Developing Web Apps with ASP.NET Core 2.0 and React - Part 3

TL;DR: This is the third part of the series about developing a Web application based on ASP.NET Core 2.0 and React. In the first post, you built a secured Web API application that provides a list of books with ASP.NET Core 2.0. In the second post, you created a Single Page Application (SPA) client based on React to consume this API. In this post, you will integrate the client-server system with scopes and authorization management by using Auth0. You can find the final code of both the React client and the ASP.NET server in this GitHub repository.

“Learn how to integrate React and ASP.NET Core apps in this practical tutorial.”

Tweet This

---

API Scopes

In order to enhance your client-server application, you are going to implement a more granular access control to the server's resources. This will guarantee that only some specific users are allowed to make specific actions. In particular, you will add to your application an API to create new books and will restrict the access to this API only to some users.

This access granularity can be obtained by using scopes. Scopes are a feature provided by OAuth2 protocol allowing to limit the access granted to an access token.

Creating Scopes for Your APIs

As a first step, you need to create scopes for your APIs. So, open the Auth0 dashboard, click the APIs section, and select the API created in the first post of the series. Then click the Scopes tab, as shown below:

Here you can define scopes for your APIs by specifying a name and a description. A scope name can be any string but, by convention, it usually takes the form of

action:resource

, where

action

represents the action granted and

resource

represents the type of object upon which the action can be executed.

So, define the

write:books

and

read:books

scopes for your APIs:

Definition of scopes for APIs just declares which actions may be performed on server's resources. As a next step, you need to associate scopes to users in order to grant them the ability to access server's resources.

Installing Authorization Extension

To associate scopes with users, you need to configure permissions and roles. This may be accomplished by using the Auth0 Authorization Extension in Auth0 dashboard. If you've never used this extension, you need to install it by accessing the Extensions section of the dashboard and by filtering the available extensions by authorization keyword, as shown in the following picture:

Then click the Auth0 Autorization box to start the installation process. You will be prompted to choose the storage where you would like to store your data. For this simple application, you will choose the default choice, Webtask Storage:

Once the extension is installed, you will see it listed in the Installed Extensions tab.

Configuring Permissions and Roles

Now that you have installed the Authorization Extension, you can use it to configure permissions and roles. To be clear, these are their definitions:

• Permissions are actions that users can do, such as reading a list of books or adding a new book to the list.
• Roles are collections of permissions. For example, you can define a role for administrators that have the permissions to read the list of books and to add a new book to the list.

So, click the Authorization Extension shown in the Installed Extensions tab and a new window will be opened. Here, select the Permissions section and then click the Create Permission button in the upper right corner. You will be prompted to provide data defining the permission, as shown below:

Here you will provide the name of the permission, a description and the client application the permission should be bound to.

Note: Ensure that the name of a permission is exactly the same as the corresponding scope.

In your case, you will add one permission for

read:books

and one for

write:books

. For both permissions specify the client application associated with your React application. The final configuration should look like the following:

Now you need to define two roles for your application: Admin, having both permissions you've just created, and User, having just the

read:books

permission. By selecting Roles section and then clicking Creating Role button, you are asked to provide a few information about the role you want to create, as you can see in the following picture:

Use this form to create both roles. The final result should be as follows:

Adding Roles to Users

Next, you will need to associate roles to users. Actually, in Users section of the Authorization Extension you will find all the users already created in your Auth0 platform. Here you can assign roles to each user by selecting him, using the Roles tab and clicking the Add role to user button. Now you can check the roles you want to assign the user, as shown in the following picture:

Publishing Permission and Rule

As a final step of Authorization Extension configuration, you need to click your domain drop-down menu in the upper right corner and select the Configuration item:

In the page that appears, make sure that Permissions are enabled and click Publish Rule button:

This operation creates a rule for your tenant that will be executed after each user login. In Auth0 platform, a rule is a JavaScript function that is executed when the user is authenticated. The newly created rule will add user's permissions to the user's profile. You can check the rule's code by using the Dashboard.

Validating Token Scopes

The last step in this configuration process is to create a new rule which ensures that scopes contained in an access token is valid accordingly to the user's permission. It means that when a user is authenticated, you need to add only the user's permissions to his access token. To add this rule, go to Rules section in the Auth0 dashboard, click the Create rule button and select the Empty rule template. Now, give a name to the rule and put the following code into the code window:

function (user, context, callback) {
  if (context.clientName !== 'Bookstore client') {
    return callback(null, user, context);
  }

  const permissions = user.permissions || [];
  const requestedScopes = context.request.body.scope || context.request.query.scope;
  const filteredScopes = requestedScopes.split(' ').filter((x) => (x.indexOf(':') < 0));
  Array.prototype.push.apply(filteredScopes, permissions);
  context.accessToken.scope = filteredScopes.join(' ');

  callback(null, user, context);
}

The first line of the JavaScript function's body checks if the current client is the Bookstore client created in the second article of this series. If not, the rule is not executed. Otherwise, only the user's permissions are assigned to the access token's

scope

property. Once you have put this code, click the Save button.

