public class FooHandler : HttpTaskAsyncHandler
{
public override async Task ProcessRequestAsync(HttpContext context)
{
return await new AdRequest().ProcessRequest();
// getting error here. "Return type of async type is void"
}
}
public class FooRequest
{
public async Task<String> ProcessRequest()
{
//return await "foo"; obviously nothing to wait here
}
}
I want to make a async handler and just want to return a string. How can i get this working? and is there a concise reference to work with Async methods and Tasks?
A few points:
You can await any Task, not just ones returned from async methods.
async methods wrap their returned value into a Task<TResult>; if there is no return value, they wrap the return itself into a Task.
There are several convenience methods available, e.g., Task.FromResult, if you don't need the overhead of an async method.
Only make a method async if you have to use await in it. If you don't need to make the method async, don't.
You may find my async/await intro helpful.
public class FooHandler : HttpTaskAsyncHandler
{
public override Task ProcessRequestAsync(HttpContext context)
{
return new AdRequest().ProcessRequest();
}
}
public class AdRequest
{
public Task<String> ProcessRequest()
{
return Task.FromResult("foo");
}
}
You shouldn't "return" the Task, the compiler will do it implicitly as it is an async function:
public override async Task ProcessRequestAsync(HttpContext context)
{
await new AdRequest().ProcessRequest();
}
public async Task<String> ProcessRequest()
{
return "foo";
}
This is another way, closer to what you were trying to do: (without async/await)
public override Task ProcessRequestAsync(HttpContext context)
{
return new AdRequest().ProcessRequest();
}
public Task<String> ProcessRequest()
{
return Task.Return("foo");
}
A general reference to async is here
Essentially adding the async modifier to a method, makes it return a Task implicitly. If you return an int, it will turn it into a Task<int>. await does the opposite, turning a Task<int> into an int.
This is a truly asynchronous method:
public Task<string> ProcessRequest()
{
var textFile = File.OpenText("file.txt");
var readTask = textFile.ReadToEndAsync();
readTask.ContinueWith(previousTask => textFile.Dispose());
return readTask;
}
If you run this method with a large file or a file on a slow drive the execution will return to caller long before file reading ends. In Stephen Cleary's example the caller will get back control only when the result ("foo") is finished calculating.
Dispose must be in ContinueWith because the method execution will return to caller before file reading is complete so file can't be closed in ProcessRequest method.
One can of course start their own task.
public Task<string> ProcessRequest(CancellationToken cancellationToken)
{
var readTask = Task.Run(() =>
{
using (var textFile = File.OpenText("file.txt"))
{
var text = textFile.ReadToEnd();
cancellationToken.ThrowIfCancellationRequested();
var processedText = text.Replace("foo", "bar");
return processedText;
}
});
return readTask;
}
It is a good practice to have a CancellationToken and periodically check if cancellation was requested to allow long running operarions to be cancelled.
Edit 1
As #Stephen Cleary highlighted the first sample and this result in approximately or maybe exactly the same CIL:
public async Task<string> ProcessRequest()
{
using (var textFile = File.OpenText("file.txt"))
{
var s = await textFile.ReadToEndAsync();
return s;
}
}
Basically the compiler will transform the code following await textFile.ReadToEndAsync() into ContinueWith.
Each syntax has its benefits, my preference is that 1-2 lines (i.e. dispose and log) go into ContinueWith, more complex continuation uses await.
Related
I have a series of methods (with variable number of parameters) that return a Task
I want to create a method that does something before and after each of this methods, by passing the Task.
I've simplified everything (removed cancellationToken, actual processing, etc) in this sample:
public async Task<string> GetDataString()
{
Console.WriteLine("Executing");
return "test";
}
public async Task<T> Process<T>(Task<T> task)
{
Console.WriteLine("Before");
var res = await task;
Console.WriteLine("After");
return res;
}
And in my main:
Task<string> task = GetDataString();
string result = await Process<string>(tasks);
Console.WriteLine(res);
the console output is
Executing
Before
After
test
What can I do to create the task but not actually starting it? And starting it only before the wait?
