaphoreSlim(concurrency))
{
foreach (var item in items)
{
tasks.Add(ExecuteInSemaphore(semaphore, task, item));
}
await Task.WhenAll(tasks);
}
}
private static async Task ExecuteInSemaphore<T>(SemaphoreSlim semaphore, Func<T, Task> task, T item)
{
await semaphore.WaitAsync();
try
{
await task(item);
}
finally
{
semaphore.Release();
}
}
}
Then you would use it like:
await fileMetadata.GroupBy(x => x.StorageType).ExecuteAsync(storageTypeFileMetadata =>
{
var fileManager = FileManagerFactory.CreateFileManager((StorageType)storageTypeFileMetadata.Key);
return fileManager.UpdateFilesAsync(storageTypeFileMetadata);
}, 4);
Related
Is it possible to use Async when using ForEach? Below is the code I am trying:
using (DataContext db = new DataLayer.DataContext())
{
db.Groups.ToList().ForEach(i => async {
await GetAdminsFromGroup(i.Gid);
});
}
I am getting the error:
The name 'Async' does not exist in the current context
The method the using statement is enclosed in is set to async.
List<T>.ForEach doesn't play particularly well with async (neither does LINQ-to-objects, for the same reasons).
In this case, I recommend projecting each element into an asynchronous operation, and you can then (asynchronously) wait for them all to complete.
using (DataContext db = new DataLayer.DataContext())
{
var tasks = db.Groups.ToList().Select(i => GetAdminsFromGroupAsync(i.Gid));
var results = await Task.WhenAll(tasks);
}
The benefits of this approach over giving an async delegate to ForEach are:
Error handling is more proper. Exceptions from async void cannot be caught with catch; this approach will propagate exceptions at the await Task.WhenAll line, allowing natural exception handling.
You know that the tasks are complete at the end of this method, since it does an await Task.WhenAll. If you use async void, you cannot easily tell when the operations have completed.
This approach has a natural syntax for retrieving the results. GetAdminsFromGroupAsync sounds like it's an operation that produces a result (the admins), and such code is more natural if such operations can return their results rather than setting a value as a side effect.
This little extension method should give you exception-safe async iteration:
public static async Task ForEachAsync<T>(this List<T> list, Func<T, Task> func)
{
foreach (var value in list)
{
await func(value);
}
}
Since we're changing the return type of the lambda from void to Task, exceptions will propagate up correctly. This will allow you to write something like this in practice:
await db.Groups.ToList().ForEachAsync(async i => {
await GetAdminsFromGroup(i.Gid);
});
Starting with C# 8.0, you can create and consume streams asynchronously.
private async void button1_Click(object sender, EventArgs e)
{
IAsyncEnumerable<int> enumerable = GenerateSequence();
await foreach (var i in enumerable)
{
Debug.WriteLine(i);
}
}
public static async IAsyncEnumerable<int> GenerateSequence()
{
for (int i = 0; i < 20; i++)
{
await Task.Delay(100);
yield return i;
}
}
More
The simple answer is to use the foreach keyword instead of the ForEach() method of List().
using (DataContext db = new DataLayer.DataContext())
{
foreach(var i in db.Groups)
{
await GetAdminsFromGroup(i.Gid);
}
}
Here is an actual working version of the above async foreach variants with sequential processing:
public static async Task ForEachAsync<T>(this List<T> enumerable, Action<T> action)
{
foreach (var item in enumerable)
await Task.Run(() => { action(item); }).ConfigureAwait(false);
}
Here is the implementation:
public async void SequentialAsync()
{
var list = new List<Action>();
Action action1 = () => {
//do stuff 1
};
Action action2 = () => {
//do stuff 2
};
list.Add(action1);
list.Add(action2);
await list.ForEachAsync();
}
What's the key difference? .ConfigureAwait(false); which keeps the context of main thread while async sequential processing of each task.
