SeamFramework.orgCommunity Documentation

Seam - Contextual Components

A Framework for Enterprise Java

2.0.2.SP1


Introduction to JBoss Seam
1. Contribute to Seam
1. Seam Tutorial
1.1. Try the examples
1.1.1. Running the examples on JBoss AS
1.1.2. Running the examples on Tomcat
1.1.3. Running the example tests
1.2. Your first Seam application: the registration example
1.2.1. Understanding the code
1.2.2. How it works
1.3. Clickable lists in Seam: the messages example
1.3.1. Understanding the code
1.3.2. How it works
1.4. Seam and jBPM: the todo list example
1.4.1. Understanding the code
1.4.2. How it works
1.5. Seam pageflow: the numberguess example
1.5.1. Understanding the code
1.5.2. How it works
1.6. A complete Seam application: the Hotel Booking example
1.6.1. Introduction
1.6.2. Overview of the booking example
1.6.3. Understanding Seam conversations
1.6.4. The Seam UI control library
1.6.5. The Seam Debug Page
1.7. A complete application featuring Seam and jBPM: the DVD Store example
1.8. An example of Seam with Hibernate: the Hibernate Booking example
1.9. A RESTful Seam application: the Blog example
1.9.1. Using "pull"-style MVC
1.9.2. Bookmarkable search results page
1.9.3. Using "push"-style MVC in a RESTful application
2. Getting started with Seam, using seam-gen
2.1. Before you start
2.2. Setting up a new Eclipse project
2.3. Creating a new action
2.4. Creating a form with an action
2.5. Generating an application from an existing database
2.6. Generating an application from existing JPA/EJB3 entities
2.7. Deploying the application as an EAR
2.8. Seam and incremental hot deployment
2.9. Using Seam with JBoss 4.0
2.9.1. Install JBoss 4.0
2.9.2. Install the JSF 1.2 RI
3. Getting started with Seam, using JBoss Tools
3.1. Before you start
3.2. Setting up a new Seam project
3.3. Creating a new action
3.4. Creating a form with an action
3.5. Generating an application from an existing database
3.6. Seam and incremental hot deployment with JBoss Tools
4. The contextual component model
4.1. Seam contexts
4.1.1. Stateless context
4.1.2. Event context
4.1.3. Page context
4.1.4. Conversation context
4.1.5. Session context
4.1.6. Business process context
4.1.7. Application context
4.1.8. Context variables
4.1.9. Context search priority
4.1.10. Concurrency model
4.2. Seam components
4.2.1. Stateless session beans
4.2.2. Stateful session beans
4.2.3. Entity beans
4.2.4. JavaBeans
4.2.5. Message-driven beans
4.2.6. Interception
4.2.7. Component names
4.2.8. Defining the component scope
4.2.9. Components with multiple roles
4.2.10. Built-in components
4.3. Bijection
4.4. Lifecycle methods
4.5. Conditional installation
4.6. Logging
4.7. The Mutable interface and @ReadOnly
4.8. Factory and manager components
5. Configuring Seam components
5.1. Configuring components via property settings
5.2. Configuring components via components.xml
5.3. Fine-grained configuration files
5.4. Configurable property types
5.5. Using XML Namespaces
6. Events, interceptors and exception handling
6.1. Seam events
6.2. Page actions
6.3. Page parameters
6.3.1. Mapping request parameters to the model
6.4. Propagating request parameters
6.5. Conversion and Validation
6.6. Navigation
6.7. Fine-grained files for definition of navigation, page actions and parameters
6.8. Component-driven events
6.9. Contextual events
6.10. Seam interceptors
6.11. Managing exceptions
6.11.1. Exceptions and transactions
6.11.2. Enabling Seam exception handling
6.11.3. Using annotations for exception handling
6.11.4. Using XML for exception handling
6.11.5. Some common exceptions
7. Conversations and workspace management
7.1. Seam's conversation model
7.2. Nested conversations
7.3. Starting conversations with GET requests
7.4. Using <s:link> and <s:button>
7.5. Success messages
7.6. Natural conversation ids
7.7. Creating a natural conversation
7.8. Redirecting to a natural conversation
7.9. Workspace management
7.9.1. Workspace management and JSF navigation
7.9.2. Workspace management and jPDL pageflow
7.9.3. The conversation switcher
7.9.4. The conversation list
7.9.5. Breadcrumbs
7.10. Conversational components and JSF component bindings
7.11. Concurrent calls to conversational components
7.11.1. RichFaces Ajax
8. Pageflows and business processes
8.1. Pageflow in Seam
8.1.1. The two navigation models
8.1.2. Seam and the back button
8.2. Using jPDL pageflows
8.2.1. Installing pageflows
8.2.2. Starting pageflows
8.2.3. Page nodes and transitions
8.2.4. Controlling the flow
8.2.5. Ending the flow
8.2.6. Pageflow composition
8.3. Business process management in Seam
8.4. Using jPDL business process definitions
8.4.1. Installing process definitions
8.4.2. Initializing actor ids
8.4.3. Initiating a business process
8.4.4. Task assignment
8.4.5. Task lists
8.4.6. Performing a task
9. Seam and Object/Relational Mapping
9.1. Introduction
9.2. Seam managed transactions
9.2.1. Disabling Seam-managed transactions
9.2.2. Configuring a Seam transaction manager
9.2.3. Transaction synchronization
9.3. Seam-managed persistence contexts
9.3.1. Using a Seam-managed persistence context with JPA
9.3.2. Using a Seam-managed Hibernate session
9.3.3. Seam-managed persistence contexts and atomic conversations
9.4. Using the JPA "delegate"
9.5. Using EL in EJB-QL/HQL
9.6. Using Hibernate filters
10. JSF form validation in Seam
11. Groovy integration
11.1. Groovy introduction
11.2. Writing Seam applications in Groovy
11.2.1. Writing Groovy components
11.2.2. seam-gen
11.3. Deployment
11.3.1. Deploying Groovy code
11.3.2. Native .groovy file deployment at development time
11.3.3. seam-gen
12. The Seam Application Framework
12.1. Introduction
12.2. Home objects
12.3. Query objects
12.4. Controller objects
13. Seam and JBoss Rules
13.1. Installing rules
13.2. Using rules from a Seam component
13.3. Using rules from a jBPM process definition
14. Security
14.1. Overview
14.1.1. Which mode is right for my application?
14.2. Requirements
14.3. Disabling Security
14.4. Authentication
14.4.1. Configuration
14.4.2. Writing an authentication method
14.4.3. Writing a login form
14.4.4. Simplified Configuration - Summary
14.4.5. Handling Security Exceptions
14.4.6. Login Redirection
14.4.7. HTTP Authentication
14.4.8. Advanced Authentication Features
14.5. Error Messages
14.6. Authorization
14.6.1. Core concepts
14.6.2. Securing components
14.6.3. Security in the user interface
14.6.4. Securing pages
14.6.5. Securing Entities
14.7. Writing Security Rules
14.7.1. Permissions Overview
14.7.2. Configuring a rules file
14.7.3. Creating a security rules file
14.8. SSL Security
14.9. CAPTCHA
14.9.1. Configuring the CAPTCHA Servlet
14.9.2. Adding a CAPTCHA to a form
14.9.3. Customising the CAPTCHA algorithm
14.10. Security Events
14.11. Run As
14.12. Extending the Identity component
15. Internationalization, localization and themes
15.1. Internationalizing your app
15.1.1. Application server configuration
15.1.2. Translated application strings
15.1.3. Other encoding settings
15.2. Locales
15.3. Labels
15.3.1. Defining labels
15.3.2. Displaying labels
15.3.3. Faces messages
15.4. Timezones
15.5. Themes
15.6. Persisting locale and theme preferences via cookies
16. Seam Text
16.1. Basic fomatting
16.2. Entering code and text with special characters
16.3. Links
16.4. Entering HTML
17. iText PDF generation
17.1. Using PDF Support
17.1.1. Creating a document
17.1.2. Basic Text Elements
17.1.3. Headers and Footers
17.1.4. Chapters and Sections
17.1.5. Lists
17.1.6. Tables
17.1.7. Document Constants
17.1.8. Configuring iText
17.2. Charting
17.3. Bar codes
17.4. Rendering Swing/AWT components
17.5. Further documentation
18. Email
18.1. Creating a message
18.1.1. Attachments
18.1.2. HTML/Text alternative part
18.1.3. Multiple recipients
18.1.4. Multiple messages
18.1.5. Templating
18.1.6. Internationalisation
18.1.7. Other Headers
18.2. Receiving emails
18.3. Configuration
18.3.1. mailSession
18.4. Meldware
18.5. Tags
19. Asynchronicity and messaging
19.1. Asynchronicity
19.1.1. Asynchronous methods
19.1.2. Asynchronous methods with the Quartz Dispatcher
19.1.3. Asynchronous events
19.2. Messaging in Seam
19.2.1. Configuration
19.2.2. Sending messages
19.2.3. Receiving messages using a message-driven bean
19.2.4. Receiving messages in the client
20. Caching
20.1. Using JBossCache in Seam
20.2. Page fragment caching
21. Web Services
21.1. Configuration and Packaging
21.2. Conversational Web Services
21.2.1. A Recommended Strategy
21.3. An example web service
22. Remoting
22.1. Configuration
22.2. The "Seam" object
22.2.1. A Hello World example
22.2.2. Seam.Component
22.2.3. Seam.Remoting
22.3. Evaluating EL Expressions
22.4. Client Interfaces
22.5. The Context
22.5.1. Setting and reading the Conversation ID
22.5.2. Remote calls within the current conversation scope
22.6. Batch Requests
22.7. Working with Data types
22.7.1. Primitives / Basic Types
22.7.2. JavaBeans
22.7.3. Dates and Times
22.7.4. Enums
22.7.5. Collections
22.8. Debugging
22.9. The Loading Message
22.9.1. Changing the message
22.9.2. Hiding the loading message
22.9.3. A Custom Loading Indicator
22.10. Controlling what data is returned
22.10.1. Constraining normal fields
22.10.2. Constraining Maps and Collections
22.10.3. Constraining objects of a specific type
22.10.4. Combining Constraints
22.11. JMS Messaging
22.11.1. Configuration
22.11.2. Subscribing to a JMS Topic
22.11.3. Unsubscribing from a Topic
22.11.4. Tuning the Polling Process
23. Seam and the Google Web Toolkit
23.1. Configuration
23.2. Preparing your component
23.3. Hooking up a GWT widget to the Seam component
23.4. GWT Ant Targets
24. Spring Framework integration
24.1. Injecting Seam components into Spring beans
24.2. Injecting Spring beans into Seam components
24.3. Making a Spring bean into a Seam component
24.4. Seam-scoped Spring beans
24.5. Using Spring PlatformTransactionManagement
24.6. Using a Seam Managed Persistence Context in Spring
24.7. Using a Seam Managed Hibernate Session in Spring
24.8. Spring Application Context as a Seam Component
24.9. Using a Spring TaskExecutor for @Asynchronous
25. Hibernate Search
25.1. Introduction
25.2. Configuration
25.3. Usage
26. Configuring Seam and packaging Seam applications
26.1. Basic Seam configuration
26.1.1. Integrating Seam with JSF and your servlet container
26.1.2. Using facelets
26.1.3. Seam Resource Servlet
26.1.4. Seam servlet filters
26.1.5. Integrating Seam with your EJB container
26.1.6. Don't forget!
26.2. Using Alternate JPA Providers
26.3. Configuring Seam in Java EE 5
26.3.1. Packaging
26.4. Configuring Seam in J2EE
26.4.1. Boostrapping Hibernate in Seam
26.4.2. Boostrapping JPA in Seam
26.4.3. Packaging
26.5. Configuring Seam in Java SE, without JBoss Embedded
26.6. Configuring Seam in Java SE, with JBoss Embedded
26.6.1. Installing Embedded JBoss
26.6.2. Packaging
26.7. Configuring jBPM in Seam
26.7.1. Packaging
26.8. Configuring SFSB and Session Timeouts in JBoss AS
26.9. Running Seam in a Portlet
27. Seam annotations
27.1. Annotations for component definition
27.2. Annotations for bijection
27.3. Annotations for component lifecycle methods
27.4. Annotations for context demarcation
27.5. Annotations for use with Seam JavaBean components in a J2EE environment
27.6. Annotations for exceptions
27.7. Annotations for Seam Remoting
27.8. Annotations for Seam interceptors
27.9. Annotations for asynchronicity
27.10. Annotations for use with JSF
27.10.1. Annotations for use with dataTable
27.11. Meta-annotations for databinding
27.12. Annotations for packaging
27.13. Annotations for integrating with the servlet container
28. Built-in Seam components
28.1. Context injection components
28.2. Utility components
28.3. Components for internationalization and themes
28.4. Components for controlling conversations
28.5. jBPM-related components
28.6. Security-related components
28.7. JMS-related components
28.8. Mail-related components
28.9. Infrastructural components
28.10. Miscellaneous components
28.11. Special components
29. Seam JSF controls
29.1. Tags
29.1.1. Navigation Controls
29.1.2. Converters and Validators
29.1.3. Formatting
29.1.4. Seam Text
29.1.5. Dropdowns
29.1.6. Other
29.2. Annotations
30. JBoss EL
30.1. Parameterized Expressions
30.1.1. Usage
30.1.2. Limitations and Hints
30.2. Projection
31. Testing Seam applications
31.1. Unit testing Seam components
31.2. Integration testing Seam components
31.2.1. Using mocks in integration tests
31.3. Integration testing Seam application user interactions
31.3.1. Configuration
31.3.2. Using SeamTest with another test framework
31.3.3. Integration Testing with Mock Data
31.3.4. Integration Testing Seam Mail
32. Seam tools
32.1. jBPM designer and viewer
32.1.1. Business process designer
32.1.2. Pageflow viewer
33. Seam on OC4J
33.1. Installation and operation of OC4J
33.2. The jee5/booking example
33.2.1. Booking Example Dependencies
33.2.2. Configuration file changes
33.2.3. Building the jee5/booking example
33.3. Deploying the Seam application to OC4J
33.4. Deploying an application created using seam-gen to OC4J
33.4.1. Generating a basic seam-gen application
33.4.2. Changes needed for deployment to OC4J
33.4.3. Building and deploying the seam-gen'd application to OC4J
33.4.4. Extending example with reverse engineered CRUD and Drools
33.5. Finishing up
34. Seam on BEA's Weblogic
34.1. Installation and operation of Weblogic
34.1.1. Installing 10.0.MP1
34.1.2. Creating your Weblogic domain
34.1.3. How to Start/Stop/Access your domain
34.2. The jee5/booking Example
34.2.1. EJB3 Issues with Weblogic
34.2.2. Getting the jee5/booking Working
34.3. The jpa booking example
34.3.1. Building and deploying jpa booking example
34.3.2. What's different with Weblogic 10.x
34.4. Deploying an application created using seam-gen on Weblogic 10.x
34.4.1. Running seam-gen setup
34.4.2. What to change for Weblogic 10.X
34.4.3. Building and Deploying your application
35. Seam on IBM's Websphere
35.1. Websphere environment and deployment information
35.1.1. Installation versions and tips
35.1.2. Required custom properties
35.2. The jee5/booking example
35.2.1. Configuration file changes
35.2.2. Building the jee5/booking example
35.2.3. Deploying the application to Websphere
35.3. The jpa booking example
35.3.1. Building the jpa example
35.3.2. Deploying the jpa example
35.3.3. Whats different for Websphere 6.1
35.4. Deploying an application created using seam-gen on Websphere 6.1.0.13
35.4.1. Running seam-gen Setup
35.4.2. Changes needed for deployment to Websphere
36. Dependencies
36.1. Project Dependencies
36.1.1. Core
36.1.2. RichFaces
36.1.3. Seam Mail
36.1.4. Seam PDF
36.1.5. JBoss Rules
36.1.6. JBPM
36.1.7. GWT
36.1.8. Spring
36.1.9. Groovy
36.2. Dependency Management using Maven

Seam is an application framework for Enterprise Java. It is inspired by the following principles:

One kind of "stuff"

Seam defines a uniform component model for all business logic in your application. A Seam component may be stateful, with the state associated with any one of several well-defined contexts, including the long-running, persistent, business process context and the conversation context, which is preserved across multiple web requests in a user interaction.

There is no distinction between presentation tier components and business logic components in Seam. You can layer your application according to whatever architecture you devise, rather than being forced to shoehorn your application logic into an unnatural layering scheme forced upon you by whatever combination of stovepipe frameworks you're using today.

Unlike plain Java EE or J2EE components, Seam components may simultaneously access state associated with the web request and state held in transactional resources (without the need to propagate web request state manually via method parameters). You might object that the application layering imposed upon you by the old J2EE platform was a Good Thing. Well, nothing stops you creating an equivalent layered architecture using Seam—the difference is that you get to architect your own application and decide what the layers are and how they work together.

Integrate JSF with EJB 3.0

JSF and EJB 3.0 are two of the best new features of Java EE 5. EJB3 is a brand new component model for server side business and persistence logic. Meanwhile, JSF is a great component model for the presentation tier. Unfortunately, neither component model is able to solve all problems in computing by itself. Indeed, JSF and EJB3 work best used together. But the Java EE 5 specification provides no standard way to integrate the two component models. Fortunately, the creators of both models foresaw this situation and provided standard extension points to allow extension and integration with other frameworks.

Seam unifies the component models of JSF and EJB3, eliminating glue code, and letting the developer think about the business problem.