I managed to do it by creating a PauseToken, as explained in this article:
https://devblogs.microsoft.com/pfxteam/cooperatively-pausing-async-methods/
but I wonder if is there a better way.
Thanks,
Mattia
Your generic ProcessAsync method could accept a task factory as argument:
public async Task<T> ProcessAsync<T>(Func<Task<T>> taskFactory)
{
Console.WriteLine("Before");
var res = await taskFactory();
Console.WriteLine("After");
return res;
}
This way the task will be created at the time you'll invoke the factory method. You are in control of its creation.
Here is an example of calling the ProcessAsync method, passing as factory a lambda:
var result = await ProcessAsync(() => GetDataStringAsync(arg1, arg2));
This way you are not restricted to a factory method without arguments.
For completeness I should mention that Task objects can also created in a cold state using the constructor new Task(), and started later using the Start method, but this approach is not recommended.
You can remove the async keyword (from GetDataString) and create a new task which will be executed when you await
so the result of the code below is : before , executing , test , after
private static async Task Start()
{
Task<string> task = GetDataString();
string result = await Process<string>(task);
Console.WriteLine(result);
Console.ReadLine();
}
public Task<string> GetDataString()
{
return new TaskFactory(TaskScheduler.Default).StartNew(() =>
{
Console.WriteLine("Executing");
return "test";
});
}
public async Task<T> Process<T>(Task<T> task)
{
Console.WriteLine("Before");
var res = await task;
Console.WriteLine("After");
return res;
}
I'm working on a series of methods that execute many different database calls using entity framework. Many of these methods can run asynchronously as I really don't care about their output, and don't rely on them.
However, when I try implementing certain methods, I get a warning from the compiler saying: "Because this call is not awaited, the current method continues to run before the call is completed"
But to me, this seems like that's my desired behavior, as I don't care what the methods do.
Here's an example of the methods
public async Task SetupAccessControl(int objectTypeId, int objectId, int? organizationId)
{
using (var context = new SupportContext(CustomerId))
{
... // omitted for brevity
if (objectTypeId == (int) ObjectType.User)
{
AddToUserRoleBridge("Everyone", objectId);//Warning on this line
AddToUserRoleBridge("Default", objectId); //Warning on this line
}
... // omitted for brevity
}
}
public async Task AddToUserRoleBridge(string role, int userId)
{
using (var context = new SupportContext(CustomerId))
{
var defaultRole = context.Roles.FirstOrDefault(n => n.Name == role);
if (defaultRole != null)
{
var urb = new UserRoleBridge
{
RoleId = defaultRole.Id,
UserId = userId
};
context.UserRoleBridges.Add(urb);
await context.SaveChangesAsync();
}
}
}
edit
Essentially, when I run the main function, I want a series of method calls to all fire off somewhat simultaneously and handle everything in their own threads so that I don't have to worry about it. Here is a pseudo-code example.
public async void RunAllAsync() {
taskA(*some value*);
taskA(*some value*);
taskB(*some value*);
taskB(*some value*);
await AllTasksCompleted
}
public async Task taskA(int item){
//do something with item
}
public async Task taskB(int item) {
subTaskB(*some value*)
subTaskB(*some value*)
}
public async Task subTaskB(int item) {
// do something
}
In the above example, when #RunAllAsync is called, every function call it makes (and the function calls they make) are fired off simultaneously. When all of these calls are completed, whatever method called #RunAllAsync would continue to execute.