This is not an old question, but .Net 6 introduced Parallel.ForeachAsync:
var collectionToIterate = db.Groups.ToList();
await Parallel.ForEachAsync(collectionToIterate, async (i, token) =>
{
await GetAdminsFromGroup(i);
});
ForeachAsync also accepts a ParallelOptions object, but usually you don't want to mess with the MaxDegreeOfParallelism property:
ParallelOptions parallelOptions = new ParallelOptions { MaxDegreeOfParallelism = 4 };
var collectionToIterate = db.Groups.ToList();
await Parallel.ForEachAsync(collectionToIterate, parallelOptions , async (i, token) =>
{
await GetAdminsFromGroup(i);
});
From Microsoft Docs: https://learn.microsoft.com/en-us/dotnet/api/system.threading.tasks.paralleloptions.maxdegreeofparallelism?view=net-6.0
By default, For and ForEach will utilize however many threads the underlying scheduler provides, so changing MaxDegreeOfParallelism from the default only limits how many concurrent tasks will be used.
Generally, you do not need to modify this setting....
Add this extension method
public static class ForEachAsyncExtension
{
public static Task ForEachAsync<T>(this IEnumerable<T> source, int dop, Func<T, Task> body)
{
return Task.WhenAll(from partition in Partitioner.Create(source).GetPartitions(dop)
select Task.Run(async delegate
{
using (partition)
while (partition.MoveNext())
await body(partition.Current).ConfigureAwait(false);
}));
}
}
And then use like so:
Task.Run(async () =>
{
var s3 = new AmazonS3Client(Config.Instance.Aws.Credentials, Config.Instance.Aws.RegionEndpoint);
var buckets = await s3.ListBucketsAsync();
foreach (var s3Bucket in buckets.Buckets)
{
if (s3Bucket.BucketName.StartsWith("mybucket-"))
{
log.Information("Bucket => {BucketName}", s3Bucket.BucketName);
ListObjectsResponse objects;
try
{
objects = await s3.ListObjectsAsync(s3Bucket.BucketName);
}
catch
{
log.Error("Error getting objects. Bucket => {BucketName}", s3Bucket.BucketName);
continue;
}
// ForEachAsync (4 is how many tasks you want to run in parallel)
await objects.S3Objects.ForEachAsync(4, async s3Object =>
{
try
{
log.Information("Bucket => {BucketName} => {Key}", s3Bucket.BucketName, s3Object.Key);
await s3.DeleteObjectAsync(s3Bucket.BucketName, s3Object.Key);
}
catch
{
log.Error("Error deleting bucket {BucketName} object {Key}", s3Bucket.BucketName, s3Object.Key);
}
});
try
{
await s3.DeleteBucketAsync(s3Bucket.BucketName);
}
catch
{
log.Error("Error deleting bucket {BucketName}", s3Bucket.BucketName);
}
}
}
}).Wait();
If you are using EntityFramework.Core there is an extension method ForEachAsync.
The example usage looks like this:
using Microsoft.EntityFrameworkCore;
using System.Threading.Tasks;
public class Example
{
private readonly DbContext _dbContext;
public Example(DbContext dbContext)
{
_dbContext = dbContext;
}
public async void LogicMethod()
{
await _dbContext.Set<dbTable>().ForEachAsync(async x =>
{
//logic
await AsyncTask(x);
});
}
public async Task<bool> AsyncTask(object x)
{
//other logic
return await Task.FromResult<bool>(true);
}
}
I would like to add that there is a Parallel class with ForEach function built in that can be used for this purpose.
The problem was that the async keyword needs to appear before the lambda, not before the body:
db.Groups.ToList().ForEach(async (i) => {
await GetAdminsFromGroup(i.Gid);
});
This is method I created to handle async scenarios with ForEach.
If one of tasks fails then other tasks will continue their execution.
You have ability to add function that will be executed on every exception.
Exceptions are being collected as aggregateException at the end and are available for you.
Can handle CancellationToken
public static class ParallelExecutor
{
/// <summary>
/// Executes asynchronously given function on all elements of given enumerable with task count restriction.
/// Executor will continue starting new tasks even if one of the tasks throws. If at least one of the tasks throwed exception then <see cref="AggregateException"/> is throwed at the end of the method run.
