ASP .NET Rest API Repeated Calls to another API - c#

How can I make repetitive calls to a URL by time intervals till I get successful result or timeout?
I upload a file to an API and it sends me a URL to let me check if my file is processed successfully. I want to make it so this url is checked in the server until the pending status is changed. There are successful, failed and pending statuses. I want to keep the user wait until the result is either fail or success.
[HttpGet]
public async Task<ActionResult> Get()
{
...
response = await client.GetAsync(other_api_url);
if(response.IsSuccessStatusCode)
{
string result = response.Content.ReadAsStringAsync().Result;
dynamic output = JsonConvert.DeserializeObject<dynamic>(result);
statusUrl = output.url_to_check_status;
//make a call to statusUrl check if content is ready
//get result and parse it
...
status = output.status;
if(status == "SUCCESS")
{
//good path
return Ok();
}
else
{
//make another call after n seconds to check again
}
}
return NotFound();
}

You can use a loop with Task.Delay, like this:
bool done = false;
while (!done)
{
await Task.Delay(1000);
done = await GetStatusFromServerAsync();
}
Although you might want a timeout:
async Task<bool> CheckForCompletion(int timeoutms)
{
var timer = StopWatch.StartNew();
while (timer.ElapsedMilliseconds < timeoutms)
{
var ok = await GetStatusFromServerAsync();
if (ok) return true;
await Task.Yield(1000);
}
return false;
}

Not that I'd agree to doing this (you should never block your request this way), but I think this should solve your problem:
[HttpGet]
public async Task<ActionResult> Get()
{
response = await client.GetAsync(other_api_url);
if(response.IsSuccessStatusCode)
{
// DO NOT USE `.Result` within async method.
string result = await response.Content.ReadAsStringAsync();
dynamic output = JsonConvert.DeserializeObject<dynamic>(result);
statusUrl = output.url_to_check_status;
bool? result = null;
while(result == null) result = await CheckIfSuccessfulAsync(statusUrl);
if (result) return Ok();
return NotFound();
}
}
// Does not *need* to be a separate method, it's just for better readability...
private async Task<bool?> CheckIfSuccessfulAsync(string statusUrl)
{
//make a call to statusUrl check if content is ready
//get result and parse it
...
status = output.status;
if (status == "SUCCESS") return true;
else if (status == "PENDING") return false;
return null;
}

Related

Trying to understand Async/Await but seem to get deadlock c#

I have an async method, and from within that method I call another Async method.
In the second method I call an API. I know that my API request is correct, so it has something to do with the async/await.
Am I creating a deadlock? If so where? And how to fix it?
public async Task<AmountInvoicedModel> CreatePaymentsAndSendAsEmail(InvoiceRequestModel model, bool calculate)
{
....
await CreateQRCodes("testMsg");
....
}
public async Task CreateQRCodes(string ocrNmbr)
{
HttpClient client = new HttpClient();
client.BaseAddress = new Uri("https://mpc.getswish.net/qrg-swish/");
client.DefaultRequestHeaders.Accept.Clear();
client.DefaultRequestHeaders.Accept.Add(
new MediaTypeWithQualityHeaderValue("application/json"));
ServicePointManager.SecurityProtocol = SecurityProtocolType.Tls12;
var json = new
{
payee = new
{
value = "01234567890",
editable = false
},
amount = new
{
value = 100,
editable = false
},
message = new
{
value = $"{ocrNmbr}",
editable = false
},
format = "jpg",
size = 300
};
try
{
HttpResponseMessage response = await client.PostAsJsonAsync(
"api/v1/prefilled", json);
var result = await response.Content.ReadAsStreamAsync();
}
catch (Exception ex)
{
throw;
}
}
UPDATE: I had to put await on the "CalculateInvoice" method too. So now it doesnt deadlock anymore, it moves on - but without giving me a response
[HttpPost]
[Route("calculateInvoice")]
public async Task<IHttpActionResult> CalculateInvoice([FromBody] InvoiceRequestModel model)
{
model.EmailAddress = AccountHelper.GetLoggedInUsername();
var result = await _paymentHandler.CreatePaymentsAndSendAsEmail(model, true);
if (result == null)
return Conflict();
return Ok(result);
}
I had to put await on the CalculateInvoice method for it to move on.
[HttpPost]
[Route("calculateInvoice")]
public async Task<IHttpActionResult> CalculateInvoice([FromBody] InvoiceRequestModel model)
{
model.EmailAddress = AccountHelper.GetLoggedInUsername();
var result = await _paymentHandler.CreatePaymentsAndSendAsEmail(model, true);
if (result == null)
return Conflict();
return Ok(result);
}