If you're not using await, the async keyword doesn't really do anything useful and you can leave it off. You can await a method returning a Task regardless of whether it's marked as async or not.
public Task DoSomethingAsync()
{
Task someTaskJustLikeANormalReturnValue = Task.Delay(1000);
return someTaskJustLikeANormalReturnValue;
}
// later...
public async Task SomeOtherFunction()
{
// You can await DoSomethingAsync just like any async method, because
// you're really awaiting the Task which got returned.
await DoSomethingAsync();
}
In your case I would probably collect the tasks and await them all together:
public async Task SetupAccessControl(int objectTypeId, int objectId, int? organizationId)
{
var tasks = new List<Task>();
using (var context = new SupportContext(CustomerId))
{
... // omitted for brevity
if (objectTypeId == (int) ObjectType.User)
{
tasks.Add(AddToUserRoleBridge("Everyone", objectId));
tasks.Add(AddToUserRoleBridge("Default", objectId));
}
... // omitted for brevity
}
await Task.WhenAll(tasks.ToArray());
}
This allows you to pass the decision up to the caller of whether to await on the subtasks or not. This will also allow the any caller to unwrap any exceptions if they happen in AddToUserRoleBridge.
You can do it two ways:
Change the signature of AddToUserRoleBridge to return void will
remove the warning (but nothing can await it then).
Store the result of the call. var ignoreme=AddToUserRoleBridge(...) will remove the warning as well.
I am implementing an interface
Task Something()
but none of the code in the implementation is async
Task Something(){
var x=1;
return null;
}
This of course results in an error when called:
await Something()
How do I send back a blank Task to make the Interface happy?
You can use Task.FromResult to return a completed Task.
Task Something()
{
var x=1;
// Do *something* here ;)
return Task.FromResult(true);
}
Note that this will run synchronously, and not be an asynchronous operation, so it'd be good to document that appropriately as well.
If async is expected, why not make it async?
Task Something()
{
return Task.Run( () => { /* your code here */ } );
}
In order to await any method, it needs to have an async signature so in order to await Something() it should be made like:
async Task Something()
{
//do stuff
}
Just remember that when you want to return a value you should return a Task and if you want to return void you should return only a Task
I am trying to block RequestHandler.ParseAll() with await ConsumerTask;, but when i set a breakpoint there, i ALWAYS get the "Done..." output first... and then Parse2() fails with a NullReferenceException. (thats my guess: "the GC starts cleaning up because _handler got out of scope")
Anyway, I can't figure out why that happens.
class MainClass
{
public async void DoWork()
{
RequestHandler _handler = new RequestHandler();
string[] mUrls;
/* fill mUrls here with values */
await Task.Run(() => _handler.ParseSpecific(mUrls));
Console.WriteLine("Done...");
}
}
static class Parser
{
public static async Task<IEnumerable<string>> QueryWebPage(string url) { /*Query the url*/ }
public static async Task Parse1(Query query)
{
Parallel.ForEach(/*Process data here*/);
}
public static async Task Parse2(Query query)
{
foreach(string line in query.WebPage)
/* Here i get a NullReference exception because query.WebPage == null */
}
}
sealed class RequestHandler
{
private BlockingCollection<Query> Queue;
private Task ConsumerTask = Task.Run(() => /* call consume() for each elem in the queue*/);
private async void Consume(Query obj)
{
await (obj.BoolField ? Parser.Parse1(obj) : Parser.Parse2(obj));
}
public async void ParseSpecific(string[] urls)
{
foreach(string v in urls)
Queue.Add(new Query(await QueryWebPage(v), BoolField: false));
Queue.CompleteAdding();
await ConsumerTask;
await ParseAll(true);
}
private async Task ParseAll(bool onlySome)
{
ReInit();
Parallel.ForEach(mCollection, v => Queue.Add(new Query(url, BoolField:false)));
Queue.CompleteAdding();
await ConsumerTask;
/* Process stuff further */
}
}
struct Query
{
public readonly string[] WebPage;
public readonly bool BoolField;
public Query(uint e, IEnumerable<string> page, bool b) : this()
{
Webpage = page.ToArray();
BoolField = b;
}
}
CodesInChaos has spotted the problem in comments. It stems from having async methods returning void, which you should almost never do - it means you've got no way to track them.
Instead, if your async methods don't have any actual value to return, you should just make them return Task.