/// </summary>
/// <typeparam name="T">Type of elements in enumerable</typeparam>
/// <param name="maxTaskCount">The maximum task count.</param>
/// <param name="enumerable">The enumerable.</param>
/// <param name="asyncFunc">asynchronous function that will be executed on every element of the enumerable. MUST be thread safe.</param>
/// <param name="onException">Acton that will be executed on every exception that would be thrown by asyncFunc. CAN be thread unsafe.</param>
/// <param name="cancellationToken">The cancellation token.</param>
public static async Task ForEachAsync<T>(int maxTaskCount, IEnumerable<T> enumerable, Func<T, Task> asyncFunc, Action<Exception> onException = null, CancellationToken cancellationToken = default)
{
using var semaphore = new SemaphoreSlim(initialCount: maxTaskCount, maxCount: maxTaskCount);
// This `lockObject` is used only in `catch { }` block.
object lockObject = new object();
var exceptions = new List<Exception>();
var tasks = new Task[enumerable.Count()];
int i = 0;
try
{
foreach (var t in enumerable)
{
await semaphore.WaitAsync(cancellationToken);
tasks[i++] = Task.Run(
async () =>
{
try
{
await asyncFunc(t);
}
catch (Exception e)
{
if (onException != null)
{
lock (lockObject)
{
onException.Invoke(e);
}
}
// This exception will be swallowed here but it will be collected at the end of ForEachAsync method in order to generate AggregateException.
throw;
}
finally
{
semaphore.Release();
}
}, cancellationToken);
if (cancellationToken.IsCancellationRequested)
{
break;
}
}
}
catch (OperationCanceledException e)
{
exceptions.Add(e);
}
foreach (var t in tasks)
{
if (cancellationToken.IsCancellationRequested)
{
break;
}
// Exception handling in this case is actually pretty fast.
// https://gist.github.com/shoter/d943500eda37c7d99461ce3dace42141
try
{
await t;
}
#pragma warning disable CA1031 // Do not catch general exception types - we want to throw that exception later as aggregate exception. Nothing wrong here.
catch (Exception e)
#pragma warning restore CA1031 // Do not catch general exception types
{
exceptions.Add(e);
}
}
if (exceptions.Any())
{
throw new AggregateException(exceptions);
}
}
}
I am using a LINQ lambda expression to generate an array of tasks and add them to a list of tasks. I want to implement a Semaphore to limit the number of tasks in process at a given point in time.
Here is code that represents what I currently have:
class Test {
private List<Task> tasks;
public void Start(){
tasks = new List<Task>();
AddTasks();
}
private void AddTasks(){
tasks.AddRange(
items
.Where(x => x.InProcess == false)
.Select( async (item) = > {
await DoWork(item);
})
.ToArray()
);
}
}
I want to achieve something similar to:
class Test {
private List<Task> tasks;
private SemaphoreSlim semaphore ;
public void Start(){
tasks = new List<Task>();
semaphore = new SemaphoreSlim(5);
AddTasks();
}
private void AddTasks(){
tasks.AddRange(
items
.Where(x => x.InProcess == false)
.Select( async (item) = > {
await semaphore.AwaitAsync();
try
{
await DoWork(item);
}
catch (System.Exception)
{
throw;
}
finally {
semaphore.Release();
}
})
.ToArray()
);
}
}
But I do not think this will work correctly, as the semaphore is inside of the task. How can I use the semaphore as part of the LINQ query to stall generating new tasks until the semaphore is awaited?
You can try this code. It should work as you expected but I haven't tried it. And don't forget to call await Task.WhenAll(tasks) when it's done and you need to get result.
private void AddTasks()
{
tasks.AddRange(items
.Where(x => x.InProcess == false)
.Select(AddTaskAsync)
.ToArray());
//later await Task.WhenAll(tasks);
}
private async Task AddTaskAsync(YourClass item)
{
await semaphore.WaitAsync();
try
{
await DoWork(item);
}
finally
{
semaphore.Release();
}
}
My example code did accomplish what I intended. Using the semaphore inside of the Task didn't cause any problems. It prevented the Task from running, which was the goal.