It is possible to write Seam applications where "everything" is an EJB. This may come as a surprise if you're used to thinking of EJBs as coarse-grained, so-called "heavyweight" objects. However, version 3.0 has completely changed the nature of EJB from the point of view of the developer. An EJB is a fine-grained object—nothing more complex than an annotated JavaBean. Seam even encourages you to use session beans as JSF action listeners!

On the other hand, if you prefer not to adopt EJB 3.0 at this time, you don't have to. Virtually any Java class may be a Seam component, and Seam provides all the functionality that you expect from a "lightweight" container, and more, for any component, EJB or otherwise.

Integrated AJAX

Seam supports the best open source JSF-based AJAX solutions: JBoss RichFaces and ICEfaces. These solutions let you add AJAX capability to your user interface without the need to write any JavaScript code.

Alternatively, Seam provides a built-in JavaScript remoting layer that lets you call components asynchronously from client-side JavaScript without the need for an intermediate action layer. You can even subscribe to server-side JMS topics and receive messages via AJAX push.

Neither of these approaches would work well, were it not for Seam's built-in concurrency and state management, which ensures that many concurrent fine-grained, asynchronous AJAX requests are handled safely and efficiently on the server side.

Business process as a first class construct

Optionally, Seam provides transparent business process management via jBPM. You won't believe how easy it is to implement complex workflows, collaboration and and task management using jBPM and Seam.

Seam even allows you to define presentation tier pageflow using the same language (jPDL) that jBPM uses for business process definition.

JSF provides an incredibly rich event model for the presentation tier. Seam enhances this model by exposing jBPM's business process related events via exactly the same event handling mechanism, providing a uniform event model for Seam's uniform component model.

Declarative state management

We're all used to the concept of declarative transaction management and declarative security from the early days of EJB. EJB 3.0 even introduces declarative persistence context management. These are three examples of a broader problem of managing state that is associated with a particular context, while ensuring that all needed cleanup occurs when the context ends. Seam takes the concept of declarative state management much further and applies it to application state. Traditionally, J2EE applications implement state management manually, by getting and setting servlet session and request attributes. This approach to state management is the source of many bugs and memory leaks when applications fail to clean up session attributes, or when session data associated with different workflows collides in a multi-window application. Seam has the potential to almost entirely eliminate this class of bugs.

Declarative application state management is made possible by the richness of the context model defined by Seam. Seam extends the context model defined by the servlet spec—request, session, application—with two new contexts—conversation and business process—that are more meaningful from the point of view of the business logic.

You'll be amazed at how many things become easier once you start using conversations. Have you ever suffered pain dealing with lazy association fetching in an ORM solution like Hibernate or JPA? Seam's conversation-scoped persistence contexts mean you'll rarely have to see a LazyInitializationException. Have you ever had problems with the refresh button? The back button? With duplicate form submission? With propagating messages across a post-then-redirect? Seam's conversation management solves these problems without you even needing to really think about them. They're all symptoms of the broken state management architecture that has been prevalent since the earliest days of the web.

Bijection

The notion of Inversion of Control or dependency injection exists in both JSF and EJB3, as well as in numerous so-called "lightweight containers". Most of these containers emphasize injection of components that implement stateless services. Even when injection of stateful components is supported (such as in JSF), it is virtually useless for handling application state because the scope of the stateful component cannot be defined with sufficient flexibility, and because components belonging to wider scopes may not be injected into components belonging to narrower scopes.

Bijection differs from IoC in that it is dynamic, contextual, and bidirectional. You can think of it as a mechanism for aliasing contextual variables (names in the various contexts bound to the current thread) to attributes of the component. Bijection allows auto-assembly of stateful components by the container. It even allows a component to safely and easily manipulate the value of a context variable, just by assigning it to an attribute of the component.

Workspace management and multi-window browsing

Seam applications let the user freely switch between multiple browser tabs, each associated with a different, safely isolated, conversation. Applications may even take advantage of workspace management, allowing the user to switch between conversations (workspaces) in a single browser tab. Seam provides not only correct multi-window operation, but also multi-window-like operation in a single window!

Prefer annotations to XML

Traditionally, the Java community has been in a state of deep confusion about precisely what kinds of meta-information counts as configuration. J2EE and popular "lightweight" containers have provided XML-based deployment descriptors both for things which are truly configurable between different deployments of the system, and for any other kinds or declaration which can not easily be expressed in Java. Java 5 annotations changed all this.

EJB 3.0 embraces annotations and "configuration by exception" as the easiest way to provide information to the container in a declarative form. Unfortunately, JSF is still heavily dependent on verbose XML configuration files. Seam extends the annotations provided by EJB 3.0 with a set of annotations for declarative state management and declarative context demarcation. This lets you eliminate the noisy JSF managed bean declarations and reduce the required XML to just that information which truly belongs in XML (the JSF navigation rules).

Integration testing is easy

Seam components, being plain Java classes, are by nature unit testable. But for complex applications, unit testing alone is insufficient. Integration testing has traditionally been a messy and difficult task for Java web applications. Therefore, Seam provides for testability of Seam applications as a core feature of the framework. You can easily write JUnit or TestNG tests that reproduce a whole interaction with a user, exercising all components of the system apart from the view (the JSP or Facelets page). You can run these tests directly inside your IDE, where Seam will automatically deploy EJB components using JBoss Embedded.

The specs ain't perfect

We think the latest incarnation of Java EE is great. But we know it's never going to be perfect. Where there are holes in the specifications (for example, limitations in the JSF lifecycle for GET requests), Seam fixes them. And the authors of Seam are working with the JCP expert groups to make sure those fixes make their way back into the next revision of the standards.

There's more to a web application than serving HTML pages

Today's web frameworks think too small. They let you get user input off a form and into your Java objects. And then they leave you hanging. A truly complete web application framework should address problems like persistence, concurrency, asynchronicity, state management, security, email, messaging, PDF and chart generation, workflow, wikitext rendering, webservices, caching and more. Once you scratch the surface of Seam, you'll be amazed at how many problems become simpler...

Seam integrates JPA and Hibernate3 for persistence, the EJB Timer Service and Quartz for lightweight asychronicity, jBPM for workflow, JBoss Rules for business rules, Meldware Mail for email, Hibernate Search and Lucene for full text search, JMS for messaging and JBoss Cache for page fragment caching. Seam layers an innovative rule-based security framework over JAAS and JBoss Rules. There's even JSF tag libraries for rendering PDF, outgoing email, charts and wikitext. Seam components may be called synchronously as a Web Service, asynchronously from client-side JavaScript or Google Web Toolkit or, of course, directly from JSF.

Get started now!

Seam works in any Java EE application server, and even works in Tomcat. If your environment supports EJB 3.0, great! If it doesn't, no problem, you can use Seam's built-in transaction management with JPA or Hibernate3 for persistence. Or, you can deploy JBoss Embedded in Tomcat, and get full support for EJB 3.0.

It turns out that the combination of Seam, JSF and EJB3 is the simplest way to write a complex web application in Java. You won't believe how little code is required!

Visit SeamFramework.org to find out how to contribute to Seam!

In this tutorial, we'll assume that you have downloaded JBoss AS 4.2. You should also have a copy of Seam downloaded and extracted to a work directory.

The directory structure of each example in Seam follows this pattern:

First, make sure you have Ant correctly installed, with $ANT_HOME and $JAVA_HOME set correctly. Next, make sure you set the location of your Tomcat 6.0 installation in the build.properties file in the root folder of your Seam installation. You will need to follow the instructions in Section 26.6.1, “Installing Embedded JBoss” for installing JBoss Embedded on Tomcat 6.0. JBoss Embedded is required to run the Seam demo applications on Tomcat. (However, it is possible to use Seam on Tomcat without JBoss Embedded.)

Now, build and deploy the example by typing ant tomcat.deploy in the examples/registration directory.

Finally, start Tomcat.

Try it out by accessing http://localhost:8080/jboss-seam-registration/ with your web browser.

When you deploy the example to Tomcat, any EJB3 components will run inside the JBoss Embeddable EJB3 container, a complete standalone EJB3 container environment.

The registration example is a fairly trivial application that lets a new user store his username, real name and password in the database. The example isn't intended to show off all of the cool functionality of Seam. However, it demonstrates the use of an EJB3 session bean as a JSF action listener, and basic configuration of Seam.

We'll go slowly, since we realize you might not yet be familiar with EJB 3.0.

The start page displays a very basic form with three input fields. Try filling them in and then submitting the form. This will save a user object in the database.

The example is implemented with two JSP pages, one entity bean and one stateless session bean.

Let's take a look at the code, starting from the "bottom".

We need an EJB entity bean for user data. This class defines persistence and validation declaratively, via annotations. It also needs some extra annotations that define the class as a Seam component.

Example 1.1. 

@Entity &lt;co id="registration-entity-annotation"/&gt;

(1)@Name("user")
(2)@Scope(SESSION)
(3)@Table(name="users")
public class User implements Serializable
{
   private static final long serialVersionUID = 1881413500711441951L;
   
(4)   private String username;
   private String password;
   private String name;
   
   public User(String name, String password, String username)
   {
      this.name = name;
      this.password = password;
      this.username = username;
   }
   
(5)   public User() {}
   
(6)   @NotNull @Length(min=5, max=15)
   public String getPassword()
   {
      return password;
   }
   public void setPassword(String password)
   {
      this.password = password;
   }
   
   @NotNull
   public String getName()
   {
      return name;
   }
   public void setName(String name)
   {
      this.name = name;
   }
   
(7)   @Id @NotNull @Length(min=5, max=15)
   public String getUsername()
   {
      return username;
   }
   public void setUsername(String username)
   {
      this.username = username;
   }
}
???

The EJB3 standard @Entity annotation indicates that the User class is an entity bean.

1

A Seam component needs a component name specified by the @Name annotation. This name must be unique within the Seam application. When JSF asks Seam to resolve a context variable with a name that is the same as a Seam component name, and the context variable is currently undefined (null), Seam will instantiate that component, and bind the new instance to the context variable. In this case, Seam will instantiate a User the first time JSF encounters a variable named user.

2

Whenever Seam instantiates a component, it binds the new instance to a context variable in the component's default context. The default context is specified using the @Scope annotation. The User bean is a session scoped component.

3

The EJB standard @Table annotation indicates that the User class is mapped to the users table.

4

name, password and username are the persistent attributes of the entity bean. All of our persistent attributes define accessor methods. These are needed when this component is used by JSF in the render response and update model values phases.

5

An empty constructor is both required by both the EJB specification and by Seam.

6

The @NotNull and @Length annotations are part of the Hibernate Validator framework. Seam integrates Hibernate Validator and lets you use it for data validation (even if you are not using Hibernate for persistence).

7

The EJB standard @Id annotation indicates the primary key attribute of the entity bean.


The most important things to notice in this example are the @Name and @Scope annotations. These annotations establish that this class is a Seam component.

We'll see below that the properties of our User class are bound directly to JSF components and are populated by JSF during the update model values phase. We don't need any tedious glue code to copy data back and forth between the JSP pages and the entity bean domain model.

However, entity beans shouldn't do transaction management or database access. So we can't use this component as a JSF action listener. For that we need a session bean.

Most Seam application use session beans as JSF action listeners (you can use JavaBeans instead if you like).

We have exactly one JSF action in our application, and one session bean method attached to it. In this case, we'll use a stateless session bean, since all the state associated with our action is held by the User bean.

This is the only really interesting code in the example!

Example 1.2. 

@Stateless    (1)
@Name("register")
public class RegisterAction implements Register
{

   @In        (2)
   private User user;
   
   @Persistenc(3)eContext
   private EntityManager em;
   
   @Logger    (4)
   private Log log;
   
   public Stri(5)ng register()
   {
      List existing = em.createQuery(
         "sele(6)ct username from User where username=#{user.username}")
         .getResultList();
         
      if (existing.size()==0)
      {
         em.persist(user);
         log.i(7)nfo("Registered new user #{user.username}");
         retur(8)n "/registered.xhtml";
      }
      else
      {
         Faces(9)Messages.instance().add("User #{user.username} already exists");
         return null;
      }
   }

}
1

The EJB standard @Stateless annotation marks this class as a stateless session bean.

2

The @In annotation marks an attribute of the bean as injected by Seam. In this case, the attribute is injected from a context variable named user (the instance variable name).

3

The EJB standard @PersistenceContext annotation is used to inject the EJB3 entity manager.

4

The Seam @Logger annotation is used to inject the component's Log instance.

5

The action listener method uses the standard EJB3 EntityManager API to interact with the database, and returns the JSF outcome. Note that, since this is a session bean, a transaction is automatically begun when the register() method is called, and committed when it completes.

6

Notice that Seam lets you use a JSF EL expression inside EJB-QL. Under the covers, this results in an ordinary JPA setParameter() call on the standard JPA Query object. Nice, huh?

7

The Log API lets us easily display templated log messages.

8

JSF action listener methods return a string-valued outcome that determines what page will be displayed next. A null outcome (or a void action listener method) redisplays the previous page. In plain JSF, it is normal to always use a JSF navigation rule to determine the JSF view id from the outcome. For complex application this indirection is useful and a good practice. However, for very simple examples like this one, Seam lets you use the JSF view id as the outcome, eliminating the requirement for a navigation rule. Note that when you use a view id as an outcome, Seam always performs a browser redirect.

9

Seam provides a number of built-in components to help solve common problems. The FacesMessages component makes it easy to display templated error or success messages. Built-in Seam components may be obtained by injection, or by calling an instance() method.


Note that we did not explicitly specify a @Scope this time. Each Seam component type has a default scope if not explicitly specified. For stateless session beans, the default scope is the stateless context. Actually, all stateless session beans belong in the stateless context.

Our session bean action listener performs the business and persistence logic for our mini-application. In more complex applications, we might need to layer the code and refactor persistence logic into a dedicated data access component. That's perfectly trivial to do. But notice that Seam does not force you into any particular strategy for application layering.

Furthermore, notice that our session bean has simultaneous access to context associated with the web request (the form values in the User object, for example), and state held in transactional resources (the EntityManager object). This is a break from traditional J2EE architectures. Again, if you are more comfortable with the traditional J2EE layering, you can certainly implement that in a Seam application. But for many applications, it's simply not very useful.

If you've used many Java frameworks before, you'll be used to having to declare all your component classes in some kind of XML file that gradually grows more and more unmanageable as your project matures. You'll be relieved to know that Seam does not require that application components be accompanied by XML. Most Seam applications require a very small amount of XML that does not grow very much as the project gets bigger.

Nevertheless, it is often useful to be able to provide for some external configuration of some components (particularly the components built in to Seam). You have a couple of options here, but the most flexible option is to provide this configuration in a file called components.xml, located in the WEB-INF directory. We'll use the components.xml file to tell Seam how to find our EJB components in JNDI:


This code configures a property named jndiPattern of a built-in Seam component named org.jboss.seam.core.init. The funny @ symbols are there because our Ant build script puts the correct JNDI pattern in when we deploy the application.

The presentation layer for our mini-application will be deployed in a WAR. So we'll need a web deployment descriptor.


This web.xml file configures Seam and JSF. The configuration you see here is pretty much identical in all Seam applications.

The view pages for a Seam application could be implemented using any technology that supports JSF. In this example we use Facelets, because we think it's better than JSP.


The only thing here that is specific to Seam is the <s:validateAll> tag. This JSF component tells JSF to validate all the contained input fields against the Hibernate Validator annotations specified on the entity bean.


This is a boring old Facelets page using some embedded EL. There is nothing specific to Seam here.

When the form is submitted, JSF asks Seam to resolve the variable named user. Since there is no value already bound to that name (in any Seam context), Seam instantiates the user component, and returns the resulting User entity bean instance to JSF after storing it in the Seam session context.

The form input values are now validated against the Hibernate Validator constraints specified on the User entity. If the constraints are violated, JSF redisplays the page. Otherwise, JSF binds the form input values to properties of the User entity bean.

Next, JSF asks Seam to resolve the variable named register. Seam finds the RegisterAction stateless session bean in the stateless context and returns it. JSF invokes the register() action listener method.

Seam intercepts the method call and injects the User entity from the Seam session context, before continuing the invocation.

The register() method checks if a user with the entered username already exists. If so, an error message is queued with the FacesMessages component, and a null outcome is returned, causing a page redisplay. The FacesMessages component interpolates the JSF expression embedded in the message string and adds a JSF FacesMessage to the view.

If no user with that username exists, the "/registered.xhtml" outcome triggers a browser redirect to the registered.xhtml page. When JSF comes to render the page, it asks Seam to resolve the variable named user and uses property values of the returned User entity from Seam's session scope.

Clickable lists of database search results are such an important part of any online application that Seam provides special functionality on top of JSF to make it easier to query data using EJB-QL or HQL and display it as a clickable list using a JSF <h:dataTable>. The messages example demonstrates this functionality.

The message list example has one entity bean, Message, one session bean, MessageListBean and one JSP.

Just like in the previous example, we have a session bean, MessageManagerBean, which defines the action listener methods for the two buttons on our form. One of the buttons selects a message from the list, and displays that message. The other button deletes a message. So far, this is not so different to the previous example.

But MessageManagerBean is also responsible for fetching the list of messages the first time we navigate to the message list page. There are various ways the user could navigate to the page, and not all of them are preceded by a JSF action—the user might have bookmarked the page, for example. So the job of fetching the message list takes place in a Seam factory method, instead of in an action listener method.

We want to cache the list of messages in memory between server requests, so we will make this a stateful session bean.

Example 1.11. 