What's happening is that ParseSpecific is only running synchronously until the first await QueryWebPage(v) that doesn't complete immediately. It's then returning... so the task started here:
await Task.Run(() => _handler.ParseSpecific(mUrls));
... completes immediately, and "Done" gets printed.
Once you've made all your async methods return Task, you can await them. You also won't need Task.Run at all. So you'd have:
public async void DoWork()
{
RequestHandler _handler = new RequestHandler();
string[] mUrls;
await _handler.ParseSpecific(mUrls);
Console.WriteLine("Done...");
}
...
public async TaskParseSpecific(string[] urls)
{
foreach(string v in urls)
{
// Refactored for readability, although I'm not sure it really
// makes sense now that it's clearer! Are you sure this is what
// you want?
var page = await QueryWebPage(v);
Queue.Add(new Query(page, false);
}
Queue.CompleteAdding();
await ConsumerTask;
await ParseAll(true);
}
Your Reinit method also needs changing, as currently the ConsumerTask will basically complete almost immediately, as Consume will return immediately as it's another async method returning void.
To be honest, what you've got looks very complex, without a proper understanding of async/await. I would read up more on async/await and then probably start from scratch. I strongly suspect you can make this much, much simpler. You might also want to read up on TPL Dataflow which is designed to make producer/consumer scenarios simpler.
I'm trying to get the hand of the new async CTP stuff and I'm probably confusing myself here..
I can have this "task method", with no problem:
public static Task<String> LongTaskAAsync() {
return Task.Run(() => {
return("AAA");
});
}
But what if I need the task to execute another task, can I mark it as "async" and use "await"? I tried this:
public async static Task<String> LongTaskAAsync() {
await Task.Delay(2000);
return Task.Run(() => {
return("AAA");
});
}
But then mysteriously get this compiler error: Since this is an async method, the return expression must be of type 'string' rather than Task<string>
What am I missing here?
You may want to read my async/await intro post.
Return values from async methods are wrapped in a Task<TResult>. Likewise, await unwraps those return values:
public static async Task<String> LongTaskAAsync() {
await Task.Delay(2000);
return await Task.Run(() => {
return("AAA");
});
}
The reasoning behind this is described in my Async "Why Do the Keywords Work That Way" Unofficial FAQ.
P.S. You can also use Task.FromResult for simple tests like this.
Edit: If you want to create and return the Task object itself, then the method should not be async. One somewhat common pattern is to have a public non-async method that calls the async portion only if necessary.
For example, some kind of asynchronous cache - if the object is in the cache, then return it immediately; otherwise, asynchronously create it, add it to the cache, and return it (this is example code - not thread-safe):
public static Task<MyClass> GetAsync(int key)
{
if (cache.Contains(key))
return Task.FromResult(cache[key]);
return CreateAndAddAsync(key);
}
private static async Task<MyClass> CreateAndAddAsync(int key)
{
var result = await CreateAsync(key);
cache.Add(key, result);
return result;
}
Can a “task method” also be an “async” method?
Yes it can be, by simply changing the method signature to public async static Task<Task<String>> LongTaskAAsync() since that is, what it will return.
If you use the async keyword, the runtime will wrap the type you return into a task, to enable asynchronousness. Say if you return a string, the runtime will wrap that into a Task<string>. int will go Task<int> and Task<string> will go Task<Task<string>>. See this console app to clearify:
public class Program
{
public static void Main(string[] args)
{
// start the main procedure asynchron
Task.Run(() => DoIt()).Wait();
}
// for async support since the static main method can't be async
public static async void DoIt()
{
Program p = new Program();
// use the methods
string s = await p.GetString();
int i = await p.GetInt();
Task<string> tsk = await p.GetTaskOfString();
// just to prove the task works:
// C# 5
string resultFromReturnedTask = await tsk;
// C# 4
string resultFromReturnedTask2 = tsk.Result;
}
public async Task<string> GetString()
{
return "string";
}
public async Task<int> GetInt()
{
return 6;
}
public async Task<Task<string>> GetTaskOfString()
{
return Task.Run(() => "string");
}
}