HttpClient in while loop only executed once

I am trying to call HttpClient request inside for loop as follows. It needs to do multiple consecutive calls to third party rest api.
But it only gives me fist service call result while loop exit before getting result from rest of the service call.
private void Search()
{
try
{
var i = 1;
using (var httpClient = new HttpClient())
{
while (i < 5)
{
string url = "https://jsonplaceholder.typicode.com/posts/" + i;
var response = httpClient.GetAsync(url).Result;
string jsonResult = response.Content.ReadAsStringAsync().Result;
Console.WriteLine(jsonResult.ToString());
i++;
}
}
}
catch (Exception ex)
{
Console.WriteLine(ex.ToString());
}
}
When I run with debug points the program gives me all the result. But when I run it without debug points it gives me only the first result.
I tried this with using async, await methods too. It also gives me same result.
As I feel Program needs to wait until the async call returns data.
Please help me to solve this.
EDIT - async way
private async Task<string> SearchNew()
{
try
{
var i = 1;
var res = string.Empty;
using (var httpClient = new HttpClient())
{
while (i < 5)
{
string url = "https://jsonplaceholder.typicode.com/posts/" + i;
var response = httpClient.GetAsync(url).Result;
string jsonResult = await response.Content.ReadAsStringAsync();
res = res + jsonResult + " --- ";
i++;
}
}
return res;
}
catch (Exception ex)
{
return ex.Message;
}
}
Both are giving same result.
There's a few things here that you should be doing. First, move the HttpClient creation outside of your method and make it static. You only need one of them and having multiple can be really bad for stability (see here):
private static HttpClient _client = new HttpClient();
Next, extract the calls to the HttpClient into a single method, something simple like this:
//Please choose a better name than this
private async Task<string> GetData(string url)
{
var response = await _client.GetAsync(url);
return await response.Content.ReadAsStringAsync();
}
And finally, you create a list of tasks and wait for them all to complete asynchronously using Task.WhenAll:
private async Task<string[]> SearchAsync()
{
var i = 1;
var tasks = new List<Task<string>>();
//Create the tasks
while (i < 5)
{
string url = "https://jsonplaceholder.typicode.com/posts/" + i;
tasks.Add(GetData(url));
i++;
}
//Wait for the tasks to complete and return
return await Task.WhenAll(tasks);
}
And to call this method:
var results = await SearchAsync();
foreach (var result in results)
{
Console.WriteLine(result);
}

(Async & await) vs (without Async & await) in Web API

I am new with Async and await using C# Programming. In WebAPI, we have created two API Controllers one with Async and await Programming and other is without that. We have done load testing using JMeter and we have got following results.
Users Sync Async
100 No Errors No Errors
500 No Errors No Errors
750 No Errors Errors - (59.0 %) - 502 Bad Gateway
763 No Errors Errors
764 No Errors Errors
765 Errors - (0.13 %) - 502 Bad Gateway Errors
1000 Errors Errors
Can you any please explain/suggest which approach is best or how can we proceed ?
API Code :
GetPersonalDetailsController - Async and await Used
public async Task<IHttpActionResult> GET([FromUri] RequestQueryListDTO objAPIRequest)
{
DateTime startResponseTime = DateTime.Now;
Response objResponse = null;
string strResponse = string.Empty;
var HeaderType = Request.Content.Headers.ContentType;
ProductBAL objProductBAL = null;
try
{
if (objAPIRequest != null)
{
Task<Response> tskGetProductDetails = Task<Response>.Run(() =>
{
objProductBAL = new ProductBAL();
return objProductBAL.GetProductDetails(objAPIRequest);
//Business Access Layer Logic calling
});
objResponse = await tskGetProductDetails;
}
else
{
objResponse = new Response();
objResponse.ReturnCode = -1;
objResponse.ReturnMessage = "Missing Parameters.";
}
}
catch (Exception ex)
{
\\ Exception Logging
}
finally
{
objProductBAL = null;
}
objResponse.ResponseTime = Math.Round((DateTime.Now - startResponseTime).TotalMilliseconds).ToString();
if (objResponse.ReturnCode == Convert.ToInt32(General.ReturnCode))
{
return Content<Response>(HttpStatusCode.BadRequest, objResponse);
}
else
{
return Ok(objResponse);
}
}
========================================================================
GetPDPController - Without using Async and await
public IHttpActionResult GET([FromUri] RequestQueryListDTO objAPIRequest)
{
DateTime startResponseTime = DateTime.Now;
Response objResponse = null;
string strResponse = string.Empty;
var HeaderType = Request.Content.Headers.ContentType;
try
{
if (objAPIRequest != null)
{
//Business Access Layer Logic calling
}
else
{
objResponse = new Response();
objResponse.ReturnCode = -1;
objResponse.ReturnMessage = "Missing Parameters.";
}
}
catch (Exception ex)
{
// Exception Logging Code
}
finally
{
objProductBAL = null;
}
objResponse.ResponseTime = Math.Round((DateTime.Now - startResponseTime).TotalMilliseconds).ToString();
if (objResponse.ReturnCode == Convert.ToInt32(General.ReturnCode))
{
return Content<Response>(HttpStatusCode.BadRequest, objResponse);
}
else
{
return Ok(objResponse);
}
}
My suggestion is have two methods, one Async and one not. That way you can test more.
GetProductDetails
GetProductDetailsAsync
you would then need to change the signature of the calling method aka GET
public IHttpActionResult GET([FromUri] RequestQueryListDTO objAPIRequest)
{
var objResponse = new Response();
//check the properties of objAPIRequest
if(bad)
{
//add stuff if you want
return Content<Response>(HttpStatusCode.BadRequest, objResponse);
}
//Business Access Layer Logic calling
//-----------------------
ProductBAL objProductBAL = new ProductBAL();
//you need to change this to async
var productDetails = objProductBAL.GetProductDetails(objAPIRequest);
//-----------------------
return Ok(objResponse);
}

