summaryrefslogtreecommitdiffstats
path: root/test/integration/routing/timelog_test.rb
blob: 4ba57709f3a34f3a9d812688975665746b9834fd (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
# frozen_string_literal: true

# Redmine - project management software
# Copyright (C) 2006-  Jean-Philippe Lang
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.

require_relative '../../test_helper'

class RoutingTimelogsTest < Redmine::RoutingTest
  def test_timelogs_global
    should_route 'GET /time_entries' => 'timelog#index'
    should_route 'GET /time_entries.csv' => 'timelog#index', :format => 'csv'
    should_route 'GET /time_entries.atom' => 'timelog#index', :format => 'atom'
    should_route 'GET /time_entries/new' => 'timelog#new'
    should_route 'POST /time_entries/new' => 'timelog#new'
    should_route 'POST /time_entries' => 'timelog#create'

    should_route 'GET /time_entries/22/edit' => 'timelog#edit', :id => '22'
    should_route 'PATCH /time_entries/22/edit' => 'timelog#edit', :id => '22'
    should_route 'PATCH /time_entries/22' => 'timelog#update', :id => '22'
    should_route 'DELETE /time_entries/22' => 'timelog#destroy', :id => '22'
  end

  def test_timelogs_scoped_under_project
    should_route 'GET /projects/foo/time_entries' => 'timelog#index', :project_id => 'foo'
    should_route 'GET /projects/foo/time_entries.csv' => 'timelog#index', :project_id => 'foo', :format => 'csv'
    should_route 'GET /projects/foo/time_entries.atom' => 'timelog#index', :project_id => 'foo', :format => 'atom'
    should_route 'GET /projects/foo/time_entries/new' => 'timelog#new', :project_id => 'foo'
    should_route 'POST /projects/foo/time_entries' => 'timelog#create', :project_id => 'foo'
  end

  def test_timelogs_scoped_under_issues
    should_route 'GET  /issues/234/time_entries/new' => 'timelog#new', :issue_id => '234'
    should_route 'POST /issues/234/time_entries' => 'timelog#create', :issue_id => '234'
  end

  def test_timelogs_report
    should_route 'GET /time_entries/report' => 'timelog#report'
    should_route 'GET /time_entries/report.csv' => 'timelog#report', :format => 'csv'

    should_route 'GET /projects/foo/time_entries/report' => 'timelog#report', :project_id => 'foo'
    should_route 'GET /projects/foo/time_entries/report.csv' => 'timelog#report', :project_id => 'foo', :format => 'csv'
  end

  def test_timelogs_bulk_edit
    should_route 'GET /time_entries/bulk_edit' => 'timelog#bulk_edit'
    should_route 'POST /time_entries/bulk_edit' => 'timelog#bulk_edit'
    should_route 'POST /time_entries/bulk_update' => 'timelog#bulk_update'
    should_route 'DELETE /time_entries/destroy' => 'timelog#destroy'
  end
end
/a> 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651
---
title: Creating Secure Vaadin Applications Using JEE6
order: 49
layout: page
---

[[creating-secure-vaadin-applications-using-jee6]]
= Creating secure Vaadin applications using JEE6

by Petter Holmström

[[introduction]]
Introduction
~~~~~~~~~~~~

In this article, we are going to look at how the security features of
JEE6 and GlassFish 3 can be used to create a secure Vaadin application.
You should be familiar with the security features of JEE6. If
not, skim through Part VII of the
http://docs.sun.com/app/docs/doc/820-7627[The Java EE 6 Tutorial, Volume
I] before continuing.

[[architecture]]
Architecture
^^^^^^^^^^^^

The example system we are going to discuss has the following
architecture:

image:img/architecture.png[System architecture]

It is a typical JEE web application consisting of several layers. At the
top, there is the client layer, i.e. the users' web browsers. Then comes
the Internet (or any other network for that matter) through which the
data travels between the client layer and the application server. On the
server side, there are the web (or presentation) layer that contains the
user interface - in this case our Vaadin application - and the
enterprise (or application) layer that contains all our business logic
in the form of Enterprise Java Beans (EJB). Finally, the domain layer
contains the system data in the form of persistent entity objects stored
in some relational database - in this case JavaDB - by some Object
Relational Mapping (ORM) solution - in this case EclipseLink 2.0.

In this article, we are going to secure all the layers, with the
exception of the client layer. This means that once the data has reached
the client it may be stored unencrypted in the browser's memory or
downloaded to a disk, but that is a risk we are willing to take in this
case ;-).

[[getting-the-example-source-code]]
Getting the Example Source Code
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

This article is based around an example application that maintains
employee records. Please note that the example system bears little
resemblance to a similar real-world system in order to keep it simple
and easy to understand.

The source code can be downloaded
https://github.com/eriklumme/doc-attachments/blob/master/attachments/SecureVaadinApplicationDemo.zip[here]
and it has been packaged as a NetBeans project. If you do not have
NetBeans, go to http://www.netbeans.org[the NetBeans web site] and
download the latest version (6.8 at the time of writing). Remember to
select the version that includes GlassFish v3.

Once NetBeans and GlassFish are installed, you can unzip the source code
archive and open it in NetBeans. The opened project should look
something like this:

image:img/nbscrshot1.png[Netbeans screenshot 1]

However, before you can try out the application, you have to create a
new JavaDB for the test data. This can be done from the Services tab
inside NetBeans:

image:img/nbscrshot2.png[Netbeans screenshot 2]

When this is done, you should update the _setup/sun-resources.xml_ file
accordingly, so that the correct username/password is used to connect to
the database and the correct JNDI-resources are registered when the
application is deployed.

[[securing-the-domain-layer]]
Securing the Domain Layer
~~~~~~~~~~~~~~~~~~~~~~~~~