@Stateful
@Scope(SESSION)
@Name("messageManager")
public class MessageManagerBean implements Serializable, MessageManager
{

   @DataModel (1)
   private List<Message> messageList;
   
   @DataModelS(2)election
   @Out(requir(3)ed=false)
   private Message message;
   
   @Persistenc(4)eContext(type=EXTENDED)
   private EntityManager em;
   
   @Factory("m(5)essageList")
   public void findMessages()
   {
      messageList = em.createQuery("from Message msg order by msg.datetime desc")
                      .getResultList();
   }
   
   public void(6) select()
   {
      message.setRead(true);
   }
   
   public void(7) delete()
   {
      messageList.remove(message);
      em.remove(message);
      message=null;
   }
   
   @Remove    (8)
   public void destroy() {}

}
1

The @DataModel annotation exposes an attibute of type java.util.List to the JSF page as an instance of javax.faces.model.DataModel. This allows us to use the list in a JSF <h:dataTable> with clickable links for each row. In this case, the DataModel is made available in a session context variable named messageList.

2

The @DataModelSelection annotation tells Seam to inject the List element that corresponded to the clicked link.

3

The @Out annotation then exposes the selected value directly to the page. So ever time a row of the clickable list is selected, the Message is injected to the attribute of the stateful bean, and the subsequently outjected to the event context variable named message.

4

This stateful bean has an EJB3 extended persistence context. The messages retrieved in the query remain in the managed state as long as the bean exists, so any subsequent method calls to the stateful bean can update them without needing to make any explicit call to the EntityManager.

5

The first time we navigate to the JSP page, there will be no value in the messageList context variable. The @Factory annotation tells Seam to create an instance of MessageManagerBean and invoke the findMessages() method to initialize the value. We call findMessages() a factory method for messages.

6

The select() action listener method marks the selected Message as read, and updates it in the database.

7

The delete() action listener method removes the selected Message from the database.

8

All stateful session bean Seam components must have a method with no parameters marked @Remove that Seam uses to remove the stateful bean when the Seam context ends, and clean up any server-side state.


Note that this is a session-scoped Seam component. It is associated with the user login session, and all requests from a login session share the same instance of the component. (In Seam applications, we usually use session-scoped components sparingly.)

The JSP page is a straightforward use of the JSF <h:dataTable> component. Again, nothing specific to Seam.

Example 1.12. 


<%@ taglib uri="http://java.sun.com/jsf/html" prefix="h" %>
<%@ taglib uri="http://java.sun.com/jsf/core" prefix="f" %>
<html>
 <head>
  <title>Messages</title>
 </head>
 <body>
  <f:view>
   <h:form>
     <h2>Message List</h2>
     <h:outputText value="No messages to display" 
                   rendered="#{messageList.rowCount==0}"/>
     <h:dataTable var="msg" value="#{messageList}" 
                  rendered="#{messageList.rowCount>0}">
        <h:column>
           <f:facet name="header">
              <h:outputText value="Read"/>
           </f:facet>
           <h:selectBooleanCheckbox value="#{msg.read}" disabled="true"/>
        </h:column>
        <h:column>
           <f:facet name="header">
              <h:outputText value="Title"/>
           </f:facet>
           <h:commandLink value="#{msg.title}" action="#{messageManager.select}"/>
        </h:column>
        <h:column>
           <f:facet name="header">
              <h:outputText value="Date/Time"/>
           </f:facet>
           <h:outputText value="#{msg.datetime}">
              <f:convertDateTime type="both" dateStyle="medium" timeStyle="short"/>
           </h:outputText>
        </h:column>
        <h:column>
           <h:commandButton value="Delete" action="#{messageManager.delete}"/>
        </h:column>
     </h:dataTable>
     <h3><h:outputText value="#{message.title}"/></h3>
     <div><h:outputText value="#{message.text}"/></div>
   </h:form>
  </f:view>
 </body>
</html>

The first time we navigate to the messages.jsp page, whether by a JSF postback (faces request) or a direct browser GET request (non-faces request), the page will try to resolve the messageList context variable. Since this context variable is not initialized, Seam will call the factory method findMessages(), which performs a query against the database and results in a DataModel being outjected. This DataModel provides the row data needed for rendering the <h:dataTable>.

When the user clicks the <h:commandLink>, JSF calls the select() action listener. Seam intercepts this call and injects the selected row data into the message attribute of the messageManager component. The action listener fires, marking the selected Message as read. At the end of the call, Seam outjects the selected Message to the context variable named message. Next, the EJB container commits the transaction, and the change to the Message is flushed to the database. Finally, the page is re-rendered, redisplaying the message list, and displaying the selected message below it.

If the user clicks the <h:commandButton>, JSF calls the delete() action listener. Seam intercepts this call and injects the selected row data into the message attribute of the messageList component. The action listener fires, removing the selected Message from the list, and also calling remove() on the EntityManager. At the end of the call, Seam refreshes the messageList context variable and clears the context variable named message. The EJB container commits the transaction, and deletes the Message from the database. Finally, the page is re-rendered, redisplaying the message list.

jBPM provides sophisticated functionality for workflow and task management. To get a small taste of how jBPM integrates with Seam, we'll show you a simple "todo list" application. Since managing lists of tasks is such core functionality for jBPM, there is hardly any Java code at all in this example.

The center of this example is the jBPM process definition. There are also two JSPs and two trivial JavaBeans (There was no reason to use session beans, since they do not access the database, or have any other transactional behavior). Let's start with the process definition:


If we view this process definition using the process definition editor provided by JBossIDE, this is what it looks like:

This document defines our business process as a graph of nodes. This is the most trivial possible business process: there is one task to be performed, and when that task is complete, the business process ends.

The first JavaBean handles the login screen login.jsp. Its job is just to initialize the jBPM actor id using the actor component. (In a real application, it would also need to authenticate the user.)


Here we see the use of @In to inject the built-in Actor component.

The JSP itself is trivial:


<%@ taglib uri="http://java.sun.com/jsf/html" prefix="h"%>
<%@ taglib uri="http://java.sun.com/jsf/core" prefix="f"%>
<html>
<head>
<title>Login</title>
</head>
<body>
<h1>Login</h1>
<f:view>
    <h:form>
      <div>
        <h:inputText value="#{login.user}"/>
        <h:commandButton value="Login" action="#{login.login}"/>
      </div>
    </h:form>
</f:view>
</body>
</html>

The second JavaBean is responsible for starting business process instances, and ending tasks.


In a more realistic example, @StartTask and @EndTask would not appear on the same method, because there is usually work to be done using the application in order to complete the task.

Finally, the meat of the application is in todo.jsp:

Example 1.16. 


<%@ taglib uri="http://java.sun.com/jsf/html" prefix="h" %>
<%@ taglib uri="http://java.sun.com/jsf/core" prefix="f" %>
<%@ taglib uri="http://jboss.com/products/seam/taglib" prefix="s" %>
<html>
<head>
<title>Todo List</title>
</head>
<body>
<h1>Todo List</h1>
<f:view>
   <h:form id="list">
      <div>
         <h:outputText value="There are no todo items." 
                       rendered="#{empty taskInstanceList}"/>
         <h:dataTable value="#{taskInstanceList}" var="task" 
                      rendered="#{not empty taskInstanceList}">
            <h:column>
                <f:facet name="header">
                    <h:outputText value="Description"/>
                </f:facet>
                <h:inputText value="#{task.description}"/>
            </h:column>
            <h:column>
                <f:facet name="header">
                    <h:outputText value="Created"/>
                </f:facet>
                <h:outputText value="#{task.taskMgmtInstance.processInstance.start}">
                    <f:convertDateTime type="date"/>
                </h:outputText>
            </h:column>
            <h:column>
                <f:facet name="header">
                    <h:outputText value="Priority"/>
                </f:facet>
                <h:inputText value="#{task.priority}" style="width: 30"/>
            </h:column>
            <h:column>
                <f:facet name="header">
                    <h:outputText value="Due Date"/>
                </f:facet>
                <h:inputText value="#{task.dueDate}" style="width: 100">
                    <f:convertDateTime type="date" dateStyle="short"/>
                </h:inputText>
            </h:column>
            <h:column>
                <s:button value="Done" action="#{todoList.done}" taskInstance="#{task}"/>
            </h:column>
         </h:dataTable>
      </div>
      <div>
      <h:messages/>
      </div>
      <div>
         <h:commandButton value="Update Items" action="update"/>
      </div>
   </h:form>
   <h:form id="new">
      <div>
         <h:inputText value="#{todoList.description}"/>
         <h:commandButton value="Create New Item" action="#{todoList.createTodo}"/>
      </div>
   </h:form>
</f:view>
</body>
</html>

Let's take this one piece at a time.

The page renders a list of tasks, which it gets from a built-in Seam component named taskInstanceList. The list is defined inside a JSF form.


Each element of the list is an instance of the jBPM class TaskInstance. The following code simply displays the interesting properties of each task in the list. For the description, priority and due date, we use input controls, to allow the user to update these values.


<h:column>
    <f:facet name="header">
       <h:outputText value="Description"/>
    </f:facet>
    <h:inputText value="#{task.description}"/>
</h:column>
<h:column>
    <f:facet name="header">
        <h:outputText value="Created"/>
    </f:facet>
    <h:outputText value="#{task.taskMgmtInstance.processInstance.start}">
        <f:convertDateTime type="date"/>
    </h:outputText>
</h:column>
<h:column>
    <f:facet name="header">
        <h:outputText value="Priority"/>
    </f:facet>
    <h:inputText value="#{task.priority}" style="width: 30"/>
</h:column>
<h:column>
    <f:facet name="header">
        <h:outputText value="Due Date"/>
    </f:facet>
    <h:inputText value="#{task.dueDate}" style="width: 100">
        <f:convertDateTime type="date" dateStyle="short"/>
    </h:inputText>
</h:column>

This button ends the task by calling the action method annotated @StartTask @EndTask. It passes the task id to Seam as a request parameter:


<h:column>
    <s:button value="Done" action="#{todoList.done}" taskInstance="#{task}"/>
</h:column>

(Note that this is using a Seam <s:button> JSF control from the seam-ui.jar package.)

This button is used to update the properties of the tasks. When the form is submitted, Seam and jBPM will make any changes to the tasks persistent. There is no need for any action listener method:


<h:commandButton value="Update Items" action="update"/>

A second form on the page is used to create new items, by calling the action method annotated @CreateProcess.


<h:form id="new">
    <div>
        <h:inputText value="#{todoList.description}"/>
        <h:commandButton value="Create New Item" action="#{todoList.createTodo}"/>
    </div>
</h:form>

There are several other files needed for the example, but they are just standard jBPM and Seam configuration and not very interesting.

For Seam applications with relatively freeform (ad hoc) navigation, JSF/Seam navigation rules are a perfectly good way to define the page flow. For applications with a more constrained style of navigation, especially for user interfaces which are more stateful, navigation rules make it difficult to really understand the flow of the system. To understand the flow, you need to piece it together from the view pages, the actions and the navigation rules.

Seam allows you to use a jPDL process definition to define pageflow. The simple number guessing example shows how this is done.

The example is implemented using one JavaBean, three JSP pages and a jPDL pageflow definition. Let's begin with the pageflow:

Example 1.18. 

<pageflow-definition 
        xmlns="http://jboss.com/products/seam/pageflow"
        xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
        xsi:schemaLocation="http://jboss.com/products/seam/pageflow 
                            http://jboss.com/products/seam/pageflow-2.1.xsd"
        name="numberGuess">
   
   <start-page(1) name="displayGuess" view-id="/numberGuess.jspx">
      <redirect/>
      <transit(2)ion name="guess" to="evaluateGuess">
         <acti(3)on expression="#{numberGuess.guess}"/>
      </transition>
      <transition name="giveup" to="giveup"/>
   </start-page>
   
   <decision n(4)ame="evaluateGuess" expression="#{numberGuess.correctGuess}">
      <transition name="true" to="win"/>
      <transition name="false" to="evaluateRemainingGuesses"/>
   </decision>
   
   <decision name="evaluateRemainingGuesses" expression="#{numberGuess.lastGuess}">
      <transition name="true" to="lose"/>
      <transition name="false" to="displayGuess"/>
   </decision>
   
   <page name="giveup" view-id="/giveup.jspx">
      <redirect/>
      <transition name="yes" to="lose"/>
      <transition name="no" to="displayGuess"/>
   </page>
   
   <page name="win" view-id="/win.jspx">
      <redirect/>
      <end-conversation/>
   </page>
   
   <page name="lose" view-id="/lose.jspx">
      <redirect/>
      <end-conversation/>
   </page>
   
</pageflow-definition>
1

The <page> element defines a wait state where the system displays a particular JSF view and waits for user input. The view-id is the same JSF view id used in plain JSF navigation rules. The redirect attribute tells Seam to use post-then-redirect when navigating to the page. (This results in friendly browser URLs.)

2

The <transition> element names a JSF outcome. The transition is triggered when a JSF action results in that outcome. Execution will then proceed to the next node of the pageflow graph, after invocation of any jBPM transition actions.

3

A transition <action> is just like a JSF action, except that it occurs when a jBPM transition occurs. The transition action can invoke any Seam component.

4

A <decision> node branches the pageflow, and determines the next node to execute by evaluating a JSF EL expression.


Here is what the pageflow looks like in the JBoss Developer Studio pageflow editor:

Now that we have seen the pageflow, it is very, very easy to understand the rest of the application!

Here is the main page of the application, numberGuess.jspx:

Example 1.19. 


<<?xml version="1.0"?>
<jsp:root xmlns:jsp="http://java.sun.com/JSP/Page" 
          xmlns:h="http://java.sun.com/jsf/html"
          xmlns:f="http://java.sun.com/jsf/core"
          xmlns:s="http://jboss.com/products/seam/taglib"
          xmlns="http://www.w3.org/1999/xhtml"
          version="2.0">
  <jsp:output doctype-root-element="html" 
              doctype-public="-//W3C//DTD XHTML 1.0 Transitional//EN"
              doctype-system="http://www.w3c.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"/>
  <jsp:directive.page contentType="text/html"/>
  <html>
  <head>
    <title>Guess a number...</title>
    <link href="niceforms.css" rel="stylesheet" type="text/css" />
    <script language="javascript" type="text/javascript" src="niceforms.js" />
  </head>
  <body>
    <h1>Guess a number...</h1>
    <f:view>
      <h:form styleClass="niceform">
        
        <div>
        <h:messages globalOnly="true"/>
        <h:outputText value="Higher!" 
               rendered="#{numberGuess.randomNumber gt numberGuess.currentGuess}"/>
        <h:outputText value="Lower!" 
               rendered="#{numberGuess.randomNumber lt numberGuess.currentGuess}"/>
        </div>
        
        <div>
        I'm thinking of a number between 
        <h:outputText value="#{numberGuess.smallest}"/> and 
        <h:outputText value="#{numberGuess.biggest}"/>. You have 
        <h:outputText value="#{numberGuess.remainingGuesses}"/> guesses.
        </div>
        
        <div>
        Your guess: 
        <h:inputText value="#{numberGuess.currentGuess}" id="inputGuess" 
                     required="true" size="3" 
                     rendered="#{(numberGuess.biggest-numberGuess.smallest) gt 20}">
          <f:validateLongRange maximum="#{numberGuess.biggest}" 
                               minimum="#{numberGuess.smallest}"/>
        </h:inputText>
        <h:selectOneMenu value="#{numberGuess.currentGuess}" 
                         id="selectGuessMenu" required="true"
                         rendered="#{(numberGuess.biggest-numberGuess.smallest) le 20 and 
                                     (numberGuess.biggest-numberGuess.smallest) gt 4}">
          <s:selectItems value="#{numberGuess.possibilities}" var="i" label="#{i}"/>
        </h:selectOneMenu>
        <h:selectOneRadio value="#{numberGuess.currentGuess}" id="selectGuessRadio" 
                          required="true"
                          rendered="#{(numberGuess.biggest-numberGuess.smallest) le 4}">
          <s:selectItems value="#{numberGuess.possibilities}" var="i" label="#{i}"/>
        </h:selectOneRadio>
        <h:commandButton value="Guess" action="guess"/>
        <s:button value="Cheat" view="/confirm.jspx"/>
        <s:button value="Give up" action="giveup"/>
        </div>
        
        <div>
        <h:message for="inputGuess" style="color: red"/>
        </div>
        
      </h:form>
    </f:view>
  </body>
  </html>
</jsp:root>

Notice how the command button names the guess transition instead of calling an action directly.

The win.jspx page is predictable:


As is lose.jspx (which I can't be bothered copy/pasting). Finally, the JavaBean Seam component:

Example 1.21. 

@Name("numberGuess")
@Scope(ScopeType.CONVERSATION)
public class NumberGuess implements Serializable {
   
   private int randomNumber;
   private Integer currentGuess;
   private int biggest;
   private int smallest;
   private int guessCount;
   private int maxGuesses;
   private boolean cheated;
   
   @Create    (1)
   public void begin()
   {
      randomNumber = new Random().nextInt(100);
      guessCount = 0;
      biggest = 100;
      smallest = 1;
   }
   
   public void setCurrentGuess(Integer guess)
   {
      this.currentGuess = guess;
   }
   
   public Integer getCurrentGuess()
   {
      return currentGuess;
   }
   
   public void guess()
   {
      if (currentGuess>randomNumber)
      {
         biggest = currentGuess - 1;
      }
      if (currentGuess<randomNumber)
      {
         smallest = currentGuess + 1;
      }
      guessCount ++;
   }
   
   public boolean isCorrectGuess()
   {
      return currentGuess==randomNumber;
   }
   
   public int getBiggest()
   {
      return biggest;
   }
   
   public int getSmallest()
   {
      return smallest;
   }
   
   public int getGuessCount()
   {
      return guessCount;
   }
   
   public boolean isLastGuess()
   {
      return guessCount==maxGuesses;
   }

   public int getRemainingGuesses() {
      return maxGuesses-guessCount;
   }

   public void setMaxGuesses(int maxGuesses) {
      this.maxGuesses = maxGuesses;
   }

   public int getMaxGuesses() {
      return maxGuesses;
   }

   public int getRandomNumber() {
      return randomNumber;
   }

   public void cheated()
   {
      cheated = true;
   }
   
   public boolean isCheat() {
      return cheated;
   }
   
   public List<Integer> getPossibilities()
   {
      List<Integer> result = new ArrayList<Integer>();
      for(int i=smallest; i<=biggest; i++) result.add(i);
      return result;
   }
   
}
1

The first time a JSP page asks for a numberGuess component, Seam will create a new one for it, and the @Create method will be invoked, allowing the component to initialize itself.


