Edit on GitHub

In this document

Problems of AJAX Operations

Performing AJAX calls is frequently used by modern applications. Especially in SPAs (Single Page Applications), it's almost the only way of communicating with the server. An AJAX call consists of some repeating steps:

On the client-side, the JavaScript code should supply an URL, optionally some data, and it should select a method (POST, GET...) to perfom an AJAX call. It must wait and handle the return value. There may be an error (network error generally) while performing a call to the server. Or there may be some error on the server-side and server may send a failed response with an error message. The client-side code must handle these and optionally notify a user (like show an error dialog). If there is no error and the server returns data, the client must also handle it. In addition to this, you generally want to block some part or a whole area of the screen and show a busy indicator until an AJAX operation completes.

Server-code should get the request, perform some server-side code, catch exceptions and return a valid response to the client. If an error occurs, it may optionally send an error message to the client. If it's a validation error, the server may want to the add descriptions to validation problems. In the case of a successful request, it may send return values to the client.

The ASP.NET Boilerplate Way

ASP.NET Boilerplate automates some of these steps by wrapping AJAX calls with the abp.ajax function. An example ajax call:

var newPerson = {
    name: 'Dougles Adams',
    age: 42
};

abp.ajax({
    url: '/People/SavePerson',
    data: JSON.stringify(newPerson)
}).done(function(data) {
    abp.notify.success('Created new person with id = ' + data.personId);
});

abp.ajax gets options as an object. You can pass any valid parameter into jQuery's $.ajax method. There are some defaults here: the dataType is 'json', the type is 'POST', and the contentType is 'application/json' (We're calling JSON.stringify to convert a JavaScript object into a JSON string before sending it to the server). You can override these defaults by passing options to abp.ajax.

abp.ajax returns a promise, so you can write done, fail, then (etc) handlers. In this example, we made a simple AJAX request to the PeopleController's SavePerson action. In the done handler, we fetched the database id for the newly created person and showed a success notification (See notification API). Let's see the MVC controller for this AJAX call:

public class PeopleController : AbpController
{
    [HttpPost]
    public JsonResult SavePerson(SavePersonModel person)
    {
        //TODO: save new person to database and return new person's id
        return Json(new {PersonId = 42});
    }
}

The SavePersonModel contains the Name and Age properties. The SavePerson action is marked with HttpPost, since abp.ajax's default method is POST. We simplified the method implementation by returning an anonymous object.

This seams pretty straightforward, but there are some important things behind the scenes that are handled by ASP.NET Boilerplate. Let's dive into those details...

AJAX Return Messages

When we directly return an object with PersonId = 2, ASP.NET Boilerplate wraps it with an MvcAjaxResponse object. The actual AJAX response is something like this:

{
  "success": true,
  "result": {
    "personId": 42
  },
  "error": null,
  "targetUrl": null,
  "unAuthorizedRequest": false,
  "__abp": true
}

Here all the properties are camelCase (since it's conventional in JavaScript) even if they are PascalCase on the server-side's code. Here's an explanation of all the fields:

  • success: A boolean value (true or false) that indicates the success status of the operation. If this is true, abp.ajax resolves the promise and calls the done handler. If it's false (if there is an exception thrown in the method call), it calls the fail handler and shows the error message using the abp.message.error function.
  • result: The actual return value of the controller action. It's valid if the request was a success and if the server sent a return value.
  • error: If success is false, this field is an object that contains the message and details fields.
  • targetUrl: This provides a way for the server to redirect the client to another url if needed.
  • unAuthorizedRequest: This provides a method for the server to inform the client that this operation is not authorized or the user is not authenticated. abp.ajax reloads the current page if this is true.
  • __abp: A special signature that is returned by an ABP wrapped responses. You don't use this yourself, but abp.ajax handles it.

This return format is recognized and handled by the abp.ajax function. The done handler in abp.ajax gets the actual return value of the controller (An object with a personId property) if there is no error.

Handling Errors

As described above, ASP.NET Boilerplate handles exceptions on the server and returns an object with an error message like this:

{
  "targetUrl": null,
  "result": null,
  "success": false,
  "error": {
    "message": "An internal error occurred during your request!",
    "details": "..."
  },
  "unAuthorizedRequest": false,
  "__abp": true
}

As you can see, success is false and result is null. abp.ajax handles this object and shows an error message to the user using the abp.message.error function. If your server-side code throws an exception type of UserFriendlyException, it directly shows the exception's message to the user. Otherwise, it hides the actual error (writes error to logs) and shows a standard ''An internal error occurred..." message to the user. All these are automatically done by ASP.NET Boilerplate.

You may want to disable displaying the message for a particular AJAX call. If so, add abpHandleError: false into the abp.ajax options.

HTTP Status Codes

ABP returns the following HTTP status codes when exceptions occur:

  • 401 for unauthenticated requests (Used has not logged in and the server action needs authentication).
  • 403 for unauthorized requests.
  • 500 for all other exception types.

WrapResult and DontWrapResult Attributes

You can control the wrapping using WrapResult and the DontWrapResult attributes for an action or all actions in a controller.

ASP.NET MVC Controllers

ASP.NET Boilerplate wraps (as described above) ASP.NET MVC action results by default if the return type is a JsonResult (or Task<JsonResult> for async actions). You can change this by using the WrapResult attribute as shown below:

public class PeopleController : AbpController
{
    [HttpPost]
    [WrapResult(WrapOnSuccess = false, WrapOnError = false)]
    public JsonResult SavePerson(SavePersonModel person)
    {
        //TODO: save new person to database and return new person's id
        return Json(new {PersonId = 42});
    }
}

As a shortcut, we can simply use the [DontWrapResult] attribute which is identical for this example.

You can change this default behaviour from the startup configuration (using Configuration.Modules.AbpMvc()...).

ASP.NET Web API Controllers

ASP.NET Boilerplate does not wrap Web API actions by default if an action has successfully executed. You can add WrapResult to actions or controllers if you need to, but by default it wraps exceptions.

You can change this default behavior from the startup configuration (using Configuration.Modules.AbpWebApi()...).

Dynamic Web API Layer

ASP.NET Boilerplate wraps dynamic web api layer methods by default. You can change this behavior using the WrapResult and DontWrapResult attributes in the interface of your application service.

You can change this default behaviour from the startup configuration (using Configuration.Modules.AbpWebApi()...).

ASP.NET Core Controllers

ABP automatically wraps results for a JsonResult, ObjectResult and any object which does not implement IActionResult.  See the ASP.NET Core documentation for more info.

You can change this default behavior from the startup configuration (using Configuration.Modules.AbpAspNetCore()...).

Dynamic Web API Layer

While ASP.NET Boilerplate provides a mechanism to make AJAX calls easy, in a real-world application it's typical to write a JavaScript function for every AJAX call. For example:

//Create a function to abstract AJAX call
var savePerson = function(person) {
    return abp.ajax({
        url: '/People/SavePerson',
        data: JSON.stringify(person)
    });
};

//Create a new person
var newPerson = {
    name: 'Dougles Adams',
    age: 42
};

//Save the person
savePerson(newPerson).done(function(data) {
    abp.notify.success('created new person with id = ' + data.personId);
});

This is good practice, but time-consuming and tedious, because you have to write a function for every ajax call. ASP.NET can automatically generate these type of functions for application services and controllers.

Read the dynamic web api layer documentation for the Web API and ASP.NET Core documentation for the ASP.NET Core integration.