In this section we are not going to cover data encryption or Access
Control Lists (ACL), so if that is what you need, you unfortunately have
to look elsewhere for the time being. Instead, we are going to point out
some things that need to be taken into account when designing the domain
model.

In a simple application where all users have read access to all data and
some users have write access, the domain model can be designed basically
in any way possible. However, if there are groups of users that should
be given read access to only some parts of the data, things get a little
more complicated.

In the case of the example system written for this article, the system
contains information that is not intended for everyone such as salaries,
competences or individual career development plans (not included in the
example code). The system will be used by different types of users
(roles):

* A payroll assistant will need read access to the employees' salaries
in order to be able to calculate the paychecks. Write access should be
prohibited.
* A project manager will need read access to the employees' competences
in order to be able to select the right people to his or her team.
However, he or she has nothing to do with how much each employee is
getting paid.
* A director will need full access to the system in order to add new
employees, change salaries, etc.

To keep our lives simple, we should design the domain model in such a
way that if a part of an object graph is readable by a certain role,
then the entire graph should also be readable by the role:

image:img/domain.png[Domain model]

Here, all roles have read access to the `Employee` domain class. Please
note that no associations point out from this class. Therefore, if we
get an instance of `Employee`, we cannot accidentally (or intentionally)
access restricted information by traversing through the object graph.

If we take a look at the `SalaryInfo` domain class, we note that it has
a unidirectional association to the `Employee` class. Thus, if we get an
instance of `SalaryInfo`, we can also access all the information in the
corresponding `Employee`. However, this is perfectly valid as the
Payroll Assistant role has read access to both domain classes.

Now it is time to move on to the enterprise layer, where the security
constraints will actually be enforced.

[[securing-the-enterprise-layer]]
Securing the Enterprise Layer
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Securing the enterprise layer is actually the easiest part, as this is
no different from securing EJBs in an ordinary JEE application. In this
example, role-based security is sufficient so we can use annotations to
secure the EJBs:

[source,java]
....
@Stateless
@TransactionManagement
public class EmployeeBean {
  // Implementations omitted

  @PersistenceContext
  private EntityManager entityManager;

  @RolesAllowed(UserRoles.ROLE_DIRECTOR)
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  public void insertEmployee(Employee employee) {
    ...
  }

  @RolesAllowed(UserRoles.ROLE_DIRECTOR)
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  public Employee updateEmployee(Employee employee) {
    ...
  }

  @RolesAllowed(UserRoles.ROLE_DIRECTOR)
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  public void deleteEmployee(Employee employee) {
    ...
  }

  @RolesAllowed({UserRoles.ROLE_DIRECTOR,
      UserRoles.ROLE_PAYROLL_ASSISTANT,
      UserRoles.ROLE_PROJECT_MANAGER})
  public Employee getEmployeeByPersonNumber(String personNumber) {
    ...
  }

  @RolesAllowed({UserRoles.ROLE_DIRECTOR, UserRoles.ROLE_PAYROLL_ASSISTANT})
  public SalaryInfo getSalaryInfo(Employee employee) {
    ...
  }


  @RolesAllowed({UserRoles.ROLE_DIRECTOR})
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  public SalaryInfo saveSalaryInfo(SalaryInfo salaryInfo) {
    ...
  }

  @RolesAllowed({UserRoles.ROLE_DIRECTOR, UserRoles.ROLE_PROJECT_MANAGER})
  public EmployeeCompetences getCompetences(Employee employee) {
    ...
  }

  @RolesAllowed({UserRoles.ROLE_DIRECTOR, UserRoles.ROLE_PROJECT_MANAGER})
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  public EmployeeCompetences saveCompetences(EmployeeCompetences competences) {
    ...
  }
}
....