The pages.xml file starts a Seam conversation (much more about that later), and specifies the pageflow definition to use for the conversation's page flow.


<?xml version="1.0" encoding="UTF-8"?>
<pages xmlns="http://jboss.com/products/seam/pages"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
       xsi:schemaLocation="http://jboss.com/products/seam/pages http://jboss.com/products/seam/pages-2.1.xsd">

  <page view-id="/numberGuess.jspx">
    <begin-conversation join="true" pageflow="numberGuess"/>
  </page>

  <page view-id="/confirm.jspx">
    <begin-conversation nested="true" pageflow="cheat"/>
  </page>

</pages>      

As you can see, this Seam component is pure business logic! It doesn't need to know anything at all about the user interaction flow. This makes the component potentially more reuseable.

The project structure is identical to the previous one, to install and deploy this application, please refer to Section 1.1, “Try the examples”. Once you've successfully started the application, you can access it by pointing your browser to http://localhost:8080/seam-booking/

Just nine classes (plus six session beans local interfaces) where used to implement this application. Six session bean action listeners contain all the business logic for the listed features.

  • BookingListAction retrieves existing bookings for the currently logged in user.

  • ChangePasswordAction updates the password of the currently logged in user.

  • HotelBookingAction implements the core functionality of the application: hotel room searching, selection, booking and booking confirmation. This functionality is implemented as a conversation, so this is the most interesting class in the application.

  • RegisterAction registers a new system user.

Three entity beans implement the application's persistent domain model.

  • Hotel is an entity bean that represent a hotel

  • Booking is an entity bean that represents an existing booking

  • User is an entity bean to represents a user who can make hotel bookings

We encourage you browse the sourcecode at your pleasure. In this tutorial we'll concentrate upon one particular piece of functionality: hotel search, selection, booking and confirmation. From the point of view of the user, everything from selecting a hotel to confirming a booking is one continuous unit of work, a conversation. Searching, however, is not part of the conversation. The user can select multiple hotels from the same search results page, in different browser tabs.

Most web application architectures have no first class construct to represent a conversation. This causes enormous problems managing state associated with the conversation. Usually, Java web applications use a combination of two techniques: first, some state is thrown into the HttpSession; second, persistable state is flushed to the database after every request, and reconstructed from the database at the beginning of each new request.

Since the database is the least scalable tier, this often results in an utterly unacceptable lack of scalability. Added latency is also a problem, due to the extra traffic to and from the database on every request. To reduce this redundant traffic, Java applications often introduce a data (second-level) cache that keeps commonly accessed data between requests. This cache is necessarily inefficient, because invalidation is based upon an LRU policy instead of being based upon when the user has finished working with the data. Furthermore, because the cache is shared between many concurrent transactions, we've introduced a whole raft of problem's associated with keeping the cached state consistent with the database.

Now consider the state held in the HttpSession. By very careful programming, we might be able to control the size of the session data. This is a lot more difficult than it sounds, since web browsers permit ad hoc non-linear navigation. But suppose we suddenly discover a system requirement that says that a user is allowed to have mutiple concurrent conversations, halfway through the development of the system (this has happened to me). Developing mechanisms to isolate session state associated with different concurrent conversations, and incorporating failsafes to ensure that conversation state is destroyed when the user aborts one of the conversations by closing a browser window or tab is not for the faint hearted (I've implemented this stuff twice so far, once for a client application, once for Seam, but I'm famously psychotic).

Now there is a better way.

Seam introduces the conversation context as a first class construct. You can safely keep conversational state in this context, and be assured that it will have a well-defined lifecycle. Even better, you won't need to be continually pushing data back and forth between the application server and the database, since the conversation context is a natural cache of data that the user is currently working with.

Usually, the components we keep in the conversation context are stateful session beans. (We can also keep entity beans and JavaBeans in the conversation context.) There is an ancient canard in the Java community that stateful session beans are a scalability killer. This may have been true in 1998 when WebFoobar 1.0 was released. It is no longer true today. Application servers like JBoss AS have extremely sophisticated mechanisms for stateful session bean state replication. (For example, the JBoss EJB3 container performs fine-grained replication, replicating only those bean attribute values which actually changed.) Note that all the traditional technical arguments for why stateful beans are inefficient apply equally to the HttpSession, so the practice of shifting state from business tier stateful session bean components to the web session to try and improve performance is unbelievably misguided. It is certainly possible to write unscalable applications using stateful session beans, by using stateful beans incorrectly, or by using them for the wrong thing. But that doesn't mean you should never use them. Anyway, Seam guides you toward a safe usage model. Welcome to 2005.

OK, I'll stop ranting now, and get back to the tutorial.

The booking example application shows how stateful components with different scopes can collaborate together to achieve complex behaviors. The main page of the booking application allows the user to search for hotels. The search results are kept in the Seam session scope. When the user navigates to one of these hotels, a conversation begins, and a conversation scoped component calls back to the session scoped component to retrieve the selected hotel.

The booking example also demonstrates the use of RichFaces Ajax to implement rich client behavior without the use of handwritten JavaScript.

The search functionality is implemented using a session-scope stateful session bean, similar to the one we saw in the message list example above.

Example 1.22. 

@Stateful     (1)
@Name("hotelSearch")
@Scope(ScopeType.SESSION)
@Restrict("#{i(2)dentity.loggedIn}")
public class HotelSearchingAction implements HotelSearching
{
   
   @PersistenceContext
   private EntityManager em;
   
   private String searchString;
   private int pageSize = 10;
   private int page;
   
   @DataModel (3)
   private List<Hotel> hotels;
   
   public void find()
   {
      page = 0;
      queryHotels();
   }
   public void nextPage()
   {
      page++;
      queryHotels();
   }
      
   private void queryHotels()
   {
      hotels = 
          em.createQuery("select h from Hotel h where lower(h.name) like #{pattern} " + 
                         "or lower(h.city) like #{pattern} " + 
                         "or lower(h.zip) like #{pattern} " +
                         "or lower(h.address) like #{pattern}")
            .setMaxResults(pageSize)
            .setFirstResult( page * pageSize )
            .getResultList();
   }
   
   public boolean isNextPageAvailable()
   {
      return hotels!=null && hotels.size()==pageSize;
   }
   
   public int getPageSize() {
      return pageSize;
   }
   
   public void setPageSize(int pageSize) {
      this.pageSize = pageSize;
   }
   
   @Factory(value="pattern", scope=ScopeType.EVENT)
   public String getSearchPattern()
   {
      return searchString==null ? 
            "%" : '%' + searchString.toLowerCase().replace('*', '%') + '%';
   }
   
   public String getSearchString()
   {
      return searchString;
   }
   
   public void setSearchString(String searchString)
   {
      this.searchString = searchString;
   }
              (4)
   @Remove
   public void destroy() {}
}
1

The EJB standard @Stateful annotation identifies this class as a stateful session bean. Stateful session beans are scoped to the conversation context by default.

2

The @Restrict annotation applies a security restriction to the component. It restricts access to the component allowing only logged-in users. The security chapter explains more about security in Seam.

3

The @DataModel annotation exposes a List as a JSF ListDataModel. This makes it easy to implement clickable lists for search screens. In this case, the list of hotels is exposed to the page as a ListDataModel in the conversation variable named hotels.

4

The EJB standard @Remove annotation specifies that a stateful session bean should be removed and its state destroyed after invocation of the annotated method. In Seam, all stateful session beans must define a method with no parameters marked @Remove. This method will be called when Seam destroys the session context.


The main page of the application is a Facelets page. Let's look at the fragment which relates to searching for hotels:

Example 1.23. 

<div class="section">
  
    <span class="errors">
       <h:messages globalOnly="true"/>
    </span>
    
    <h1>Search Hotels</h1>

	<h:form id="searchCriteria">
	<fieldset> 
	   <h:inputText id="searchString" value="#{hotelSearch.searchString}" 
                    style="width: 165px;">
         <a:support event="onkeyup" actionListener="#{hotelSearch.find}" 
              (1)      reRender="searchResults" />
       </h:inputText>
       &#160;
	   <a:commandButton id="findHotels" value="Find Hotels" action="#{hotelSearch.find}" 
                        reRender="searchResults"/>
       &#160;
       <a:stat(2)us>
          <f:facet name="start">
             <h:graphicImage value="/img/spinner.gif"/>
          </f:facet>
       </a:status>
	   <br/>
       <h:outputLabel for="pageSize">Maximum results:</h:outputLabel>&#160;
       <h:selectOneMenu value="#{hotelSearch.pageSize}" id="pageSize">
          <f:selectItem itemLabel="5" itemValue="5"/>
          <f:selectItem itemLabel="10" itemValue="10"/>
          <f:selectItem itemLabel="20" itemValue="20"/>
       </h:selectOneMenu>
    </fieldset>
    </h:form>
    
</div>

<a:outputPanel(3) id="searchResults">
  <div class="section">
    <h:outputText value="No Hotels Found"
                  rendered="#{hotels != null and hotels.rowCount==0}"/>
    <h:dataTable id="hotels" value="#{hotels}" var="hot" 
                 rendered="#{hotels.rowCount>0}">
        <h:column>
            <f:facet name="header">Name</f:facet>
            #{hot.name}
        </h:column>
        <h:column>
            <f:facet name="header">Address</f:facet>
            #{hot.address}
        </h:column>
        <h:column>
            <f:facet name="header">City, State</f:facet>
            #{hot.city}, #{hot.state}, #{hot.country}
        </h:column> 
        <h:column>
            <f:facet name="header">Zip</f:facet>
            #{hot.zip}
        </h:column>
        <h:column>
            <f:facet name="header">Action</f:facet>
            <s(4):link id="viewHotel" value="View Hotel" 
                    action="#{hotelBooking.selectHotel(hot)}"/>
        </h:column>
    </h:dataTable>
    <s:link value="More results" action="#{hotelSearch.nextPage}" 
            rendered="#{hotelSearch.nextPageAvailable}"/>
  </div>
</a:outputPanel>    
1

The RichFaces Ajax <a:support> tag allows a JSF action event listener to be called by asynchronous XMLHttpRequest when a JavaScript event like onkeyup occurs. Even better, the reRender attribute lets us render a fragment of the JSF page and perform a partial page update when the asynchronous response is received.

2

The RichFaces Ajax <a:status> tag lets us display a cheesy annimated image while we wait for asynchronous requests to return.

3

The RichFaces Ajax <a:outputPanel> tag defines a region of the page which can be re-rendered by an asynchronous request.

4

The Seam <s:link> tag lets us attach a JSF action listener to an ordinary (non-JavaScript) HTML link. The advantage of this over the standard JSF <h:commandLink> is that it preserves the operation of "open in new window" and "open in new tab". Also notice that we use a method binding with a parameter: #{hotelBooking.selectHotel(hot)}. This is not possible in the standard Unified EL, but Seam provides an extension to the EL that lets you use parameters on any method binding expression.

If you're wondering how navigation occurs, you can find all the rules in WEB-INF/pages.xml; this is discussed in Section 6.6, “Navigation”.


This page displays the search results dynamically as we type, and lets us choose a hotel and pass it to the selectHotel() method of the HotelBookingAction, which is where the really interesting stuff is going to happen.

Now let's see how the booking example application uses a conversation-scoped stateful session bean to achieve a natural cache of persistent data related to the conversation. The following code example is pretty long. But if you think of it as a list of scripted actions that implement the various steps of the conversation, it's understandable. Read the class from top to bottom, as if it were a story.

Example 1.24. 

@Stateful
@Name("hotelBooking")
@Restrict("#{identity.loggedIn}")
public class HotelBookingAction implements HotelBooking
{
   
   @Persistenc(1)eContext(type=EXTENDED)
   private EntityManager em;
   
   @In 
   private User user;
   
   @In(required=false) @Out
   private Hotel hotel;
   
   @In(required=false) 
   @Out(requir(2)ed=false)
   private Booking booking;
     
   @In
   private FacesMessages facesMessages;
      
   @In
   private Events events;
   
   @Logger 
   private Log log;
   
   private boolean bookingValid;
   
   @Begin     (3)
   public void selectHotel(Hotel selectedHotel)
   {
      hotel = em.merge(selectedHotel);
   }
   
   public void bookHotel()
   {      
      booking = new Booking(hotel, user);
      Calendar calendar = Calendar.getInstance();
      booking.setCheckinDate( calendar.getTime() );
      calendar.add(Calendar.DAY_OF_MONTH, 1);
      booking.setCheckoutDate( calendar.getTime() );
   }
   
   public void setBookingDetails()
   {
      Calendar calendar = Calendar.getInstance();
      calendar.add(Calendar.DAY_OF_MONTH, -1);
      if ( booking.getCheckinDate().before( calendar.getTime() ) )
      {
         facesMessages.addToControl("checkinDate", "Check in date must be a future date");
         bookingValid=false;
      }
      else if ( !booking.getCheckinDate().before( booking.getCheckoutDate() ) )
      {
         facesMessages.addToControl("checkoutDate", 
                                    "Check out date must be later than check in date");
         bookingValid=false;
      }
      else
      {
         bookingValid=true;
      }
   }
   
   public boolean isBookingValid()
   {
      return bookingValid;
   }
   
   @End       (4)
   public void confirm()
   {
      em.persist(booking);
      facesMessages.add("Thank you, #{user.name}, your confimation number " + 
                        " for #{hotel.name} is #{booki g.id}");
      log.info("New booking: #{booking.id} for #{user.username}");
      events.raiseTransactionSuccessEvent("bookingConfirmed");
   }
   
   @End
   public void cancel() {}
   
   @Remove    (5)
   public void destroy() {}
1

This bean uses an EJB3 extended persistence context, so that any entity instances remain managed for the whole lifecycle of the stateful session bean.

2

The @Out annotation declares that an attribute value is outjected to a context variable after method invocations. In this case, the context variable named hotel will be set to the value of the hotel instance variable after every action listener invocation completes.

3

The @Begin annotation specifies that the annotated method begins a long-running conversation, so the current conversation context will not be destroyed at the end of the request. Instead, it will be reassociated with every request from the current window, and destroyed either by timeout due to conversation inactivity or invocation of a matching @End method.

4

The @End annotation specifies that the annotated method ends the current long-running conversation, so the current conversation context will be destroyed at the end of the request.

5

This EJB remove method will be called when Seam destroys the conversation context. Don't forget to define this method!


HotelBookingAction contains all the action listener methods that implement selection, booking and booking confirmation, and holds state related to this work in its instance variables. We think you'll agree that this code is much cleaner and simpler than getting and setting HttpSession attributes.

Even better, a user can have multiple isolated conversations per login session. Try it! Log in, run a search, and navigate to different hotel pages in multiple browser tabs. You'll be able to work on creating two different hotel reservations at the same time. If you leave any one conversation inactive for long enough, Seam will eventually time out that conversation and destroy its state. If, after ending a conversation, you backbutton to a page of that conversation and try to perform an action, Seam will detect that the conversation was already ended, and redirect you to the search page.

Seam makes it very easy to implement applications which keep state on the server-side. However, server-side state is not always appropriate, especially in for functionality that serves up content. For this kind of problem we often need to let the user bookmark pages and have a relatively stateless server, so that any page can be accessed at any time, via the bookmark. The Blog example shows how to a implement RESTful application using Seam. Every page of the application can be bookmarked, including the search results page.

The Blog example demonstrates the use of "pull"-style MVC, where instead of using action listener methods to retrieve data and prepare the data for the view, the view pulls data from components as it is being rendered.

This snippet from the index.xhtml facelets page displays a list of recent blog entries:


If we navigate to this page from a bookmark, how does the data used by the <h:dataTable> actually get initialized? Well, what happens is that the Blog is retrieved lazily—"pulled"—when needed, by a Seam component named blog. This is the opposite flow of control to what is usual in traditional web action-based frameworks like Struts.


This is good so far, but what about bookmarking the result of form submissions, such as a search results page?

The blog example has a tiny form in the top right of each page that allows the user to search for blog entries. This is defined in a file, menu.xhtml, included by the facelets template, template.xhtml:


Then the form would have looked like this:


<div id="search">
   <h:form>
      <h:inputText value="#{searchAction.searchPattern}"/>
      <h:commandButton value="Search" action="searchResults"/>
   </h:form>
</div>

But when we redirect, we need to include the values submitted with the form as request parameters, to get a bookmarkable URL like http://localhost:8080/seam-blog/search.seam?searchPattern=seam. JSF does not provide an easy way to do this, but Seam does. We use a Seam page parameter, defined in WEB-INF/pages.xml:


This tells Seam to include the value of #{searchService.searchPattern} as a request parameter named searchPattern when redirecting to the page, and then re-apply the value of that parameter to the model before rendering the page.