I have a simple class that does a synchronous thing,
public static class Synchronous
{
public static void DoTheWholeThing()
{
AStuff aStuff;
using (var a = new A())
{
aStuff = a.GetStuff();
}
BStuff bStuff;
using (var b = new B())
{
bStuff = b.GetStuff();
}
var combination = CombineStuff(aStuff, bStuff);
}
private static Combination CombineStuff(AStuff aStuff, BStuff bStuff)
{
//// Magic Here
}
}
Obviously, this code is not fully defined but it does illustrate my question.
Now, the classes A and B are both responsible for retrieving data from different remote sources. Consequently, the developers of A and B have implemented asynchronous entry points called GetStuffAsync which return Task<AStuff> and Task<BStuff> respectively.
I want to take maximum advantage of the asynchronous methods and call them concurrently so I can reduce the overall wait time of my code.
Here is what I've concocted, so far.
public static class Asynchronous
{
public async static Task DoTheWholeThing(CancellationToken cancellationToken)
{
var getAStuffTask = new Func<Task<AStuff>>(
async () =>
{
using (var a = new A())
{
return await a.GetStuffAsync(cancellationToken);
}
})();
var getBStuffTask = new Func<Task<BStuff>>(
async () =>
{
using (var b = new B())
{
return await b.GetStuffAsync(cancellationToken);
}
})();
var combination = CombineStuff(
await getAStuffTask,
await getBStuffTask);
}
private Combination CombineStuff(AStuff aStuff, BStuff bStuff)
{
//// Magic Here
}
}
Aside from this code looking curiously like the javascript module pattern, is this the correct approach. I don't think I should be using Task.Run as this code is clearly not CPU bound.
It seems a bit "clunky" that I need to instantiate typed delegates to do this. Is there a better way?
EDIT
following two good answers I'm in a quandary between named functions and continuations.
The code becomes radically simpler when you simply extract the anonymous methods out into named methods:
public async static Task DoTheWholeThing(CancellationToken cancellationToken)
{
var getAStuffTask = GetAStuffAsync(cancellationToken);
var getBStuffTask = GetBStuffAsync(cancellationToken);
var combination = CombineStuff(
await getAStuffTask,
await getBStuffTask);
}
private static async Task<AStuff> GetAStuffAsync(CancellationToken cancellationToken)
{
using (var a = new A())
{
return await a.GetStuffAsync(cancellationToken);
}
}
private static async Task<BStuff> GetBStuffAsync(CancellationToken cancellationToken)
{
using (var b = new B())
{
return await b.GetStuffAsync(cancellationToken);
}
}
That said, if you really want to stick with the anonymous methods, you can create a helper method that will allow generic type inference and lambdas to implicitly figure out the type of the delegate:
public async static Task DoTheWholeThing(CancellationToken cancellationToken)
{
var getAStuffTask = Start(async () =>
{
using (var a = new A())
{
return await a.GetStuffAsync(cancellationToken);
}
});
var getBStuffTask = Start(async () =>
{
using (var b = new B())
{
return await b.GetStuffAsync(cancellationToken);
}
});
var combination = CombineStuff(
await getAStuffTask,
await getBStuffTask);
}
public static Task<T> Start<T>(Func<Task<T>> asyncOperation)
{
return asyncOperation();
}
Use TPL continuations to call Dispose as soon as the task is complete.
public async static Task DoTheWholeThing(CancellationToken cancellationToken)
{
var a = new A();
var b = new B();
// start the tasks and store them for awaiting later
var getAStuffTask = a.GetStuffAsync(cancellationToken);
var getBStuffTask = b.GetStuffAsync(cancellationToken);
// queue up continuations to dispose of the resource as soon as it is not needed
getAStuffTask.ContinueWith(() => a.Dispose());
getBStuffTask.ContinueWith(() => b.Dispose());
// await as normal
var combination = CombineStuff(
await getAStuffTask,
await getBStuffTask);
}
I am unsure if wrapping the whole method in an addition using block will accomplish anything but it may provide peace of mind.