The `UserRoles` class is a helper class that defines constants for all
the role names:

[source,java]
....
public class UserRoles {
  public static final String ROLE_DIRECTOR = "DIRECTOR";
  public static final String ROLE_PAYROLL_ASSISTANT = "PAYROLL_ASSISTANT";
  public static final String ROLE_PROJECT_MANAGER = "PROJECT_MANAGER";
}
....

This is actually all there is to it - the container will take care of
the rest. Note, that there are separate lookup methods for basic
employee information and salary information, and that the methods
require different roles. This is how the security constraints discussed
in the previous section are enforced in practice.

[[securing-the-web-layer]]
Securing the Web Layer
~~~~~~~~~~~~~~~~~~~~~~

As all of the application's data and logic should now be protected
inside the enterprise layer, securing the web layer really comes down to
two basic tasks: handling user authentication and disabling the
restricted parts of your user interface. In the example application, the
user interface has not been restricted in order to make it possible to
test the security of the enterprise layer, e.g. what happens when a
restriction actions is attempted.

As the Vaadin application runs entirely on the server, this can be done
inside the application in the same manner as in a Swing desktop
application. However, an (arguably) better approach is to rely on
standard JEE web layer security.

To keep things simple, a Vaadin application should be designed in such a
way that when the application starts, the user is already authenticated
and when the user logs out, the application is closed. In this way, the
JEE container handles the authentication and it is even possible to move
from e.g. form-based authentication to certificate-based authentication
without having to change a single line of code inside the Vaadin
application.

[[the-vaadin-application-servlet]]
The Vaadin Application Servlet
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Here is the code for the application servlet:

[source,java]
....
@WebServlet(urlPatterns={"/ui/*", "/VAADIN/*"})
public class DemoAppServlet extends AbstractApplicationServlet {

  @Inject Instance<DemoApp> application;

  @Override
  protected Class<? extends Application> getApplicationClass() throws
          ClassNotFoundException {
    return DemoApp.class;
  }

  @Override
  protected Application getNewApplication(HttpServletRequest request) throws
          ServletException {
    DemoApp app = application.get();
    Principal principal = request.getUserPrincipal();
    if (principal == null) {
      throw new ServletException("Access denied");
    }

    // In this example, a user can be in one role only
    if (request.isUserInRole(UserRoles.ROLE_DIRECTOR)) {
      app.setUserRole(UserRoles.ROLE_DIRECTOR);
    } else if (request.isUserInRole(UserRoles.ROLE_PAYROLL_ASSISTANT)) {
      app.setUserRole(UserRoles.ROLE_PAYROLL_ASSISTANT);
    } else if (request.isUserInRole(UserRoles.ROLE_PROJECT_MANAGER)) {
      app.setUserRole(UserRoles.ROLE_PROJECT_MANAGER);
    } else {
      throw new ServletException("Access denied");
    }

    app.setUser(principal);
    app.setLogoutURL(request.getContextPath() + "/logout.jsp");
    return app;
  }
}
....

Please note the URL patterns that this servlet handles. The URL for the
Vaadin application will be _$CONTEXT_PATH/ui_. However, the servlet also
has to handle requests to _$CONTEXT_PATH/VAADIN/*_, as the widgetsets
and themes will not load otherwise.

Next, in the `getNewApplication(..)` method, the user principal is
fetched from the request and passed to the Vaadin application using the
`setUser(..)` method (this is not a requirement, but is useful if the
Vaadin application needs to know the identity of the current user). If
the application will act differently depending on the user's roles,
these have to be passed in as well - in this case using a custom setter
defined in the `DemoApp` class. Finally, the logout URL is set to point
to a custom JSP which we will look at in a moment.

[[the-deployment-descriptor]]
The Deployment Descriptor
^^^^^^^^^^^^^^^^^^^^^^^^^

To make sure the user is authenticated when the Vaadin application is
started, all requests to the Vaadin application should require
authentication. In this example we are going to use form-based
authentication using ordinary JSPs for the login, logout and error
screens, but we could just as well use some other form of authentication
such as certificates. In order to achieve this, we add the following to
the `web.xml` deployment descriptor:

[source,xml]
....
<web-app>
  ...
  <security-constraint>
    <display-name>SecureApplicationConstraint</display-name>
    <web-resource-collection>
      <web-resource-name>Vaadin application</web-resource-name>
      <description>The entire Vaadin application is protected</description>