The redirect takes us to the search.xhtml page:


<h:dataTable value="#{searchResults}" var="blogEntry">
   <h:column>
      <div>
         <h:outputLink value="entry.seam">
            <f:param name="blogEntryId" value="#{blogEntry.id}"/>
            #{blogEntry.title}
         </h:outputLink>
         posted on 
         <h:outputText value="#{blogEntry.date}">
            <f:convertDateTime timeZone="#{blog.timeZone}" locale="#{blog.locale}" type="both"/>
         </h:outputText>
      </div>
   </h:column>
</h:dataTable>

Which again uses "pull"-style MVC to retrieve the actual search results:

@Name("searchService")

public class SearchService 
{
   
   @In
   private EntityManager entityManager;
   
   private String searchPattern;
   
   @Factory("searchResults")
   public List<BlogEntry> getSearchResults()
   {
      if (searchPattern==null)
      {
         return null;
      }
      else
      {
         return entityManager.createQuery("select be from BlogEntry be "+ 
                      "where lower(be.title) like :searchPattern " + 
                      "lower(be.body) like :searchPattern order by be.date desc")
               .setParameter( "searchPattern", getSqlSearchPattern() )
               .setMaxResults(100)
               .getResultList();
      }
   }
   private String getSqlSearchPattern()
   {
      return searchPattern==null ? "" :
             '%' + searchPattern.toLowerCase().replace('*', '%').replace('?', '_') + '%';
   }
   public String getSearchPattern()
   {
      return searchPattern;
   }
   public void setSearchPattern(String searchPattern)
   {
      this.searchPattern = searchPattern;
   }
}

Very occasionally, it makes more sense to use push-style MVC for processing RESTful pages, and so Seam provides the notion of a page action. The Blog example uses a page action for the blog entry page, entry.xhtml. Note that this is a little bit contrived, it would have been easier to use pull-style MVC here as well.

The entryAction component works much like an action class in a traditional push-MVC action-oriented framework like Struts:

@Name("entryAction")

@Scope(STATELESS)
public class EntryAction
{
   @In(create=true) 
   private Blog blog;
   
   @Out
   private BlogEntry blogEntry;
   
   public void loadBlogEntry(String id) throws EntryNotFoundException
   {
      blogEntry = blog.getBlogEntry(id);
      if (blogEntry==null) throw new EntryNotFoundException(id);
   }
   
}

Page actions are also declared in pages.xml:


<pages>
   ...

   <page view-id="/entry.xhtml" action="#{entryAction.loadBlogEntry(blogEntry.id)}">
      <param name="blogEntryId" value="#{blogEntry.id}"/>
   </page>

   <page view-id="/post.xhtml" action="#{loginAction.challenge}"/>

   <page view-id="*" action="#{blog.hitCount.hit}"/>

</pages>

Notice that the example is using page actions for some other functionality—the login challenge, and the pageview counter. Also notice the use of a parameter in the page action method binding. This is not a standard feature of JSF EL, but Seam lets you use it, not just for page actions, but also in JSF method bindings.

When the entry.xhtml page is requested, Seam first binds the page parameter blogEntryId to the model, then runs the page action, which retrieves the needed data—the blogEntry—and places it in the Seam event context. Finally, the following is rendered:


<div class="blogEntry">
   <h3>#{blogEntry.title}</h3>
   <div>
      <h:outputText escape="false" value="#{blogEntry.body}"/>
   </div>
   <p>
      [Posted on&#160;
      <h:outputText value="#{blogEntry.date}">
         <f:convertDateTime timezone="#{blog.timeZone}" 
                            locale="#{blog.locale}" type="both"/>
      </h:outputText>]
   </p>
</div>

If the blog entry is not found in the database, the EntryNotFoundException exception is thrown. We want this exception to result in a 404 error, not a 505, so we annotate the exception class:

@ApplicationException(rollback=true)

@HttpError(errorCode=HttpServletResponse.SC_NOT_FOUND)
public class EntryNotFoundException extends Exception
{
   EntryNotFoundException(String id)
   {
      super("entry not found: " + id);
   }
}

An alternative implementation of the example does not use the parameter in the method binding:

@Name("entryAction")

@Scope(STATELESS)
public class EntryAction
{
   @In(create=true) 
   private Blog blog;
   
   @In @Out
   private BlogEntry blogEntry;
   
   public void loadBlogEntry() throws EntryNotFoundException
   {
      blogEntry = blog.getBlogEntry( blogEntry.getId() );
      if (blogEntry==null) throw new EntryNotFoundException(id);
   }
   
}

<pages>
   ...

   <page view-id="/entry.xhtml" action="#{entryAction.loadBlogEntry}">
      <param name="blogEntryId" value="#{blogEntry.id}"/>
   </page>
   
   ...
</pages>

It is a matter of taste which implementation you prefer.

The Seam distribution includes a command line utility that makes it really easy to set up an Eclipse project, generate some simple Seam skeleton code, and reverse engineer an application from a preexisting database.

This is the easy way to get your feet wet with Seam, and gives you some ammunition for next time you find yourself trapped in an elevator with one of those tedious Ruby guys ranting about how great and wonderful his new toy is for building totally trivial applications that put things in databases.

In this release, seam-gen works best for people with JBoss AS. You can use the generated project with other J2EE or Java EE 5 application servers by making a few manual changes to the project configuration.

You can use seam-gen without Eclipse, but in this tutorial, we want to show you how to use it in conjunction with Eclipse for debugging and integration testing. If you don't want to install Eclipse, you can still follow along with this tutorial—all steps can be performed from the command line.

Seam-gen is basically just a big ugly Ant script wrapped around Hibernate Tools, together with some templates. That makes it easy to customize if you need to.

The first thing we need to do is configure seam-gen for your environment: JBoss AS installation directory, Eclipse workspace, and database connection. It's easy, just type:

cd jboss-seam-2.0.x
seam setup

And you will be prompted for the needed information:

~/workspace/jboss-seam$ ./seam setup
Buildfile: build.xml

init:

setup:
     [echo] Welcome to seam-gen :-)
    [input] Enter your Java project workspace (the directory that contains your Seam projects) [C:/Projects] [C:/Projects]
/Users/pmuir/workspace
    [input] Enter your JBoss home directory [C:/Program Files/jboss-4.2.2.GA] [C:/Program Files/jboss-4.2.2.GA]
/Applications/jboss-4.2.2.GA
    [input] Enter the project name [myproject] [myproject]
helloworld
     [echo] Accepted project name as: helloworld
    [input] Select a RichFaces skin (not applicable if using ICEFaces) [blueSky] ([blueSky], classic, ruby, wine, deepMarine, emeraldTown, sakura, DEFAULT)

    [input] Is this project deployed as an EAR (with EJB components) or a WAR (with no EJB support) [ear]  ([ear], war, )

    [input] Enter the Java package name for your session beans [com.mydomain.helloworld] [com.mydomain.helloworld]
org.jboss.helloworld
    [input] Enter the Java package name for your entity beans [org.jboss.helloworld] [org.jboss.helloworld]

    [input] Enter the Java package name for your test cases [org.jboss.helloworld.test] [org.jboss.helloworld.test]

    [input] What kind of database are you using? [hsql]  ([hsql], mysql, oracle, postgres, mssql, db2, sybase, enterprisedb, h2)
mysql
    [input] Enter the Hibernate dialect for your database [org.hibernate.dialect.MySQLDialect] [org.hibernate.dialect.MySQLDialect]

    [input] Enter the filesystem path to the JDBC driver jar [lib/hsqldb.jar] [lib/hsqldb.jar]
/Users/pmuir/java/mysql.jar
    [input] Enter JDBC driver class for your database [com.mysql.jdbc.Driver] [com.mysql.jdbc.Driver]

    [input] Enter the JDBC URL for your database [jdbc:mysql:///test] [jdbc:mysql:///test]
jdbc:mysql:///helloworld
    [input] Enter database username [sa] [sa]
pmuir
    [input] Enter database password [] []

    [input] skipping input as property hibernate.default_schema.new has already been set.
    [input] Enter the database catalog name (it is OK to leave this blank) [] []

    [input] Are you working with tables that already exist in the database? [n]  (y, [n], )
y
    [input] Do you want to drop and recreate the database tables and data in import.sql each time you deploy? [n]  (y, [n], )
n
    [input] Enter your ICEfaces home directory (leave blank to omit ICEfaces) [] []

[propertyfile] Creating new property file: /Users/pmuir/workspace/jboss-seam/seam-gen/build.properties
     [echo] Installing JDBC driver jar to JBoss server
     [echo] Type 'seam create-project' to create the new project

BUILD SUCCESSFUL
Total time: 1 minute 32 seconds
~/workspace/jboss-seam $ 

The tool provides sensible defaults, which you can accept by just pressing enter at the prompt.

The most important choice you need to make is between EAR deployment and WAR deployment of your project. EAR projects support EJB 3.0 and require Java EE 5. WAR projects do not support EJB 3.0, but may be deployed to a J2EE environment. The packaging of a WAR is also simpler to understand. If you installed an EJB3-ready application server like JBoss, choose ear. Otherwise, choose war. We'll assume that you've chosen an EAR deployment for the rest of the tutorial, but you can follow exactly the same steps for a WAR deployment.

If you are working with an existing data model, make sure you tell seam-gen that the tables already exist in the database.

The settings are stored in seam-gen/build.properties, but you can also modify them simply by running seam setup a second time.

Now we can create a new project in our Eclipse workspace directory, by typing:

seam new-project
C:\Projects\jboss-seam>seam new-project
Buildfile: build.xml

...

new-project:
     [echo] A new Seam project named 'helloworld' was created in the C:\Projects directory
     [echo] Type 'seam explode' and go to http://localhost:8080/helloworld
     [echo] Eclipse Users: Add the project into Eclipse using File > New > Project and select General > Project (not Java Project)
     [echo] NetBeans Users: Open the project in NetBeans

BUILD SUCCESSFUL
Total time: 7 seconds
C:\Projects\jboss-seam>

This copies the Seam jars, dependent jars and the JDBC driver jar to a new Eclipse project, and generates all needed resources and configuration files, a facelets template file and stylesheet, along with Eclipse metadata and an Ant build script. The Eclipse project will be automatically deployed to an exploded directory structure in JBoss AS as soon as you add the project using New -> Project... -> General -> Project -> Next, typing the Project name (helloworld in this case), and then clicking Finish. Do not select Java Project from the New Project wizard.

If your default JDK in Eclipse is not a Java SE 5 or Java SE 6 JDK, you will need to select a Java SE 5 compliant JDK using Project -> Properties -> Java Compiler.

Alternatively, you can deploy the project from outside Eclipse by typing seam explode.

Go to http://localhost:8080/helloworld to see a welcome page. This is a facelets page, view/home.xhtml, using the template view/layout/template.xhtml. You can edit this page, or the template, in eclipse, and see the results immediately, by clicking refresh in your browser.

Don't get scared by the XML configuration documents that were generated into the project directory. They are mostly standard Java EE stuff, the stuff you need to create once and then never look at again, and they are 90% the same between all Seam projects. (They are so easy to write that even seam-gen can do it.)

The generated project includes three database and persistence configurations. The persistence-test.xml and import-test.sql files are used when running the TestNG unit tests against HSQLDB. The database schema and the test data in import-test.sql is always exported to the database before running tests. The myproject-dev-ds.xml, persistence-dev.xmland import-dev.sql files are for use when deploying the application to your development database. The schema might be exported automatically at deployment, depending upon whether you told seam-gen that you are working with an existing database. The myproject-prod-ds.xml, persistence-prod.xmland import-prod.sql files are for use when deploying the application to your production database. The schema is not exported automatically at deployment.

If you're used to traditional action-style web frameworks, you're probably wondering how you can create a simple web page with a stateless action method in Java. If you type:

seam new-action

Seam will prompt for some information, and generate a new facelets page and Seam component for your project.

C:\Projects\jboss-seam>seam new-action
Buildfile: build.xml

validate-workspace:

validate-project:

action-input:
    [input] Enter the Seam component name
ping
    [input] Enter the local interface name [Ping]

    [input] Enter the bean class name [PingBean]

    [input] Enter the action method name [ping]

    [input] Enter the page name [ping]


setup-filters:

new-action:
     [echo] Creating a new stateless session bean component with an action method
     [copy] Copying 1 file to C:\Projects\helloworld\src\action\org\jboss\helloworld
     [copy] Copying 1 file to C:\Projects\helloworld\src\action\org\jboss\helloworld
     [copy] Copying 1 file to C:\Projects\helloworld\src\action\org\jboss\helloworld\test
     [copy] Copying 1 file to C:\Projects\helloworld\src\action\org\jboss\helloworld\test
     [copy] Copying 1 file to C:\Projects\helloworld\view
     [echo] Type 'seam restart' and go to http://localhost:8080/helloworld/ping.seam

BUILD SUCCESSFUL
Total time: 13 seconds
C:\Projects\jboss-seam>

Because we've added a new Seam component, we need to restart the exploded directory deployment. You can do this by typing seam restart, or by running the restart target in the generated project build.xml file from inside Eclipse. Another way to force a restart is to edit the file resources/META-INF/application.xml in Eclipse. Note that you do not need to restart JBoss each time you change the application.

Now go to http://localhost:8080/helloworld/ping.seam and click the button. You can see the code behind this action by looking in the project src directory. Put a breakpoint in the ping() method, and click the button again.

Finally, locate the PingTest.xml file in the test package and run the integration tests using the TestNG plugin for Eclipse. Alternatively, run the tests using seam test or the test target of the generated build.

When you deploy your Seam application as an exploded directory, you'll get some support for incremental hot deployment at development time. You need to enable debug mode in both Seam and Facelets, by adding this line to components.xml:


<core:init debug="true">

Now, the following files may be redeployed without requiring a full restart of the web application:

But if we want to change any Java code, we still need to do a full restart of the application. (In JBoss this may be accomplished by touching the top level deployment descriptor: application.xml for an EAR deployment, or web.xml for a WAR deployment.)

But if you really want a fast edit/compile/test cycle, Seam supports incremental redeployment of JavaBean components. To make use of this functionality, you must deploy the JavaBean components into the WEB-INF/dev directory, so that they will be loaded by a special Seam classloader, instead of by the WAR or EAR classloader.

You need to be aware of the following limitations:

If you create a WAR project using seam-gen, incremental hot deployment is available out of the box for classes in the src/action source directory. However, seam-gen does not support incremental hot deployment for EAR projects.

JBoss Tools is a collection of Eclipse plugins. JBoss Tools a project creation wizard for Seam, Content Assist for the Unified Expression Language (EL) in both facelets and Java code, a graphical editor for jPDL, a graphical editor for Seam configuration files, support for running Seam integration tests from within Eclipse, and much more.

In short, if you are an Eclipse user, then you'll want JBoss Tools!

JBoss Tools, as with seam-gen, works best with JBoss AS, but it's possible with a few tweaks to get your app running on other application servers. The changes are much like those described for seam-gen later in this reference manual.

Start up Eclipse and select the Seam perspective.

Go to File -> New -> Seam Web Project.

First, enter a name for your new project. For this tutorial, we're going to use helloworld .

Now, we need to tell JBoss Tools about JBoss AS. This is a two stage process, first we need to define a runtime, make sure you select JBoss AS 4.2:

Enter a name for the runtime, and locate it on your hard drive:

Next, we need to define a server JBoss Tools can deploy the project to. Make sure to again select JBoss AS 4.2, and also the runtime you just defined:

On the next screen give the server a name, and hit Finish:

Make sure the runtime and server you just created are selected, select Dynamic Web Project with Seam 2.0 (technology preview) and hit Next:

The next 3 screens allow you to further customize your new project, but for us the defaults are fine. So just hit <empahsis>Next</empahsis> until you reach the final screen.

The first step here is to tell JBoss Tools about the Seam download you want to use. Add a new Seam Runtime - make sure to give it a name, and select 2.0 as the version:

The most important choice you need to make is between EAR deployment and WAR deployment of your project. EAR projects support EJB 3.0 and require Java EE 5. WAR projects do not support EJB 3.0, but may be deployed to a J2EE environment. The packaging of a WAR is also simpler to understand. If you installed an EJB3-ready application server like JBoss, choose EAR. Otherwise, choose WAR. We'll assume that you've chosen a WAR deployment for the rest of the tutorial, but you can follow exactly the same steps for a EAR deployment.

Next, select your database type. We'll assume you have MySQL installed, with an existing schema. You'll need to tell JBoss Tools about the database, select MySQL as the database, and create a new connection profile. Select Generic JDBC Connection:

Give it a name:

JBoss Tools doesn't come with drivers for any databases, so you need to tell JBoss Tools where the MySQL JDBC driver is. Tell it about the driver by clicking ....

Locate MySQL 5, and hit Add...:

Choose the MySQL JDBC Driver template:

Locate the jar on your computer by choosing Edit Jar/Zip:

Review the username and password used to connect, and if correct, hit Ok.

Finally, choose the newly created driver:

If you are working with an existing data model, make sure you tell JBoss Tools that the tables already exist in the database.

Review the username and password used to connect, test the connection using the Test Connection button, and if it works, hit Finish:

Finally, review the package names for your generated beans, and if you are happy, click Finish:

JBoss has sophisticated support for hot re-deployment of WARs and EARs. Unfortunately, due to bugs in the JVM, repeated redeployment of an EAR—which is common during development—eventually causes the JVM to run out of perm gen space. For this reason, we recommend running JBoss in a JVM with a large perm gen space at development time. We suggest the following values:

         -Xms512m -Xmx1024m -XX:PermSize=256m -XX:MaxPermSize=512
      

If you don't have so much memory available, the following is our minimum recommendation:

         -Xms256m -Xmx512m -XX:PermSize=128m -XX:MaxPermSize=256
      

Locate the server in the JBoss Server View, right click on the server and select Edit Launch Configuration:

Then, alter the VM arguements:

If you don't want to bother with this stuff now, you don't have to—come back to it later, when you get your first OutOfMemoryException.