Note: Keep in mind that the rules are executed in the order they are displayed in the Rules section.

Adding Scope Management to ASP.NET Core 2.0 APIs

After this configuration process on the Auth0 platform side, you need to change the code of the ASP.NET Core 2 Web API application built in part 1 of this series. As a first step, you need to implement a way to add an access control to the Web API and to check if a request has the rights to access the Web API. You can apply the Policy-based Authorization provided by ASP.NET Core.

Following this approach, you need to define an Authorization requirement and an Authorization handler. The former is a collection of data containing the current user's rights. The latter evaluates the Authorization requirement to determine if access is allowed.

Now, you have to define the Authorization requirement. So, add a

HasScopeRequirement.cs

file to the ASP.NET Core project and write the following code inside:

public class HasScopeRequirement : IAuthorizationRequirement
{
    public string Issuer { get; }
    public string Scope { get; }

    public HasScopeRequirement(string scope, string issuer)
    {
        Scope = scope ?? throw new ArgumentNullException(nameof(scope));
        Issuer = issuer ?? throw new ArgumentNullException(nameof(issuer));
    }
}

The

HasScopeRequirement

class inherits from

IAuthorizationRequirement

and its constructor simply assigns the

scope

and the

issuer

passed as parameters to its public properties

Scope

and

Issuer

.

The Authorization handler will be implemented in a

HasScopeHandler.cs

file as shown in the following code:

public class HasScopeHandler : AuthorizationHandler<HasScopeRequirement>
{
    protected override Task HandleRequirementAsync(AuthorizationHandlerContext context,             HasScopeRequirement requirement)
    {
        if (!context.User.HasClaim(c => c.Type == "scope" && c.Issuer == requirement.Issuer))
            return Task.CompletedTask;

        var scopes = context.User
            .FindFirst(c => c.Type == "scope" && c.Issuer == requirement.Issuer)
            .Value.Split(' ');

        if (scopes.Any(s => s == requirement.Scope)) context.Succeed(requirement);

        return Task.CompletedTask;
    }
}

The class

HasScopeHandler

inherits from

AuthorizationHandler

. It overrides the

HandleRequirementAsync()

method by taking the current

context

and authorization

requirement

as parameters.

If the current user principal hasn't a

scope

claim issued by the trusted

Issuer

defined in the authorization requirement, the authorization is denied. Otherwise, the user's scopes are compared with the requirement's scope. Only if al least one user's scope matches the requirement's scope, the authorization is granted.

Now, you need to change the

ConfigurationServices()

method in the

Startup

class in order to make available the newly created Authorization requirement and handler, as in the following:

public void ConfigureServices(IServiceCollection services)
{
    services.AddAuthentication(options =>
    {
        options.DefaultAuthenticateScheme = JwtBearerDefaults.AuthenticationScheme;
        options.DefaultChallengeScheme = JwtBearerDefaults.AuthenticationScheme;

    }).AddJwtBearer(options =>
    {
        options.Authority = Configuration["Auth0:Authority"];
        options.Audience = Configuration["Auth0:Audience"];
    });

    services.AddAuthorization(options =>
    {
        options.AddPolicy("read:books", policy => policy.Requirements.Add(new                             HasScopeRequirement("read:books", Configuration["Auth0:Authority"])));
    });

    services.AddSingleton<IAuthorizationHandler, HasScopeHandler>();
    services.AddMvc();
}

The differences from the previous version of the method concern the addition of the new policy through

services.AddAuthorization()

method and the registration of

HasScopeHandler

as a singleton through

AddSingleton()

.

The final step to setup scope management at the API side is to pass the

read:books

scope to the

Authorize

attribute in order to enable the access control for the Web API:

[Route("api/[controller]")]
public class BooksController : Controller
{
  [HttpGet, Authorize("read:books")]
  public IEnumerable<Book> Get()
  {
    var resultBookList = new Book[] { ... };

    return resultBookList;
  }
}

Testing the Authorization Policy for Read Permission

Now you can test Web API access control by simply running the Single Page Application created in part 2 of this series. If you access the application by using a user with User or Admin role, you will be able to get the list of books provided by the Web API. Otherwise, you will not be able to get the list and will get a page like the following:

If you look at the Developer tools console or some other similar tool in your browser, you will find something like this:

As expected, the Web API refused to provide you the requested data since you have not the required permissions.