      <url-pattern>/ui/*</url-pattern>
    </web-resource-collection>
    <auth-constraint>
      <description>Only valid users are allowed</description>
      <role-name>DIRECTOR</role-name>
      <role-name>PAYROLL_ASSISTANT</role-name>
      <role-name>PROJECT_MANAGER</role-name>
    </auth-constraint>
  </security-constraint>
  <login-config>
    <auth-method>FORM</auth-method>
    <realm-name>file</realm-name>
    <form-login-config>
      <form-login-page>/login.jsp</form-login-page>
      <form-error-page>/loginError.jsp</form-error-page>
    </form-login-config>
  </login-config>
  <security-role>
    <description/>
    <role-name>DIRECTOR</role-name>
  </security-role>
  <security-role>
    <description/>
    <role-name>PAYROLL_ASSISTANT</role-name>
  </security-role>
  <security-role>
    <description/>
    <role-name>PROJECT_MANAGER</role-name>
  </security-role>
  ...
</web-app>
....

Basically, this file tells the container that this web application:

* uses the roles DIRECTOR, PAYROLL_ASSISTANT and PROJECT_MANAGER,
* requires the user to be in any of these roles when accessing the
Vaadin application,
* requires users to be in the _file_ realm (a built-in realm manageable
from the GlassFish administration console), and
* uses form-based authentication with a JSP for displaying the login
form and another for displaying login errors.

For more information about configuring security for JEE web
applications, please see the JEE6 documentation.

[[the-jsps]]
The JSPs
^^^^^^^^

Now we are going to write the JSPs that will be used for logging users
in and out. These files are well covered in the JEE6 documentation, so
we are just going to list them here without further commenting. First up
is _login.jsp_:

[source,html]
....
<%@page contentType="text/html" pageEncoding="UTF-8"%>
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
    <title>Secure Vaadin Application Demo Login</title>
  </head>
  <body>
    <h1>Please login</h1>
    <form method="post" action="j_security_check">
      <p>
        Username: <input type="text" name="j_username"/>
      </p>
      <p>
        Password: <input type="password" name="j_password"/>
      </p>
      <p>
        <input type="submit" value="Login"/>
      </p>
    </form>
  </body>
</html>
....

Then we move on to _loginError.jsp_:

[source,html]
....
<%@page contentType="text/html" pageEncoding="UTF-8"%>
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
    <title>Secure Vaadin Application Demo Login Failure</title>
  </head>
  <body>
    <h1>Login Failed!</h1>
    <p>
      Please <a href="login.jsp">try again</a>.
    </p>
  </body>
</html>
....

Coming up next is _logout.jsp_:

[source,html]
....
<%@page contentType="text/html" pageEncoding="UTF-8"%>
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
    <title>Secure Vaadin Application Demo</title>
  </head>
  <body>
    <h1>You have been logged out</h1>
    <p>
      <a href="login.jsp">Log in</a> again.
    </p>
  </body>
</html>
<%
  session.invalidate();
%>
....

Please note that this file contains a single line of code at the end
that invalidates the session, effectively logging the user out.

Finally, an _index.jsp_ file is needed in order to make sure that any
requests to the context path are redirected to the Vaadin application:

[source,html]
....
<%
  response.sendRedirect("ui/");
%>
....

There! Now the login and logout mechanisms are in place.

[[securing-the-transport-layer]]
Securing the Transport Layer
~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Even though both the web layer and the enterprise layer are now secured,
the data still has to travel across the Internet to reach the client
layer and, as we know, the Internet is full of people with questionable
intentions. Therefore, we need to make sure that the data reaches its
destination undisclosed and unmodified. In other words, we need SSL.

Provided that the application server has been properly configured to use
SSL (GlassFish v3 should be out of the box, though with a self-signed
certificate), it is very easy to force a web application to use SSL. We
just have to add the following security constraint to the _web.xml_
file:

[source,xml]
....
<security-constraint>
  <display-name>SecureChannelConstraint</display-name>
  <web-resource-collection>
    <web-resource-name>Entire site</web-resource-name>
    <description/>
    <url-pattern>/*</url-pattern>
  </web-resource-collection>
  <user-data-constraint>
    <description>Require encrypted channel</description>
    <transport-guarantee>CONFIDENTIAL</transport-guarantee>
  </user-data-constraint>
</security-constraint>
....

This will force all requests to the application to go over an encrypted
SSL link.

[[configuring-glassfish]]
Configuring GlassFish
~~~~~~~~~~~~~~~~~~~~~