How to wait for a callback before returning

I am trying to port a code from .Net to Unity C# and I am stuck on a syntax including a callback.
Basically, I had to replace the .Net 'HttpClient' library by this one, also called 'HttpClient'. But the 'Get' syntax is not the same and uses a Callback. I am quite new to C# and Http queries and don't know how to deal with this syntax to get the expected return.
The original function written in .Net:
internal static JObject GetToBackOffice(string action)
{
var url = backOfficeUrl;
url = url + action;
HttpClient httpClient = new HttpClient();
var response =
httpClient.GetAsync(url
).Result;
if (response.StatusCode == System.Net.HttpStatusCode.OK)
{
var idProcess = Newtonsoft.Json.Linq.JObject.Parse(response.Content.ReadAsStringAsync().Result);
return idProcess;
}
else
return null;
The C# code I am writing for Unity:
internal class Utils
{
internal static JObject GetToBackOffice(string action)
{
var url = backOfficeUrl;
url = url + action;
HttpClient httpClient = new HttpClient();
JObject idProcess = new JObject();
httpClient.GetString(new Uri(url),
(response) =>
{
// Raised when the download completes
if (response.StatusCode == System.Net.HttpStatusCode.OK)
{
idProcess = Newtonsoft.Json.Linq.JObject.Parse(response.Data);
}
else
{
idProcess = null;
}
});
// Here I would like to wait for the response, so that idProcess is filled with the received data before returning
return idProcess;
}
public class Action
{
public bool SendData(string id, string secretKey, FakeData data)
{
var idProcess = Utils.GetToBackOffice(String.Format("Events/{0}/infos",id));
//...I do then something with idProcess
//Currently, when I use idProcess here, it is still empty since the GetString response hasn't been received yet when this line is executed
return true;
}
}
public class EventHubSimulator : MonoBehaviour
{
void Start()
{
//Fill the parameters (I skip the details)
string oId = ...;
string secretKey = ...;
var vh = ...;
Action action = new Action();
action.SendData(oId, secretKey, vh);
}
}
My issue is that after the GetToBackOffice function, my code directly uses 'idProcess' for something else but this object is empty because the response was not received yet. I would like to wait for the response before my function returns.
I hope I was clear enough. I know that similar question have already been posted but couldn't find a solution to my specific issue.
Edit:
Finally I used a coroutine as Nain suggested but couldn't get what I expected the way he said. This way seems to work (event though it might not be a good way to do it).
public class EventHubSimulator : MonoBehaviour
{
void Start()
{
//Fill the parameters (I skip the details)
string oId = ...;
string secretKey = ...;
var vh = ...;
Utils utils = new Utils();
StartCoroutine(utils.SendData(oId, secretKey, vh));
}
}
public class Utils: MonoBehaviour
{
private const string backOfficeUrl = "http://myurl/api/";
public CI.HttpClient.HttpResponseMessage<string> response;
public IEnumerator SendData(string id, string secretKey, FakeData data)
{
response = null;
yield return GetToBackOffice(String.Format("Events/{0}/infos", id)); //Make a Http Get request
//The next lines are executed once the response has been received
//Do something with response
Foo(response);
}
IEnumerator GetToBackOffice(string action)
{
var url = backOfficeUrl;
url = url + action;
//Make a Http Get request
HttpClient httpClient = new HttpClient();
httpClient.GetString(new Uri(url), (r) =>
{
// Raised when the download completes
if (r.StatusCode == System.Net.HttpStatusCode.OK)
{
//Once the response has been received, write it in the global variable
response = r;
Debug.Log("Response received : " + response);
}
else
{
Debug.Log("ERROR =============================================");
Debug.Log(r.ReasonPhrase);
throw new Exception(r.ReasonPhrase);
}
});
//Wait for the response to be received
yield return WaitForResponse();
Debug.Log("GetToBackOffice coroutine end ");
}
IEnumerator WaitForResponse()
{
Debug.Log("WaitForResponse Coroutine started");
//Wait for response to become be assigned
while (response == null)
{
yield return new WaitForSeconds(0.02f);
}
Debug.Log("WaitForResponse Coroutine ended");
}
}
One solution is polling for completion as Nain submitted as an answer. If you don't want polling you can use a TaskCompletionSource. This Q&A dives a bit deeper into the why and how.
Your code can then be written like this:
async Task CallerMethod()
{
JObject result = await GetToBackOffice(...);
// Do something with result
}
internal static Task<JObject> GetToBackOffice(string action)
{
var tsc = new TaskCompletionSource<JObject>();
var url = backOfficeUrl;
url = url + action;
HttpClient httpClient = new HttpClient();
JObject idProcess = new JObject();
httpClient.GetString(new Uri(url),
(response) =>
{
// Raised when the download completes
if (response.StatusCode == System.Net.HttpStatusCode.OK)
{
tsc.SetResult(Newtonsoft.Json.Linq.JObject.Parse(response.Data));
}
else
{
tsc.SetResult(null);
}
});
return tsc.Task;
}
See also the section Async Method Calls a Coroutine And Wait for Completion of this msdn blogpost.
NOTE Tasks, and async/await support is only available as beta functionality in Unity. See also this post.
Write a coroutine like
//Class scope variable is neede to hold Response other wise it will
//be destroied as soon as function is ended
ResponseType response;
IEnumerator WaitForResponce(ResponseType response)
{
this.response = response;
while(this.response.Data == null)
yield return new WaitForSeconds (0.02f);
//do what you want here
}
and call the coroutine
httpClient.GetString(new Uri(url),
(response) =>
{
// Raised when the download completes
if (response.StatusCode == System.Net.HttpStatusCode.OK)
{
//idProcess = Newtonsoft.Json.Linq.JObject.Parse(response.Data);
StartCoroutine(WaitForResponse(response));
}
else
{
idProcess = null;
}
});
If the GetString method is returning a Task object then you can use the Wait() method to let the Task finished processing before a result is returned.
var task = httpClient.GetString({
impl here..
});
task.Wait();
return idProcess;