To start JBoss, and deploy the project, just right click on the server you created, and click Start, (or Debug to start in debug mode):

Don't get scared by the XML configuration documents that were generated into the project directory. They are mostly standard Java EE stuff, the stuff you need to create once and then never look at again, and they are 90% the same between all Seam projects.

The two core concepts in Seam are the notion of a context and the notion of a component. Components are stateful objects, usually EJBs, and an instance of a component is associated with a context, and given a name in that context. Bijection provides a mechanism for aliasing internal component names (instance variables) to contextual names, allowing component trees to be dynamically assembled, and reassembled by Seam.

Let's start by describing the contexts built in to Seam.

Seam contexts are created and destroyed by the framework. The application does not control context demarcation via explicit Java API calls. Context are usually implicit. In some cases, however, contexts are demarcated via annotations.

The basic Seam contexts are:

You will recognize some of these contexts from servlet and related specifications. However, two of them might be new to you: conversation context, and business process context. One reason state management in web applications is so fragile and error-prone is that the three built-in contexts (request, session and application) are not especially meaningful from the point of view of the business logic. A user login session, for example, is a fairly arbitrary construct in terms of the actual application work flow. Therefore, most Seam components are scoped to the conversation or business process contexts, since they are the contexts which are most meaningful in terms of the application.

Let's look at each context in turn.

The conversation context is a truly central concept in Seam. A conversation is a unit of work from the point of view of the user. It might span several interactions with the user, several requests, and several database transactions. But to the user, a conversation solves a single problem. For example, "book hotel", "approve contract", "create order" are all conversations. You might like to think of a conversation implementing a single "use case" or "user story", but the relationship is not necessarily quite exact.

A conversation holds state associated with "what the user is doing now, in this window". A single user may have multiple conversations in progress at any point in time, usually in multiple windows. The conversation context allows us to ensure that state from the different conversations does not collide and cause bugs.

It might take you some time to get used to thinking of applications in terms of conversations. But once you get used to it, we think you'll love the notion, and never be able to not think in terms of conversations again!

Some conversations last for just a single request. Conversations that span multiple requests must be demarcated using annotations provided by Seam.

Some conversations are also tasks. A task is a conversation that is significant in terms of a long-running business process, and has the potential to trigger a business process state transition when it is successfully completed. Seam provides a special set of annotations for task demarcation.

Conversations may be nested, with one conversation taking place "inside" a wider conversation. This is an advanced feature.

Usually, conversation state is actually held by Seam in the servlet session between requests. Seam implements configurable conversation timeout, automatically destroying inactive conversations, and thus ensuring that the state held by a single user login session does not grow without bound if the user abandons conversations.

Seam serializes processing of concurrent requests that take place in the same long-running conversation context, in the same process.

Alternatively, Seam may be configured to keep conversational state in the client browser.

Neither the servlet nor EJB specifications define any facilities for managing concurrent requests originating from the same client. The servlet container simply lets all threads run concurrently and leaves enforcing threadsafeness to application code. The EJB container allows stateless components to be accessed concurrently, and throws an exception if multiple threads access a stateful session bean.

This behavior might have been okay in old-style web applications which were based around fine-grained, synchronous requests. But for modern applications which make heavy use of many fine-grained, asynchronous (AJAX) requests, concurrency is a fact of life, and must be supported by the programming model. Seam weaves a concurrency management layer into its context model.

The Seam session and application contexts are multithreaded. Seam will allow concurrent requests in a context to be processed concurrently. The event and page contexts are by nature single threaded. The business process context is strictly speaking multi-threaded, but in practice concurrency is sufficiently rare that this fact may be disregarded most of the time. Finally, Seam enforces a single thread per conversation per process model for the conversation context by serializing concurrent requests in the same long-running conversation context.

Since the session context is multithreaded, and often contains volatile state, session scope components are always protected by Seam from concurrent access. Seam serializes requests to session scope session beans and JavaBeans by default (and detects and breaks any deadlocks that occur). This is not the default behaviour for application scoped components however, since application scoped components do not usually hold volatile state and because synchronization at the global level is extremely expensive. However, you can force a serialized threading model on any session bean or JavaBean component by adding the @Synchronized annotation.

This concurrency model means that AJAX clients can safely use volatile session and conversational state, without the need for any special work on the part of the developer.

Seam components are POJOs (Plain Old Java Objects). In particular, they are JavaBeans or EJB 3.0 enterprise beans. While Seam does not require that components be EJBs and can even be used without an EJB 3.0 compliant container, Seam was designed with EJB 3.0 in mind and includes deep integration with EJB 3.0. Seam supports the following component types.

All seam components need a name. We can assign a name to a component using the @Name annotation:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    ... 
}

This name is the seam component name and is not related to any other name defined by the EJB specification. However, seam component names work just like JSF managed bean names and you can think of the two concepts as identical.

@Name is not the only way to define a component name, but we always need to specify the name somewhere. If we don't, then none of the other Seam annotations will function.

Just like in JSF, a seam component instance is usually bound to a context variable with the same name as the component name. So, for example, we would access the LoginAction using Contexts.getStatelessContext().get("loginAction"). In particular, whenever Seam itself instantiates a component, it binds the new instance to a variable with the component name. However, again like JSF, it is possible for the application to bind a component to some other context variable by programmatic API call. This is only useful if a particular component serves more than one role in the system. For example, the currently logged in User might be bound to the currentUser session context variable, while a User that is the subject of some administration functionality might be bound to the user conversation context variable.

For very large applications, and for built-in seam components, qualified names are often used.

@Name("com.jboss.myapp.loginAction")

@Stateless
public class LoginAction implements Login { 
    ... 
}

We may use the qualified component name both in Java code and in JSF's expression language:


<h:commandButton type="submit" value="Login"
                 action="#{com.jboss.myapp.loginAction.login}"/>

Since this is noisy, Seam also provides a means of aliasing a qualified name to a simple name. Add a line like this to the components.xml file:


<factory name="loginAction" scope="STATELESS" value="#{com.jboss.myapp.loginAction}"/>

All of the built-in Seam components have qualified names, but most of them are aliased to a simple name by the components.xml file included in the Seam jar.

Dependency injection or inversion of control is by now a familiar concept to most Java developers. Dependency injection allows a component to obtain a reference to another component by having the container "inject" the other component to a setter method or instance variable. In all dependency injection implementations that we have seen, injection occurs when the component is constructed, and the reference does not subsequently change for the lifetime of the component instance. For stateless components, this is reasonable. From the point of view of a client, all instances of a particular stateless component are interchangeable. On the other hand, Seam emphasizes the use of stateful components. So traditional dependency injection is no longer a very useful construct. Seam introduces the notion of bijection as a generalization of injection. In contrast to injection, bijection is:

In essence, bijection lets you alias a context variable to a component instance variable, by specifying that the value of the instance variable is injected, outjected, or both. Of course, we use annotations to enable bijection.

The @In annotation specifies that a value should be injected, either into an instance variable:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    @In User user;
    ... 
}

or into a setter method:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    User user;
    
    @In
    public void setUser(User user) {
        this.user=user;
    }
    
    ... 
}

By default, Seam will do a priority search of all contexts, using the name of the property or instance variable that is being injected. You may wish to specify the context variable name explicitly, using, for example, @In("currentUser").

If you want Seam to create an instance of the component when there is no existing component instance bound to the named context variable, you should specify @In(create=true). If the value is optional (it can be null), specify @In(required=false).

For some components, it can be repetitive to have to specify @In(create=true) everywhere they are used. In such cases, you can annotate the component @AutoCreate, and then it will always be created, whenever needed, even without the explicit use of create=true.

You can even inject the value of an expression:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    @In("#{user.username}") String username;
    ... 
}

Injected values are disinjected (i.e, set to null) immediately after method completion and outjection.

(There is much more information about component lifecycle and injection in the next chapter.)

The @Out annotation specifies that an attribute should be outjected, either from an instance variable:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    @Out User user;
    ... 
}

or from a getter method:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    User user;
    
    @Out
    public User getUser() {
        return user;
    }
    
    ... 
}

An attribute may be both injected and outjected:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    @In @Out User user;
    ... 
}

or:

@Name("loginAction")

@Stateless
public class LoginAction implements Login { 
    User user;
    
    @In
    public void setUser(User user) {
        this.user=user;
    }
    
    @Out
    public User getUser() {
        return user;
    }
    
    ... 
}

The @Install annotation lets you control conditional installation of components that are required in some deployment scenarios and not in others. This is useful if:

@Install works by letting you specify precedence and dependencies.

The precedence of a component is a number that Seam uses to decide which component to install when there are multiple classes with the same component name in the classpath. Seam will choose the component with the higher precendence. There are some predefined precedence values (in ascending order):

Suppose we have a component named messageSender that talks to a JMS queue.

@Name("messageSender") 

public class MessageSender {
    public void sendMessage() {
        //do something with JMS
    }
}

In our unit tests, we don't have a JMS queue available, so we would like to stub out this method. We'll create a mock component that exists in the classpath when unit tests are running, but is never deployed with the application:

@Name("messageSender") 

@Install(precedence=MOCK)
public class MockMessageSender extends MessageSender {
    public void sendMessage() {
        //do nothing!
    }
}

The precedence helps Seam decide which version to use when it finds both components in the classpath.

This is nice if we are able to control exactly which classes are in the classpath. But if I'm writing a reusable framework with many dependecies, I don't want to have to break that framework across many jars. I want to be able to decide which components to install depending upon what other components are installed, and upon what classes are available in the classpath. The @Install annotation also controls this functionality. Seam uses this mechanism internally to enable conditional installation of many of the built-in components. However, you probably won't need to use it in your application.

Who is not totally fed up with seeing noisy code like this?

private static final Log log = LogFactory.getLog(CreateOrderAction.class);

        
public Order createOrder(User user, Product product, int quantity) {
    if ( log.isDebugEnabled() ) {
        log.debug("Creating new order for user: " + user.username() + 
            " product: " + product.name() 
            + " quantity: " + quantity);
    }
    return new Order(user, product, quantity);
}

It is difficult to imagine how the code for a simple log message could possibly be more verbose. There is more lines of code tied up in logging than in the actual business logic! I remain totally astonished that the Java community has not come up with anything better in 10 years.

Seam provides a logging API that simplifies this code significantly:

@Logger private Log log;

        
public Order createOrder(User user, Product product, int quantity) {
    log.debug("Creating new order for user: #0 product: #1 quantity: #2", user.username(), product.name(), quantity);
    return new Order(user, product, quantity);
}

It doesn't matter if you declare the log variable static or not—it will work either way, except for entity bean components which require the log variable to be static.

Note that we don't need the noisy if ( log.isDebugEnabled() ) guard, since string concatenation happens inside the debug() method. Note also that we don't usually need to specify the log category explicitly, since Seam knows what component it is injecting the Log into.

If User and Product are Seam components available in the current contexts, it gets even better:

@Logger private Log log;

        
public Order createOrder(User user, Product product, int quantity) {
    log.debug("Creating new order for user: #{user.username} product: #{product.name} quantity: #0", quantity);
    return new Order(user, product, quantity);
}

Seam logging automagically chooses whether to send output to log4j or JDK logging. If log4j is in the classpath, Seam with use it. If it is not, Seam will use JDK logging.

Many application servers feature an amazingly broken implementation of HttpSession clustering, where changes to the state of mutable objects bound to the session are only replicated when the application calls setAttribute() explicitly. This is a source of bugs that can not effectively be tested for at development time, since they will only manifest when failover occurs. Furthermore, the actual replication message contains the entire serialized object graph bound to the session attribute, which is inefficient.

Of course, EJB stateful session beans must perform automatic dirty checking and replication of mutable state and a sophisticated EJB container can introduce optimizations such as attribute-level replication. Unfortunately, not all Seam users have the good fortune to be working in an environment that supports EJB 3.0. So, for session and conversation scoped JavaBean and entity bean components, Seam provides an extra layer of cluster-safe state management over the top of the web container session clustering.

For session or conversation scoped JavaBean components, Seam automatically forces replication to occur by calling setAttribute() once in every request that the component was invoked by the application. Of course, this strategy is inefficient for read-mostly components. You can control this behavior by implementing the org.jboss.seam.core.Mutable interface, or by extending org.jboss.seam.core.AbstractMutable, and writing your own dirty-checking logic inside the component. For example,

@Name("account")

public class Account extends AbstractMutable
{
    private BigDecimal balance;
    
    public void setBalance(BigDecimal balance)
    {
        setDirty(this.balance, balance);
        this.balance = balance;
    }
    
    public BigDecimal getBalance()
    {
        return balance;
    }
    
    ...
    
}

Or, you can use the @ReadOnly annotation to achieve a similar effect:

@Name("account")

public class Account
{
    private BigDecimal balance;
    
    public void setBalance(BigDecimal balance)
    {
        this.balance = balance;
    }
    
    @ReadOnly
    public BigDecimal getBalance()
    {
        return balance;
    }
    
    ...
    
}

For session or conversation scoped entity bean components, Seam automatically forces replication to occur by calling setAttribute() once in every request, unless the (conversation-scoped) entity is currently associated with a Seam-managed persistence context, in which case no replication is needed. This strategy is not necessarily efficient, so session or conversation scope entity beans should be used with care. You can always write a stateful session bean or JavaBean component to "manage" the entity bean instance. For example,

@Stateful

@Name("account")
public class AccountManager extends AbstractMutable
{
    private Account account; // an entity bean
    
    @Unwrap
    public void getAccount()
    {
        return account;
    }
    
    ...
    
}

Note that the EntityHome class in the Seam Application Framework provides a great example of managing an entity bean instance using a Seam component.

We often need to work with objects that are not Seam components. But we still want to be able to inject them into our components using @In and use them in value and method binding expressions, etc. Sometimes, we even need to tie them into the Seam context lifecycle (@Destroy, for example). So the Seam contexts can contain objects which are not Seam components, and Seam provides a couple of nice features that make it easier to work with non-component objects bound to contexts.

The factory component pattern lets a Seam component act as the instantiator for a non-component object. A factory method will be called when a context variable is referenced but has no value bound to it. We define factory methods using the @Factory annotation. The factory method binds a value to the context variable, and determines the scope of the bound value. There are two styles of factory method. The first style returns a value, which is bound to the context by Seam:

@Factory(scope=CONVERSATION)

public List<Customer> getCustomerList() { 
    return ... ;
} 

The second style is a method of type void which binds the value to the context variable itself:

@DataModel List<Customer> customerList;


@Factory("customerList")
public void initCustomerList() { 
    customerList = ...  ;
} 

In both cases, the factory method is called when we reference the customerList context variable and its value is null, and then has no further part to play in the lifecycle of the value. An even more powerful pattern is the manager component pattern. In this case, we have a Seam component that is bound to a context variable, that manages the value of the context variable, while remaining invisible to clients.

A manager component is any component with an @Unwrap method. This method returns the value that will be visable to clients, and is called every time a context variable is referenced.

@Name("customerList")

@Scope(CONVERSATION)
public class CustomerListManager
{
    ...
    
    @Unwrap
    public List<Customer> getCustomerList() { 
        return ... ;
    }
}

The manager component pattern is especially useful if we have an object where you need more control over the lifecycle of the component. For example, if you have a heavyweight object that needs a cleanup operation when the context ends you could @Unwrap the object, and perform cleanup in the @Destroy method of the manager component.

@Name("hens")

@Scope(APPLICATION) 
public class HenHouse {
    
    Set<Hen> hens;
    
    @In(required=false) Hen hen;
    
    @Unwrap
    public List<Hen> getHens() {
        if (hens == null) {
            // Setup our hens
        }
        return hens;
    }
    
    @Observer({"chickBorn", "chickenBoughtAtMarket"})
    public addHen() {
        hens.add(hen);
    }
    
    @Observer("chickenSoldAtMarket")
    public removeHen() {
        hens.remove(hen);
    }
    
    @Observer("foxGetsIn")
    public removeAllHens() {
        hens.clear();
    }
    ...
} 

Here the managed component observes many events which change the underlying object. The component manages these actions itself, and because the object is unwrapped on every access, a consistent view is provided.

The philosophy of minimizing XML-based configuration is extremely strong in Seam. Nevertheless, there are various reasons why we might want to configure a Seam component using XML: to isolate deployment-specific information from the Java code, to enable the creation of re-usable frameworks, to configure Seam's built-in functionality, etc. Seam provides two basic approaches to configuring components: configuration via property settings in a properties file or in web.xml, and configuration via components.xml.

The components.xml file is a bit more powerful than property settings. It lets you:

A components.xml file may appear in one of three different places:

Usually, Seam components are installed when the deployment scanner discovers a class with a @Name annotation sitting in an archive with a seam.properties file or a META-INF/components.xml file. (Unless the component has an @Install annotation indicating it should not be installed by default.) The components.xml file lets us handle special cases where we need to override the annotations.