You don't need to wrap your async calls in delegates to get them to execute immediately. If you call the GetStuffAsync methods directly without awaiting them you will have the same result.
public static class Asynchronous
{
public async static Task DoTheWholeThing(CancellationToken cancellationToken)
{
using (var a = new A())
using (var b = new B()) {
var taskA = a.GetStuffAsync(cancellationToken);
var taskB = b.GetStuffAsync(cancellationToken);
await Task.WhenAll(new [] { taskA, taskB });
var combination = CombineStuff(taskA.Result, taskB.Result);
}
}
private Combination CombineStuff(AStuff aStuff, BStuff bStuff)
{
//// Magic Here
}
}
Note that this does keep the a and b objects alive during the call to CombineStuff as #Servy notes. If that is a problem the declaration of the Task objects can be moved outside of the using blocks as below:
public static class Asynchronous
{
public async static Task DoTheWholeThing(CancellationToken cancellationToken)
{
Task taskA;
Task taskB;
using (var a = new A())
using (var b = new B()) {
taskA = a.GetStuffAsync(cancellationToken);
taskB = b.GetStuffAsync(cancellationToken);
await Task.WhenAll(new [] { taskA, taskB });
}
var combination = CombineStuff(taskA.Result, taskB.Result);
}
private Combination CombineStuff(AStuff aStuff, BStuff bStuff)
{
//// Magic Here
}
}
Although this still holds onto a and b as long as both tasks are running, rather than disposing of each as they return.
Is it possible to use Async when using ForEach? Below is the code I am trying:
using (DataContext db = new DataLayer.DataContext())
{
db.Groups.ToList().ForEach(i => async {
await GetAdminsFromGroup(i.Gid);
});
}
I am getting the error:
The name 'Async' does not exist in the current context
The method the using statement is enclosed in is set to async.
List<T>.ForEach doesn't play particularly well with async (neither does LINQ-to-objects, for the same reasons).
In this case, I recommend projecting each element into an asynchronous operation, and you can then (asynchronously) wait for them all to complete.
using (DataContext db = new DataLayer.DataContext())
{
var tasks = db.Groups.ToList().Select(i => GetAdminsFromGroupAsync(i.Gid));
var results = await Task.WhenAll(tasks);
}
The benefits of this approach over giving an async delegate to ForEach are:
Error handling is more proper. Exceptions from async void cannot be caught with catch; this approach will propagate exceptions at the await Task.WhenAll line, allowing natural exception handling.
You know that the tasks are complete at the end of this method, since it does an await Task.WhenAll. If you use async void, you cannot easily tell when the operations have completed.
This approach has a natural syntax for retrieving the results. GetAdminsFromGroupAsync sounds like it's an operation that produces a result (the admins), and such code is more natural if such operations can return their results rather than setting a value as a side effect.
This little extension method should give you exception-safe async iteration:
public static async Task ForEachAsync<T>(this List<T> list, Func<T, Task> func)
{
foreach (var value in list)
{
await func(value);
}
}
Since we're changing the return type of the lambda from void to Task, exceptions will propagate up correctly. This will allow you to write something like this in practice:
await db.Groups.ToList().ForEachAsync(async i => {
await GetAdminsFromGroup(i.Gid);
});
Starting with C# 8.0, you can create and consume streams asynchronously.
private async void button1_Click(object sender, EventArgs e)
{
IAsyncEnumerable<int> enumerable = GenerateSequence();
await foreach (var i in enumerable)
{
Debug.WriteLine(i);
}
}
public static async IAsyncEnumerable<int> GenerateSequence()
{
for (int i = 0; i < 20; i++)
{
await Task.Delay(100);
yield return i;
}
}
More
The simple answer is to use the foreach keyword instead of the ForEach() method of List().
using (DataContext db = new DataLayer.DataContext())
{
foreach(var i in db.Groups)
{
await GetAdminsFromGroup(i.Gid);
}
}
Here is an actual working version of the above async foreach variants with sequential processing:
public static async Task ForEachAsync<T>(this List<T> enumerable, Action<T> action)
{
foreach (var item in enumerable)
await Task.Run(() => { action(item); }).ConfigureAwait(false);
}
Here is the implementation:
public async void SequentialAsync()
{
var list = new List<Action>();
Action action1 = () => {
//do stuff 1
};
Action action2 = () => {
//do stuff 2
};
list.Add(action1);
list.Add(action2);
await list.ForEachAsync();
}
What's the key difference? .ConfigureAwait(false); which keeps the context of main thread while async sequential processing of each task.