As we are going to let GlassFish handle the user database, we have to do
some additional configuration before the application can be deployed.
Users created using the GlassFish administration console are assigned to
groups, which in turn can be mapped to application roles. It is possible
to configure GlassFish to automatically map a group name to a role with
the same name, but in this case we are going to define the mapping
manually by adding the following definitions to the _sun-web.xml_ file:

[source,xml]
....
<security-role-mapping>
  <role-name>DIRECTOR</role-name>
  <group-name>Directors</group-name>
</security-role-mapping>
<security-role-mapping>
  <role-name>PAYROLL_ASSISTANT</role-name>
  <group-name>Payroll Assistants</group-name>
</security-role-mapping>
<security-role-mapping>
  <role-name>PROJECT_MANAGER</role-name>
  <group-name>Project Managers</group-name>
</security-role-mapping>
....

These definitions tell GlassFish that all users that belong to the
_Directors_ group should hold the `DIRECTOR` role, etc.

The application is now secured. However, in order to try it out we need
to add some users to the _file_ realm using the GlassFish Administration
Console:

image:img/glassfish_console1.png[Glassfish console 1]

image:img/glassfish_console2.png[Glassfish console 2]

Now, we can deploy the application, login with different users and
explore what happens.

[[adding-auditing]]
Adding Auditing
~~~~~~~~~~~~~~~

Although the application is now protected from unauthorized users, it
has not yet been protected from illegal use by authorized users. As the
application deals with sensitive personal information, it should be
possible to see what the users have done with the data while using the
system.

GlassFish has an auditing system that, when turned on, automatically
records access decisions (such as successful or failed logins). However,
in this case we need some more fine-grained auditing. One way of
accomplishing this is to use CDI and interceptors (go
http://docs.jboss.org/webbeans/reference/1.0.0.PREVIEW1/en-US/html/interceptors.html[here]
for more information).

We begin by defining the annotation that will be used to annotate the
methods that are to be subject to auditing:

[source,java]
....
@InterceptorBinding
@Target({ElementType.METHOD, ElementType.TYPE})
@Retention(RetentionPolicy.RUNTIME)
public @interface AuditLog {
}
....

Next, we implement the actual interceptor:

[source,java]
....
@AuditLog
@Interceptor
public class AuditLogInterceptor {
  @Resource
  SessionContext sessionContext;

  @EJB
  AuditService auditService;

  @AroundInvoke
  public Object recordAuditLogEntry(InvocationContext ctx) throws Exception {
    Object result = ctx.proceed();
    StringBuilder sb = new StringBuilder();
    sb.append(ctx.getMethod().getName());
    sb.append("(");
    for (Object p : ctx.getParameters()) {
      sb.append(p);
      sb.append(",");
    }
    sb.append(")");
    String userName = sessionContext.getCallerPrincipal().getName();
    auditService.recordEntry(userName, sb.toString());
    return result;
  }
}
....

Before we can use the interceptor, we have to activate it by adding the
following to the _beans.xml_ file:

[source,xml]
....
<interceptors>
  <class>demoapp.security.AuditLogInterceptor</class>
</interceptors>
....

Finally, we annotate the enterprise methods that should be subject to
auditing:

[source,java]
....
@Stateless
@TransactionManagement
public class EmployeeBean {
  ...

  @RolesAllowed(UserRoles.ROLE_DIRECTOR)
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  @AuditLog
  public void insertEmployee(Employee employee) {
    ...
  }

  @RolesAllowed(UserRoles.ROLE_DIRECTOR)
  @TransactionAttribute(TransactionAttributeType.REQUIRED)
  @AuditLog
  public Employee updateEmployee(Employee employee) {
    ...
  }
  ...
}
....

There! Now, every time a method annotated with `@AuditLog` is
successfully invoked, it will be recorded together with a timestamp and
the name of the user who invoked it.

[[summary]]
Summary
~~~~~~~

In this article, we have discussed how a typical Vaadin/JEE6 application
can be secured. We have secured the enterprise layer using annotations,
secured the web and channel layers by declaring security constraints in
the deployment descriptor and shown how Vaadin can be used together with
form-based authentication. Finally, we have looked at a way of
implementing auditing using interceptors.