Task not returning IsCompleted state

Following code has a bug: result doesn't contain any state; IsCompleted, IsCanceled and IsFaulted are always false, but I tested that Task works correctly, where is a problem?
var result = _dataService.SyncPoll(webApiPoll);
if (result.IsCompleted)
{
_logger.Info("Execute sync, poll was completed");
poll.IsSynchronized = true;
poll.ServerStatus = ServerStatus.Active;
ctx.SaveChanges();
}
//////
public Task SyncPoll(PollDto poll)
{
if (!_isAuthorized)
{
return null;
}
var client = new ApiClient(_baseApiUrl, _authToken);
Task result = Task.Run(async () => await client.SyncPollWeb(poll));
return result;
}
///////
public async Task<HttpResponseMessage> SyncPollWeb(PollDto poll)
{
HttpResponseMessage resp;
//System.Diagnostics.Debugger.Launch();
using (var client = GetClient())
{
client.DefaultRequestHeaders.Authorization = new System.Net.Http.Headers.AuthenticationHeaderValue(_authType, _accessToken);
resp = await client.PostAsJsonAsync<PollDto>("api/poll", poll);
}
return resp;
}
what you are doing actually is a fire and forget so you don't have the status with such principle
The issue come from the await
you have to add await in order to have back the status
var result = await _dataService.SyncPoll(webApiPoll);
//and here
Task result = await client.SyncPollWeb(poll);
That's happened because when you checking the state your task is not finished (and running) yet - you just create it. You should wait until any state is deternined for your task. Use
result.Wait();
before checking for result.IsCompleted

Categories