For example, the following components.xml file installs jBPM:


<components xmlns="http://jboss.com/products/seam/components" 
            xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
            xmlns:bpm="http://jboss.com/products/seam/bpm">
    <bpm:jbpm/>
</components>

This example does the same thing:


<components>
    <component class="org.jboss.seam.bpm.Jbpm"/>
</components>

This one installs and configures two different Seam-managed persistence contexts:


<components xmlns="http://jboss.com/products/seam/components" 
            xmlns:persistence="http://jboss.com/products/seam/persistence"

    <persistence:managed-persistence-context name="customerDatabase"
                       persistence-unit-jndi-name="java:/customerEntityManagerFactory"/>
        
    <persistence:managed-persistence-context name="accountingDatabase"
                       persistence-unit-jndi-name="java:/accountingEntityManagerFactory"/>            

</components>

As does this one:


<components>
    <component name="customerDatabase" 
              class="org.jboss.seam.persistence.ManagedPersistenceContext">
        <property name="persistenceUnitJndiName">java:/customerEntityManagerFactory</property>
    </component>
    
    <component name="accountingDatabase"
              class="org.jboss.seam.persistence.ManagedPersistenceContext">
        <property name="persistenceUnitJndiName">java:/accountingEntityManagerFactory</property>
    </component>
</components>

This example creates a session-scoped Seam-managed persistence context (this is not recommended in practice):


<components xmlns="http://jboss.com/products/seam/components" 
            xmlns:persistence="http://jboss.com/products/seam/persistence"

  <persistence:managed-persistence-context name="productDatabase" 
                                          scope="session"
                     persistence-unit-jndi-name="java:/productEntityManagerFactory"/>        

</components>

<components>
            
    <component name="productDatabase"
              scope="session"
              class="org.jboss.seam.persistence.ManagedPersistenceContext">
        <property name="persistenceUnitJndiName">java:/productEntityManagerFactory</property>
    </component>

</components>

It is common to use the auto-create option for infrastructural objects like persistence contexts, which saves you from having to explicitly specify create=true when you use the @In annotation.


<components xmlns="http://jboss.com/products/seam/components" 
            xmlns:persistence="http://jboss.com/products/seam/persistence"

  <persistence:managed-persistence-context name="productDatabase" 
                                    auto-create="true"
                     persistence-unit-jndi-name="java:/productEntityManagerFactory"/>        

</components>

<components>
            
    <component name="productDatabase"
        auto-create="true"
              class="org.jboss.seam.persistence.ManagedPersistenceContext">
        <property name="persistenceUnitJndiName">java:/productEntityManagerFactory</property>
    </component>

</components>

The <factory> declaration lets you specify a value or method binding expression that will be evaluated to initialize the value of a context variable when it is first referenced.


<components>

    <factory name="contact" method="#{contactManager.loadContact}" scope="CONVERSATION"/>

</components>

You can create an "alias" (a second name) for a Seam component like so:


<components>

    <factory name="user" value="#{actor}" scope="STATELESS"/>

</components>

You can even create an "alias" for a commonly used expression:


<components>

    <factory name="contact" value="#{contactManager.contact}" scope="STATELESS"/>

</components>

It is especially common to see the use of auto-create="true" with the <factory> declaration:


<components>

    <factory name="session" value="#{entityManager.delegate}" scope="STATELESS" auto-create="true"/>

</components>

Sometimes we want to reuse the same components.xml file with minor changes during both deployment and testing. Seam lets you place wildcards of the form @wildcard@ in the components.xml file which can be replaced either by your Ant build script (at deployment time) or by providing a file named components.properties in the classpath (at development time). You'll see this approach used in the Seam examples.

Properties of string, primitive or primitive wrapper type may be configured just as you would expect:

org.jboss.seam.core.manager.conversationTimeout 60000

<core:manager conversation-timeout="60000"/>

<component name="org.jboss.seam.core.manager">
    <property name="conversationTimeout">60000</property>
</component>

Arrays, sets and lists of strings or primitives are also supported:

org.jboss.seam.bpm.jbpm.processDefinitions order.jpdl.xml, return.jpdl.xml, inventory.jpdl.xml

<bpm:jbpm>
    <bpm:process-definitions>
        <value>order.jpdl.xml</value>
        <value>return.jpdl.xml</value>
        <value>inventory.jpdl.xml</value>
    </bpm:process-definitions>
</bpm:jbpm>

<component name="org.jboss.seam.bpm.jbpm">
    <property name="processDefinitions">
        <value>order.jpdl.xml</value>
        <value>return.jpdl.xml</value>
        <value>inventory.jpdl.xml</value>
    </property>
</component>

Even maps with String-valued keys and string or primitive values are supported:


<component name="issueEditor">
    <property name="issueStatuses">
        <key>open</key> <value>open issue</value>
        <key>resolved</key> <value>issue resolved by developer</value>
        <key>closed</key> <value>resolution accepted by user</value>
    </property>
</component>

Finally, you may wire together components using a value-binding expression. Note that this is quite different to injection using @In, since it happens at component instantiation time instead of invocation time. It is therefore much more similar to the dependency injection facilities offered by traditional IoC containers like JSF or Spring.


<drools:managed-working-memory name="policyPricingWorkingMemory" rule-base="#{policyPricingRules}"/>

<component name="policyPricingWorkingMemory"
          class="org.jboss.seam.drools.ManagedWorkingMemory">
    <property name="ruleBase">#{policyPricingRules}</property>
</component>

Throughout the examples, there have been two competing ways of declaring components: with and without the use of XML namespaces. The following shows a typical components.xml file without namespaces:


<?xml version="1.0" encoding="UTF-8"?>
<components xmlns="http://jboss.com/products/seam/components"
            xsi:schemaLocation="http://jboss.com/products/seam/components http://jboss.com/products/seam/components-2.1.xsd">

    <component class="org.jboss.seam.core.init">
        <property name="debug">true</property>
        <property name="jndiPattern">@jndiPattern@</property>
    </component>
    
</components>

As you can see, this is somewhat verbose. Even worse, the component and attribute names cannot be validated at development time.

The namespaced version looks like this:


<?xml version="1.0" encoding="UTF-8"?>
<components xmlns="http://jboss.com/products/seam/components"
            xmlns:core="http://jboss.com/products/seam/core"
            xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
            xsi:schemaLocation=
                "http://jboss.com/products/seam/core http://jboss.com/products/seam/core-2.1.xsd 
                 http://jboss.com/products/seam/components http://jboss.com/products/seam/components-2.1.xsd">

    <core:init debug="true" jndi-pattern="@jndiPattern@"/>

</components>

Even though the schema declarations are verbose, the actual XML content is lean and easy to understand. The schemas provide detailed information about each component and the attributes available, allowing XML editors to offer intelligent autocomplete. The use of namespaced elements makes generating and maintaining correct components.xml files much simpler.

Now, this works great for the built-in Seam components, but what about user components? There are two options. First, Seam supports mixing the two models, allowing the use of the generic <component> declarations for user components, along with namespaced declarations for built-in components. But even better, Seam allows you to quickly declare namespaces for your own components.

Any Java package can be associated with an XML namespace by annotating the package with the @Namespace annotation. (Package-level annotations are declared in a file named package-info.java in the package directory.) Here is an example from the seampay demo:

@Namespace(value="http://jboss.com/products/seam/examples/seampay")

package org.jboss.seam.example.seampay;
import org.jboss.seam.annotations.Namespace;

That is all you need to do to use the namespaced style in components.xml! Now we can write:


<components xmlns="http://jboss.com/products/seam/components"
            xmlns:pay="http://jboss.com/products/seam/examples/seampay"
            ... >

    <pay:payment-home new-instance="#{newPayment}"
                      created-message="Created a new payment to #{newPayment.payee}" />

    <pay:payment name="newPayment"
                 payee="Somebody"
                 account="#{selectedAccount}"
                 payment-date="#{currentDatetime}"
                 created-date="#{currentDatetime}" />
     ...
</components>

Or:


<components xmlns="http://jboss.com/products/seam/components"
            xmlns:pay="http://jboss.com/products/seam/examples/seampay"
            ... >

    <pay:payment-home>
        <pay:new-instance>"#{newPayment}"</pay:new-instance>
        <pay:created-message>Created a new payment to #{newPayment.payee}</pay:created-message>
    </pay:payment-home>
    
    <pay:payment name="newPayment">
        <pay:payee>Somebody"</pay:payee>
        <pay:account>#{selectedAccount}</pay:account>
        <pay:payment-date>#{currentDatetime}</pay:payment-date>
        <pay:created-date>#{currentDatetime}</pay:created-date>
     </pay:payment>
     ...
</components>

These examples illustrate the two usage models of a namespaced element. In the first declaration, the <pay:payment-home> references the paymentHome component:

package org.jboss.seam.example.seampay;

...
@Name("paymentHome")
public class PaymentController
    extends EntityHome<Payment>
{
    ... 
}

The element name is the hyphenated form of the component name. The attributes of the element are the hyphenated form of the property names.

In the second declaration, the <pay:payment> element refers to the Payment class in the org.jboss.seam.example.seampay package. In this case Payment is an entity that is being declared as a Seam component:

package org.jboss.seam.example.seampay;

...
@Entity
public class Payment
    implements Serializable
{
    ...
}

If we want validation and autocompletion to work for user-defined components, we will need a schema. Seam does not yet provide a mechanism to automatically generate a schema for a set of components, so it is necessary to generate one manually. The schema definitions for the standard Seam packages can be used for guidance.

The following are the the namespaces used by Seam:

Complementing the contextual component model, there are two further basic concepts that facilitate the extreme loose-coupling that is the distinctive feature of Seam applications. The first is a strong event model where events may be mapped to event listeners via JSF-like method binding expressions. The second is the pervasive use of annotations and interceptors to apply cross-cutting concerns to components which implement business logic.

A Seam page action is an event that occurs just before we render a page. We declare page actions in WEB-INF/pages.xml. We can define a page action for either a particular JSF view id:


<pages>
    <page view-id="/hello.jsp" action="#{helloWorld.sayHello}"/>
</pages>

Or we can use a * wildcard as a suffix to the view-id to specify an action that applies to all view ids that match the pattern:


<pages>
    <page view-id="/hello/*" action="#{helloWorld.sayHello}"/>
</pages>

If multiple wildcarded page actions match the current view-id, Seam will call all the actions, in order of least-specific to most-specific.

The page action method can return a JSF outcome. If the outcome is non-null, Seam will use the defined navigation rules to navigate to a view.

Furthermore, the view id mentioned in the <page> element need not correspond to a real JSP or Facelets page! So, we can reproduce the functionality of a traditional action-oriented framework like Struts or WebWork using page actions. For example:

TODO: translate struts action into page action

This is quite useful if you want to do complex things in response to non-faces requests (for example, HTTP GET requests).

Multiple or conditional page actions my be specified using the <action> tag:


<pages>
    <page view-id="/hello.jsp">
        <action execute="#{helloWorld.sayHello}" if="#{not validation.failed}"/>
        <action execute="#{hitCount.increment}"/>
    </page>
</pages>

A JSF faces request (a form submission) encapsulates both an "action" (a method binding) and "parameters" (input value bindings). A page action might also needs parameters!

Since GET requests are bookmarkable, page parameters are passed as human-readable request parameters. (Unlike JSF form inputs, which are anything but!)

You can use page parameters with or without an action method.

Seam lets us provide a value binding that maps a named request parameter to an attribute of a model object.


<pages>
      <page view-id="/hello.jsp" action="#{helloWorld.sayHello}">
          <param name="firstName" value="#{person.firstName}"/>
          <param name="lastName" value="#{person.lastName}"/>
      </page>
  </pages>

The <param> declaration is bidirectional, just like a value binding for a JSF input:

The essential idea behind all this is that however we get from any other page to /hello.jsp (or from /hello.jsp back to /hello.jsp), the value of the model attribute referred to in the value binding is "remembered", without the need for a conversation (or other server-side state).

If just the name attribute is specified then the request parameter is propagated using the PAGE context (it isn't mapped to model property).


<pages>
      <page view-id="/hello.jsp" action="#{helloWorld.sayHello}">
          <param name="firstName" />
          <param name="lastName" />
      </page>
  </pages>

Propagation of page parameters is especially useful if you want to build multi-layer master-detail CRUD pages. You can use it to "remember" which view you were previously on (e.g. when pressing the Save button), and which entity you were editing.

This all sounds pretty complex, and you're probably wondering if such an exotic construct is really worth the effort. Actually, the idea is very natural once you "get it". It is definitely worth taking the time to understand this stuff. Page parameters are the most elegant way to propagate state across a non-faces request. They are especially cool for problems like search screens with bookmarkable results pages, where we would like to be able to write our application code to handle both POST and GET requests with the same code. Page parameters eliminate repetitive listing of request parameters in the view definition and make redirects much easier to code.

You can specify a JSF converter for complex model propreties:


<pages>
   <page view-id="/calculator.jsp" action="#{calculator.calculate}">
      <param name="x" value="#{calculator.lhs}"/>
      <param name="y" value="#{calculator.rhs}"/>
      <param name="op" converterId="com.my.calculator.OperatorConverter" value="#{calculator.op}"/>
   </page>
</pages>

Alternatively:


<pages>
   <page view-id="/calculator.jsp" action="#{calculator.calculate}">
      <param name="x" value="#{calculator.lhs}"/>
      <param name="y" value="#{calculator.rhs}"/>
      <param name="op" converter="#{operatorConverter}" value="#{calculator.op}"/>
   </page>
</pages>

JSF validators, and required="true" may also be used:


<pages>
    <page view-id="/blog.xhtml">
        <param name="date" 
               value="#{blog.date}" 
               validatorId="com.my.blog.PastDate" 
               required="true"/>
    </page>
</pages>

Alternatively:


<pages>
    <page view-id="/blog.xhtml">
        <param name="date" 
               value="#{blog.date}" 
               validator="#{pastDateValidator}" 
               required="true"/>
    </page>
</pages>

Even better, model-based Hibernate validator annotations are automatically recognized and validated.

When type conversion or validation fails, a global FacesMessage is added to the FacesContext.

You can use standard JSF navigation rules defined in faces-config.xml in a Seam application. However, JSF navigation rules have a number of annoying limitations:

A further problem is that "orchestration" logic gets scattered between pages.xml and faces-config.xml. It's better to unify this logic into pages.xml.

This JSF navigation rule:


<navigation-rule>
   <from-view-id>/editDocument.xhtml</from-view-id>
    
   <navigation-case>
      <from-action>#{documentEditor.update}</from-action>
      <from-outcome>success</from-outcome>
      <to-view-id>/viewDocument.xhtml</to-view-id>
      <redirect/>
   </navigation-case>
    
</navigation-rule>

Can be rewritten as follows:


<page view-id="/editDocument.xhtml">
    
    <navigation from-action="#{documentEditor.update}">
        <rule if-outcome="success">
            <redirect view-id="/viewDocument.xhtml"/>
        </rule>
    </navigation>
    
</page>

But it would be even nicer if we didn't have to pollute our DocumentEditor component with string-valued return values (the JSF outcomes). So Seam lets us write:


<page view-id="/editDocument.xhtml">
    
    <navigation from-action="#{documentEditor.update}" 
                   evaluate="#{documentEditor.errors.size}">
        <rule if-outcome="0">
            <redirect view-id="/viewDocument.xhtml"/>
        </rule>
    </navigation>
    
</page>

Or even:


<page view-id="/editDocument.xhtml">
    
    <navigation from-action="#{documentEditor.update}">
        <rule if="#{documentEditor.errors.empty}">
            <redirect view-id="/viewDocument.xhtml"/>
        </rule>
    </navigation>
    
</page>

The first form evaluates a value binding to determine the outcome value to be used by the subsequent rules. The second approach ignores the outcome and evaluates a value binding for each possible rule.

Of course, when an update succeeds, we probably want to end the current conversation. We can do that like this:


<page view-id="/editDocument.xhtml">
    
    <navigation from-action="#{documentEditor.update}">
        <rule if="#{documentEditor.errors.empty}">
            <end-conversation/>
            <redirect view-id="/viewDocument.xhtml"/>
        </rule>
    </navigation>
    
</page>

As we've ended conversation any subsequent requests won't know which document we are interested in. We can pass the document id as a request parameter which also makes the view bookmarkable:


<page view-id="/editDocument.xhtml">
    
    <navigation from-action="#{documentEditor.update}">
        <rule if="#{documentEditor.errors.empty}">
            <end-conversation/>
            <redirect view-id="/viewDocument.xhtml">
                <param name="documentId" value="#{documentEditor.documentId}"/>
            </redirect>
        </rule>
    </navigation>
    
</page>

Null outcomes are a special case in JSF. The null outcome is interpreted to mean "redisplay the page". The following navigation rule matches any non-null outcome, but not the null outcome:


<page view-id="/editDocument.xhtml">
    
    <navigation from-action="#{documentEditor.update}">
        <rule>
            <render view-id="/viewDocument.xhtml"/>
        </rule>
    </navigation>
    
</page>

If you want to perform navigation when a null outcome occurs, use the following form instead:


<page view-id="/editDocument.xhtml">
    
    <navigation from-action="#{documentEditor.update}">
        <render view-id="/viewDocument.xhtml"/>
    </navigation>
    
</page>

The view-id may be given as a JSF EL expression:


<page view-id="/editDocument.xhtml">

    <navigation>
        <rule if-outcome="success">
            <redirect view-id="/#{userAgent}/displayDocument.xhtml"/>
        </rule>
    </navigation>
    
</page>

Seam components can interact by simply calling each others methods. Stateful components may even implement the observer/observable pattern. But to enable components to interact in a more loosely-coupled fashion than is possible when the components call each others methods directly, Seam provides component-driven events.

We specify event listeners (observers) in components.xml.


<components>
    <event type="hello">
        <action execute="#{helloListener.sayHelloBack}"/>
        <action execute="#{logger.logHello}"/>
    </event>
</components>

Where the event type is just an arbitrary string.

When an event occurs, the actions registered for that event will be called in the order they appear in components.xml. How does a component raise an event? Seam provides a built-in component for this.

@Name("helloWorld")

public class HelloWorld {
    public void sayHello() {
        FacesMessages.instance().add("Hello World!");
        Events.instance().raiseEvent("hello");
    }
}

Or you can use an annotation.