This is not an old question, but .Net 6 introduced Parallel.ForeachAsync:
var collectionToIterate = db.Groups.ToList();
await Parallel.ForEachAsync(collectionToIterate, async (i, token) =>
{
await GetAdminsFromGroup(i);
});
ForeachAsync also accepts a ParallelOptions object, but usually you don't want to mess with the MaxDegreeOfParallelism property:
ParallelOptions parallelOptions = new ParallelOptions { MaxDegreeOfParallelism = 4 };
var collectionToIterate = db.Groups.ToList();
await Parallel.ForEachAsync(collectionToIterate, parallelOptions , async (i, token) =>
{
await GetAdminsFromGroup(i);
});
From Microsoft Docs: https://learn.microsoft.com/en-us/dotnet/api/system.threading.tasks.paralleloptions.maxdegreeofparallelism?view=net-6.0
By default, For and ForEach will utilize however many threads the underlying scheduler provides, so changing MaxDegreeOfParallelism from the default only limits how many concurrent tasks will be used.
Generally, you do not need to modify this setting....
Add this extension method
public static class ForEachAsyncExtension
{
public static Task ForEachAsync<T>(this IEnumerable<T> source, int dop, Func<T, Task> body)
{
return Task.WhenAll(from partition in Partitioner.Create(source).GetPartitions(dop)
select Task.Run(async delegate
{
using (partition)
while (partition.MoveNext())
await body(partition.Current).ConfigureAwait(false);
}));
}
}
And then use like so:
Task.Run(async () =>
{
var s3 = new AmazonS3Client(Config.Instance.Aws.Credentials, Config.Instance.Aws.RegionEndpoint);
var buckets = await s3.ListBucketsAsync();
foreach (var s3Bucket in buckets.Buckets)
{
if (s3Bucket.BucketName.StartsWith("mybucket-"))
{
log.Information("Bucket => {BucketName}", s3Bucket.BucketName);
ListObjectsResponse objects;
try
{
objects = await s3.ListObjectsAsync(s3Bucket.BucketName);
}
catch
{
log.Error("Error getting objects. Bucket => {BucketName}", s3Bucket.BucketName);
continue;
}
// ForEachAsync (4 is how many tasks you want to run in parallel)
await objects.S3Objects.ForEachAsync(4, async s3Object =>
{
try
{
log.Information("Bucket => {BucketName} => {Key}", s3Bucket.BucketName, s3Object.Key);
await s3.DeleteObjectAsync(s3Bucket.BucketName, s3Object.Key);
}
catch
{
log.Error("Error deleting bucket {BucketName} object {Key}", s3Bucket.BucketName, s3Object.Key);
}
});
try
{
await s3.DeleteBucketAsync(s3Bucket.BucketName);
}
catch
{
log.Error("Error deleting bucket {BucketName}", s3Bucket.BucketName);
}
}
}
}).Wait();
If you are using EntityFramework.Core there is an extension method ForEachAsync.
The example usage looks like this:
using Microsoft.EntityFrameworkCore;
using System.Threading.Tasks;
public class Example
{
private readonly DbContext _dbContext;
public Example(DbContext dbContext)
{
_dbContext = dbContext;
}
public async void LogicMethod()
{
await _dbContext.Set<dbTable>().ForEachAsync(async x =>
{
//logic
await AsyncTask(x);
});
}
public async Task<bool> AsyncTask(object x)
{
//other logic
return await Task.FromResult<bool>(true);
}
}
I would like to add that there is a Parallel class with ForEach function built in that can be used for this purpose.
The problem was that the async keyword needs to appear before the lambda, not before the body:
db.Groups.ToList().ForEach(async (i) => {
await GetAdminsFromGroup(i.Gid);
});
This is method I created to handle async scenarios with ForEach.
If one of tasks fails then other tasks will continue their execution.
You have ability to add function that will be executed on every exception.
Exceptions are being collected as aggregateException at the end and are available for you.
Can handle CancellationToken
public static class ParallelExecutor
{
/// <summary>
/// Executes asynchronously given function on all elements of given enumerable with task count restriction.