You could make the user experience a bit more fluent by managing the

403

HTTP status code on the client side. For example, you could change the

ComponentDidMount

method for the

Home

React component as follows:

  componentDidMount() {
    const accessToken = this.props.auth.getAccessToken();

    fetch("/api/books", {headers: new Headers({
        "Accept": "application/json",
        "Authorization": `Bearer ${accessToken}`
    })})
        .then(response => {
          if (response.ok) {
            return response.json()
          }
          if (response.status === 403) {
            alert("You are not authorized!")
          }
        })
        .then(books => this.setState({bookList: books || []}))
        .catch(error => console.log(error))
  }

In this case, if the status code returned by the server is

403

, the application will simply show an alert saying that the user is not authorized to access that page.

Managing Write Permissions in ASP.NET Core 2.0 APIs

So far, you have only managed the

read:books

permission. Now, you are going to change the ASP.NET Core 2 Web API application to support

write:books

permission.

As a first step, you will add the authorization policy in

Startup

class. You simply need to change the lambda function passed to the

AddAuthorization()

method as in the following:

    services.AddAuthorization(options =>
    {
        options.AddPolicy("read:books", policy => policy.Requirements.Add(
        new HasScopeRequirement("read:books", Configuration["Auth0:Authority"])
      ));

          options.AddPolicy("write:books", policy => policy.Requirements.Add(
        new HasScopeRequirement("write:books", Configuration["Auth0:Authority"])
      ));
    });

Then, you will change the

BooksController

class in order to allow adding new books to the existing list.

First of all, you need to transform the current array of books into a static list, as shown below:

[Route("api/[controller]")]
public class BooksController : Controller
{
    private static List<Book> bookList = new List<Book>() {
    new Book { Author = "Ray Bradbury", Title = "Fahrenheit 451", AgeRestriction = false },
    new Book { Author = "Gabriel García Márquez", Title = "One Hundred years of Solitude" AgeRestriction = false },
        new Book { Author = "George Orwell", Title = "1984", AgeRestriction = false },
        new Book { Author = "Anais Nin", Title = "Delta of Venus", AgeRestriction = true }
    };

    [HttpGet, Authorize("read:books")]
  public IEnumerable<Book> Get()
  {
    return bookList;
    }

    //... etc
}

This is necessary because you need to keep the same list among the various requests and get any changes to it. Of course, in a real world scenario this data should be stored in a database.

Now you can add a new API to add a book to the current book list:

[Route("api/[controller]")]
public class BooksController : Controller
{
  //... bookList definition

  [HttpPost, Authorize("write:books")]
  public void Post([FromBody] Book book)
  {
    bookList.Add(book);
  }

  //... etc
}

As you can see, the new API simply adds to the list of books the book representation received from the body of the HTTP POST request. The important thing is that the API has been marked with the

Authorize

attribute by specifying that the

write:books

permission is required.

Allowing to Add Books from the React Client

On the React client side, you will implement the addition of books.

First, you add in the home page a link that will open a page allowing to submit a new book. This can be done by starting with a small change in the JSX code of the

Home

component, as shown below:

class Home extends React.Component {
  //... constructor and other statements

  render() {
    //... other statements
    return  (
      <div>
        <Link to="/bookForm">Add a book</Link>
        <ul>
          {bookList}
        </ul>
      </div>
    );
  }
}

This simple addition makes the home page looking as follows:

Then you need to define the route used in the

Link

component by creating it in the

App

component. The new route definition will be as follows:

//... other imports
import BookForm from './BookForm';

class App extends Component {
  //... other statements

  render() {
    const logoutButton =  this.createLogoutButton();

    return (
      <div className="App">
        <header className="App-header">
          {logoutButton}
          <h1 className="App-title">My Bookstore</h1>
        </header>
        <Switch>
          <Route exact path="/" render={() => this.renderHome()}/>
          <Route path="/startSession" render={({history}) => this.startSession(history)}/>
          <Route path="/bookForm" render={({history}) => <BookForm history={history} auth={this.authService}/>}/>
        </Switch>
      </div>
    );
  }
}

The

/bookForm

route is mapped to the

BookForm

component along with two props representing the browser history and the authentication service implemented in

AuthService

module.

Then, you will implement the

BookForm

component by creating a file named

BookForm.js

and putting inside it the following code:

import React from 'react';
import './BookForm.css'

class BookForm extends React.Component {
  constructor() {
    super();
    this.state = {
      author: "",
      title: ""
    };
  }

  handleAuthorChange(e) {
    this.setState({author: e.target.value});
  }

  handleTitleChange(e) {
    this.setState({title: e.target.value});
  }

  render() {
    return (
      <div className="formContainer">
        <form onSubmit={(e) => this.handleFormSubmit(e)}>
          <div className="row">
            <label className="col-50" htmlFor="author">Author</label>
            <input type="text" name="author" value={this.state.author}
              onChange={(e)=> this.handleAuthorChange(e)}/>
          </div>
          <div className="row">
            <label className="col-50" htmlFor="title">Title</label>
            <input type="text" name="title" value={this.state.title}
              onChange={(e)=> this.handleTitleChange(e)}/>
          </div>
          <div className="row">
            <input type="submit" value="Submit book"/>
          </div>
        </form>
      </div>
    );
  }

  handleFormSubmit(e) {
    //TO BE IMPLEMENTED
  }
}

export default BookForm;