/// Executor will continue starting new tasks even if one of the tasks throws. If at least one of the tasks throwed exception then <see cref="AggregateException"/> is throwed at the end of the method run.
/// </summary>
/// <typeparam name="T">Type of elements in enumerable</typeparam>
/// <param name="maxTaskCount">The maximum task count.</param>
/// <param name="enumerable">The enumerable.</param>
/// <param name="asyncFunc">asynchronous function that will be executed on every element of the enumerable. MUST be thread safe.</param>
/// <param name="onException">Acton that will be executed on every exception that would be thrown by asyncFunc. CAN be thread unsafe.</param>
/// <param name="cancellationToken">The cancellation token.</param>
public static async Task ForEachAsync<T>(int maxTaskCount, IEnumerable<T> enumerable, Func<T, Task> asyncFunc, Action<Exception> onException = null, CancellationToken cancellationToken = default)
{
using var semaphore = new SemaphoreSlim(initialCount: maxTaskCount, maxCount: maxTaskCount);
// This `lockObject` is used only in `catch { }` block.
object lockObject = new object();
var exceptions = new List<Exception>();
var tasks = new Task[enumerable.Count()];
int i = 0;
try
{
foreach (var t in enumerable)
{
await semaphore.WaitAsync(cancellationToken);
tasks[i++] = Task.Run(
async () =>
{
try
{
await asyncFunc(t);
}
catch (Exception e)
{
if (onException != null)
{
lock (lockObject)
{
onException.Invoke(e);
}
}
// This exception will be swallowed here but it will be collected at the end of ForEachAsync method in order to generate AggregateException.
throw;
}
finally
{
semaphore.Release();
}
}, cancellationToken);
if (cancellationToken.IsCancellationRequested)
{
break;
}
}
}
catch (OperationCanceledException e)
{
exceptions.Add(e);
}
foreach (var t in tasks)
{
if (cancellationToken.IsCancellationRequested)
{
break;
}
// Exception handling in this case is actually pretty fast.
// https://gist.github.com/shoter/d943500eda37c7d99461ce3dace42141
try
{
await t;
}
#pragma warning disable CA1031 // Do not catch general exception types - we want to throw that exception later as aggregate exception. Nothing wrong here.
catch (Exception e)
#pragma warning restore CA1031 // Do not catch general exception types
{
exceptions.Add(e);
}
}
if (exceptions.Any())
{
throw new AggregateException(exceptions);
}
}
}
I have a long running Task that uses callbacks to feed data incrementally (rather than a single ContinueWith() callback at the end).
I want to be able to pass the Task object back in this callback for task identification purposes (using Task.CurrentId)
However, I cannot work out how to pass the Task object into the task delegate. There appears to be no overload to do this, and I can't use the closure to do it since the task object is not defined at that point.
eg.
public Task StartDoingSomeStuff(CallbackDelegate callback)
{
Task task = Task.Factory.StartNew(() =>
{
while(whatever)
{
var results = DoSomeStuff();
callback(results, task); //CS0165. How do I get hold of the task?
}
});
return task;
}
gives:
error CS0165: Use of unassigned local variable 'task'
Split declaring the variable and assigning the task to it into two statements. Make sure that you don't use the variable before the task has been assigned:
public Task StartDoingSomeStuff(CallbackDelegate callback)
{
var gate = new object();
lock (gate)
{
Task task = null;
task = Task.Factory.StartNew(() =>
{
lock (gate)
{
while (whatever)
{
var results = DoSomeStuff();
callback(results, task);
}
}
});
return task;
}
}
Another solution is to create a task key and a dictionary mapping keys to tasks, then pass the key as state to the task action:
var taskMap = new Dictionary<object, Task>();
var taskKey = new object();
taskMap.Add(taskKey, Task.Factory.StartNew(key => { callback(results, key); }, taskKey));
of course you then have to look up the task from the key, which may or may not suit your scenario.
This should work although it is bad practice:
public Task StartDoingSomeStuff(CallbackDelegate callback)
{
Task task = null;
task = Task.Factory.StartNew(() =>
{
while(whatever)
{
var results = DoSomeStuff();
callback(results, task); //CS0165. How do I get hold of the task?
}
});
return task;
}