The markup generated by the

render()

method of the component defines a classic form with two textbox and a submit button. Its appearance is as shown in the following picture:

The styles of the form are defined in the

BookForm.css

file imported by the component. Its content is as follows:

.formContainer {
    border-radius: 5px;
    background-color: #f2f2f2;
    padding: 20px;
    margin-left: 20%;
    margin-right: 20%;
}

.row:after {
    content: "";
    display: table;
    clear: both;
}

.col-50 {
    float: left;
    width: 40%;
    margin-top: 6px;
    text-align: right;
}

input[type=text] {
    width: 50%;
}

Looking at the component's code, you can find that the constructor initializes the state of the component. It is defined as an object with an empty string for the author and the title of a book. You can also find two methods,

handleAuthorChange()

and

handleTitleChange()

, used to apply changes from the form elements to the component's state. The

handleFormSubmit()

method, bound to the

onSubmit

event of the form, is not actually implemented in the code above. You can implement it as follows:

handleFormSubmit(e) {
  e.preventDefault();
  const accessToken = this.props.auth.getAccessToken();

  fetch("/api/books", {
    method: "POST",
    body: JSON.stringify(this.state),
    headers: new Headers({
      "Content-Type": "application/json",
      "Authorization": `Bearer ${accessToken}`
    })
  }).then((response) => {
    if (response.ok) {
      this.props.history.push("/");
    } else {
      alert(response.statusText);
    }
  });
}

As you can see, the first line of the method prevents the default behaviour of form submission. This is needed to avoid the standard page reloading performed by the browser. Then you get the access token by using the

getAccessToken()

method provided by

AuthService

module. Finally, you submit the current component state to the server by using

fetch()

. When a response is received from the server, if its status code is successful, the current route is changed with the root, that is you move back to the

Home

component. Otherwise, the message returned by the server is shown in an alert.

Disabling Book Addition for Not Authorized Users

The implementation you made so far works fine. If you access the application by using an admin user, you will be able to add a new book by clicking the Add a book link, filling the form and submitting it to the server. On the other hand, if you access the application by using a simple user (i.e. a non-admin user) you will be able to click on the Add a button link and fill the book form, but when you will try to submit the form's data to the server you will receive a

403

HTTP status code.

Even if this behaviour is functionally correct, letting the user to access the book form when he is not authorized to send data to the server is very annoying. You should make sure that only admin users access the book form. In other words, only admin users should see the Add a button link.

You can obtain this result by making a few changes to the React application.

The first thing you need to do is getting the scopes returned by the authorization server and store them into the session data. That's mean to add the following line to the

setSession()

method of the

AuthService

class:

setSession(authResult) {
  //... other statements
  localStorage.setItem('scopes', JSON.stringify(authResult.scope || ""));
}

As you can see, you can find the user's scope in the

authResult

object returned by the server after the authentication. In particular, the

scope

property will be persisted as a string with all the other session info.

Now you can add the

hasScopes()

method to the

AuthService

class:

export default class AuthService {
  //... other statements

  hasScopes(scopes) {
    const grantedScopes = JSON.parse(localStorage.getItem('scopes')).split(' ');
    return scopes.every(scope => grantedScopes.includes(scope));
  }
}

This method takes an array of strings representing scopes and returns a boolean value saying if the passed scopes are included in the scopes granted to the user (

true

) or not (

false

).

You will use the

hasScopes()

method inside the

render()

method of the

Home

component as shown by the following snippet of code:

class Home extends React.Component {
  //... other statements

  render() {
    const bookList = this.state.bookList.map((book) => (
      <li><i>{book.author}</i> - <h3>{book.title}</h3></li>
    ));
    const addBookButton = this.props.auth.hasScopes(["write:books"]) ?
      <Link to="/bookForm">Add a book</Link>
      : null;

    return (
      <div>
        {addBookButton}
        <ul>
          {bookList}
        </ul>
      </div>
    );
  }
}

As you can see, the

addBookButton

variable will contain the

<Link>

element when the user has the

write:books

permission. Otherwise, its value will be

null

. This conditional rendering of the Add a book link solves your issue.

“Managing OAuth 2.0 Scopes on ASP.NET Core 2.0 is easy.”

Tweet This

Summary

In this part of the series, you made a few settings on the Auth0 dashboard in order to define and manage scopes and roles for users. Then you changed the Web API application implemented in the first part and the React-based client built in the second part of the series in order to support user's permission previously defined.

You can find the final code and download it from the GitHub repository.