@Name("helloWorld")

public class HelloWorld {
    @RaiseEvent("hello")
    public void sayHello() {
        FacesMessages.instance().add("Hello World!");
    }
}

Notice that this event producer has no dependency upon event consumers. The event listener may now be implemented with absolutely no dependency upon the producer:

@Name("helloListener")

public class HelloListener {
    public void sayHelloBack() {
        FacesMessages.instance().add("Hello to you too!");
    }
}

The method binding defined in components.xml above takes care of mapping the event to the consumer. If you don't like futzing about in the components.xml file, you can use an annotation instead:

@Name("helloListener")

public class HelloListener {
    @Observer("hello")
    public void sayHelloBack() {
        FacesMessages.instance().add("Hello to you too!");
    }
}

You might wonder why I've not mentioned anything about event objects in this discussion. In Seam, there is no need for an event object to propagate state between event producer and listener. State is held in the Seam contexts, and is shared between components. However, if you really want to pass an event object, you can:

@Name("helloWorld")

public class HelloWorld {
    private String name;
    public void sayHello() {
        FacesMessages.instance().add("Hello World, my name is #0.", name);
        Events.instance().raiseEvent("hello", name);
    }
}
@Name("helloListener")

public class HelloListener {
    @Observer("hello")
    public void sayHelloBack(String name) {
        FacesMessages.instance().add("Hello #0!", name);
    }
}

Seam defines a number of built-in events that the application can use to perform special kinds of framework integration. The events are:

Seam components may observe any of these events in just the same way they observe any other component-driven events.

EJB 3.0 introduced a standard interceptor model for session bean components. To add an interceptor to a bean, you need to write a class with a method annotated @AroundInvoke and annotate the bean with an @Interceptors annotation that specifies the name of the interceptor class. For example, the following interceptor checks that the user is logged in before allowing invoking an action listener method:

public class LoggedInInterceptor {


   @AroundInvoke
   public Object checkLoggedIn(InvocationContext invocation) throws Exception {
   
      boolean isLoggedIn = Contexts.getSessionContext().get("loggedIn")!=null;
      if (isLoggedIn) {
         //the user is already logged in
         return invocation.proceed();
      }
      else {
         //the user is not logged in, fwd to login page
         return "login";
      }
   }
}

To apply this interceptor to a session bean which acts as an action listener, we must annotate the session bean @Interceptors(LoggedInInterceptor.class). This is a somewhat ugly annotation. Seam builds upon the interceptor framework in EJB3 by allowing you to use @Interceptors as a meta-annotation for class level interceptors (those annotated @Target(TYPE)). In our example, we would create an @LoggedIn annotation, as follows:

@Target(TYPE)

@Retention(RUNTIME)
@Interceptors(LoggedInInterceptor.class)
public @interface LoggedIn {}

We can now simply annotate our action listener bean with @LoggedIn to apply the interceptor.

@Stateless

@Name("changePasswordAction")
@LoggedIn
@Interceptors(SeamInterceptor.class)
public class ChangePasswordAction implements ChangePassword { 
    
    ...
    
    public String changePassword() { ... }
    
}

If interceptor ordering is important (it usually is), you can add @Interceptor annotations to your interceptor classes to specify a partial order of interceptors.

@Interceptor(around={BijectionInterceptor.class,

                     ValidationInterceptor.class,
                     ConversationInterceptor.class},
             within=RemoveInterceptor.class)
public class LoggedInInterceptor
{
    ...
}

You can even have a "client-side" interceptor, that runs around any of the built-in functionality of EJB3:

@Interceptor(type=CLIENT)

public class LoggedInInterceptor
{
    ...
}

EJB interceptors are stateful, with a lifecycle that is the same as the component they intercept. For interceptors which do not need to maintain state, Seam lets you get a performance optimization by specifying @Interceptor(stateless=true).

Much of the functionality of Seam is implemented as a set of built-in Seam interceptors, including the interceptors named in the previous example. You don't have to explicitly specify these interceptors by annotating your components; they exist for all interceptable Seam components.

You can even use Seam interceptors with JavaBean components, not just EJB3 beans!

EJB defines interception not only for business methods (using @AroundInvoke), but also for the lifecycle methods @PostConstruct, @PreDestroy, @PrePassivate and @PostActive. Seam supports all these lifecycle methods on both component and interceptor not only for EJB3 beans, but also for JavaBean components (except @PreDestroy which is not meaningful for JavaBean components).

JSF is surprisingly limited when it comes to exception handling. As a partial workaround for this problem, Seam lets you define how a particular class of exception is to be treated by annotating the exception class, or declaring the exception class in an XML file. This facility is meant to be combined with the EJB 3.0-standard @ApplicationException annotation which specifies whether the exception should cause a transaction rollback.

Since we can't add annotations to all the exception classes we are interested in, Seam also lets us specify this functionality in pages.xml.


<pages>
   
   <exception class="javax.persistence.EntityNotFoundException">
      <http-error error-code="404"/>
   </exception>
   
   <exception class="javax.persistence.PersistenceException">
      <end-conversation/>
      <redirect view-id="/error.xhtml">
          <message>Database access failed</message>
      </redirect>
   </exception>
   
   <exception>
      <end-conversation/>
      <redirect view-id="/error.xhtml">
          <message>Unexpected failure</message>
      </redirect>
   </exception>
   
</pages>

The last <exception> declaration does not specify a class, and is a catch-all for any exception for which handling is not otherwise specified via annotations or in pages.xml.

You can also use EL to specify the view-id to redirect to.

You can also access the handled exception instance through EL, Seam places it in the conversation context, e.g. to access the message of the exception:


...
throw new AuthorizationException("You are not allowed to do this!");

<pages>

    <exception class="org.jboss.seam.security.AuthorizationException">
        <end-conversation/>
        <redirect view-id="/error.xhtml">
            <message severity="WARN">#{org.jboss.seam.handledException.message}</message>
        </redirect>
    </exception>

</pages>

org.jboss.seam.handledException holds the nested exception that was actually handled by an exception handler. The outermost (wrapper) exception is also available, as org.jboss.seam.exception.

If you are using JPA:


<exception class="javax.persistence.EntityNotFoundException">
   <redirect view-id="/error.xhtml">
      <message>Not found</message>
   </redirect>
</exception>

<exception class="javax.persistence.OptimisticLockException">
   <end-conversation/>
   <redirect view-id="/error.xhtml">
      <message>Another user changed the same data, please try again</message>
   </redirect>
</exception>

If you are using the Seam Application Framework:


<exception class="org.jboss.seam.framework.EntityNotFoundException">
   <redirect view-id="/error.xhtml">
      <message>Not found</message>
   </redirect>
</exception>

If you are using Seam Security:


<exception class="org.jboss.seam.security.AuthorizationException">
   <redirect>
      <message>You don't have permission to do this</message>
   </redirect>
</exception>
    
<exception class="org.jboss.seam.security.NotLoggedInException">
   <redirect view-id="/login.xhtml">
      <message>Please log in first</message>
   </redirect>
</exception>

And, for JSF:


<exception class="javax.faces.application.ViewExpiredException">
   <redirect view-id="/error.xhtml">
      <message>Your session has timed out, please try again</message>
   </redirect>
</exception>

A ViewExpiredException occurs if the user posts back to a page once their session has expired. no-conversation-view-id and conversation-required give you finer grained control over session expiration if you are inside a conversation.

It's time to understand Seam's conversation model in more detail.

Historically, the notion of a Seam "conversation" came about as a merger of three different ideas:

  • The idea of a workspace, which I encountered in a project for the Victorian government in 2002. In this project I was forced to implement workspace management on top of Struts, an experience I pray never to repeat.

  • The idea of an application transaction with optimistic semantics, and the realization that existing frameworks based around a stateless architecture could not provide effective management of extended persistence contexts. (The Hibernate team is truly fed up with copping the blame for LazyInitializationExceptions, which are not really Hibernate's fault, but rather the fault of the extremely limiting persistence context model supported by stateless architectures such as the Spring framework or the traditional stateless session facade (anti)pattern in J2EE.)

  • The idea of a workflow task.

By unifying these ideas and providing deep support in the framework, we have a powerful construct that lets us build richer and more efficient applications with less code than before.

The examples we have seen so far make use of a very simple conversation model that follows these rules:

Seam transparently propagates the conversation context (including the temporary conversation context) across JSF postbacks and redirects. If you don't do anything special, a non-faces request (a GET request for example) will not propagate the conversation context and will be processed in a new temporary conversation. This is usually - but not always - the desired behavior.

If you want to propagate a Seam conversation across a non-faces request, you need to explicitly code the Seam conversation id as a request parameter:


<a href="main.jsf?conversationId=#{conversation.id}">Continue</a>

Or, the more JSF-ish:


<h:outputLink value="main.jsf">
    <f:param name="conversationId" value="#{conversation.id}"/>
    <h:outputText value="Continue"/>
</h:outputLink>

If you use the Seam tag library, this is equivalent:


<h:outputLink value="main.jsf">
    <s:conversationId/>
    <h:outputText value="Continue"/>
</h:outputLink>

If you wish to disable propagation of the conversation context for a postback, a similar trick is used:


<h:commandLink action="main" value="Exit">
    <f:param name="conversationPropagation" value="none"/>
</h:commandLink>

If you use the Seam tag library, this is equivalent:


<h:commandLink action="main" value="Exit">
    <s:conversationPropagation type="none"/>
</h:commandLink>

Note that disabling conversation context propagation is absolutely not the same thing as ending the conversation.

The conversationPropagation request parameter, or the <s:conversationPropagation> tag may even be used to begin and end conversation, or begin a nested conversation.


<h:commandLink action="main" value="Exit">
    <s:conversationPropagation type="end"/>
</h:commandLink>

<h:commandLink action="main" value="Select Child">
    <s:conversationPropagation type="nested"/>
</h:commandLink>

<h:commandLink action="main" value="Select Hotel">
    <s:conversationPropagation type="begin"/>
</h:commandLink>

<h:commandLink action="main" value="Select Hotel">
    <s:conversationPropagation type="join"/>
</h:commandLink>

This conversation model makes it easy to build applications which behave correctly with respect to multi-window operation. For many applications, this is all that is needed. Some complex applications have either or both of the following additional requirements:

A nested conversation is created by invoking a method marked @Begin(nested=true) inside the scope of an existing conversation. A nested conversation has its own conversation context, and also has read-only access to the context of the outer conversation. (It can read the outer conversation's context variables, but not write to them.) When an @End is subsequently encountered, the nested conversation will be destroyed, and the outer conversation will resume, by "popping" the conversation stack. Conversations may be nested to any arbitrary depth.

Certain user activity (workspace management, or the back button) can cause the outer conversation to be resumed before the inner conversation is ended. In this case it is possible to have multiple concurrent nested conversations belonging to the same outer conversation. If the outer conversation ends before a nested conversation ends, Seam destroys all nested conversation contexts along with the outer context.

A conversation may be thought of as a continuable state. Nested conversations allow the application to capture a consistent continuable state at various points in a user interaction, thus insuring truly correct behavior in the face of backbuttoning and workspace management.

TODO: an example to show how a nested conversation prevents bad stuff happening when you backbutton.

Usually, if a component exists in a parent conversation of the current nested conversation, the nested conversation will use the same instance. Occasionally, it is useful to have a different instance in each nested conversation, so that the component instance that exists in the parent conversation is invisible to its child conversations. You can achieve this behavior by annotating the component @PerNestedConversation.

JSF does not define any kind of action listener that is triggered when a page is accessed via a non-faces request (for example, a HTTP GET request). This can occur if the user bookmarks the page, or if we navigate to the page via an <h:outputLink>.

Sometimes we want to begin a conversation immediately the page is accessed. Since there is no JSF action method, we can't solve the problem in the usual way, by annotating the action with @Begin.

A further problem arises if the page needs some state to be fetched into a context variable. We've already seen two ways to solve this problem. If that state is held in a Seam component, we can fetch the state in a @Create method. If not, we can define a @Factory method for the context variable.

If none of these options works for you, Seam lets you define a page action in the pages.xml file.


<pages>
    <page view-id="/messageList.jsp" action="#{messageManager.list}"/>
    ...
</pages>

This action method is called at the beginning of the render response phase, any time the page is about to be rendered. If a page action returns a non-null outcome, Seam will process any appropriate JSF and Seam navigation rules, possibly resulting in a completely different page being rendered.

If all you want to do before rendering the page is begin a conversation, you could use a built-in action method that does just that:


<pages>
    <page view-id="/messageList.jsp" action="#{conversation.begin}"/>
    ...
</pages>

Note that you can also call this built-in action from a JSF control, and, similarly, you can use #{conversation.end} to end conversations.

If you want more control, to join existing conversations or begin a nested conversion, to begin a pageflow or an atomic conversation, you should use the <begin-conversation> element.


<pages>
    <page view-id="/messageList.jsp">
       <begin-conversation nested="true" pageflow="AddItem"/>
    <page>
    ...
</pages>

There is also an <end-conversation> element.


<pages>
    <page view-id="/home.jsp">
       <end-conversation/>
    <page>
    ...
</pages>

To solve the first problem, we now have five options:

JSF command links always perform a form submission via JavaScript, which breaks the web browser's "open in new window" or "open in new tab" feature. In plain JSF, you need to use an <h:outputLink> if you need this functionality. But there are two major limitations to <h:outputLink>.

Seam provides the notion of a page action to help solve the first problem, but this does nothing to help us with the second problem. We could work around this by using the RESTful approach of passing a request parameter and requerying for the selected object on the server side. In some cases—such as the Seam blog example application—this is indeed the best approach. The RESTful style supports bookmarking, since it does not require server-side state. In other cases, where we don't care about bookmarks, the use of @DataModel and @DataModelSelection is just so convenient and transparent!

To fill in this missing functionality, and to make conversation propagation even simpler to manage, Seam provides the <s:link> JSF tag.

The link may specify just the JSF view id:


<s:link view="/login.xhtml" value="Login"/>

Or, it may specify an action method (in which case the action outcome determines the page that results):


<s:link action="#{login.logout}" value="Logout"/>

If you specify both a JSF view id and an action method, the 'view' will be used unless the action method returns a non-null outcome:


<s:link view="/loggedOut.xhtml"  action="#{login.logout}" value="Logout"/>

The link automatically propagates the selected row of a DataModel using inside <h:dataTable>:


<s:link view="/hotel.xhtml" action="#{hotelSearch.selectHotel}" value="#{hotel.name}"/>

You can leave the scope of an existing conversation:


<s:link view="/main.xhtml" propagation="none"/>

You can begin, end, or nest conversations:


<s:link action="#{issueEditor.viewComment}" propagation="nest"/>

If the link begins a conversation, you can even specify a pageflow to be used:


<s:link action="#{documentEditor.getDocument}" propagation="begin"
        pageflow="EditDocument"/>

The taskInstance attribute if for use in jBPM task lists:


<s:link action="#{documentApproval.approveOrReject}" taskInstance="#{task}"/>

(See the DVD Store demo application for examples of this.)

Finally, if you need the "link" to be rendered as a button, use <s:button>:


<s:button action="#{login.logout}" value="Logout"/>

When working with conversations that deal with persistent objects, it may be desirable to use the natural business key of the object instead of the standard, "surrogate" conversation id:

Easy redirect to existing conversation

It can be useful to redirect to an existing conversation if the user requests the same operation twice. Take this example: “ You are on ebay, half way through paying for an item you just won as a Christmas present for your parents. Lets say you're sending it straight to them - you enter your payment details but you can't remember their address. You accidentally reuse the same browser window finding out their address. Now you need to return to the payment for the item.

With a natural conversation its really easy to have the user rejoin the existing conversation, and pick up where they left off - just have them to rejoin the payForItem conversation with the itemId as the conversation id.

User friendly URLs

For me this consists of a navigable hierarchy (I can navigate by editing the url) and a meaningful URL (like this Wiki uses - so don't identify things by random ids). For some applications user friendly URLs are less important, of course.

With a natural conversations, when you are building your hotel booking system (or, of course, whatever your app is) you can generate a URL like http://seam-hotels/book.seam?hotel=BestWesternAntwerpen (of course, whatever parameter hotel maps to on your domain model must be unique) and with URLRewrite easily transform this to http://seam-hotels/book/BestWesternAntwerpen.

Much better!

Natural conversations are defined in pages.xml:


  <conversation name="PlaceBid"
                  parameter-name="auctionId"
                  parameter-value="#{auction.auctionId}"/>

The first thing to note from the above definition is that the conversation has a name, in this case PlaceBid. This name uniquely identifies this particular named conversation, and is used by the page definition to identify a named conversation to participate in.

The next attribute, parameter-name defines the request parameter that will contain the natural conversation id, in place of the default conversation id parameter. In this example, the parameter-name is auctionId. This means that instead of a conversation parameter like cid=123 appearing in the URL for your page, it will contain auctionId=765432 instead.

The last attribute in the above configuration, parameter-value, defines an EL expression used to evaluate the value of the natural business key to use as the conversation id. In this example, the conversation id will be the primary key value of the auction instance currently in scope.

Next, we define which pages will participate in the named conversation. This is done by specifying the conversation attribute for a page definition:


  <page view-id="/bid.xhtml" conversation="PlaceBid" login-required="true">
      <navigation from-action="#{bidAction.confirmBid}">        
          <rule if-outcome="success">
              <redirect view-id="/auction.xhtml">
                  <param name="id" value="#{bidAction.bid.auction.auctionId}"/>
              </redirect>
          </rule>        
      </navigation>
  </page>

When starting, or redirecting to, a natural conversation there are a number of options for specifying the natural conversation name. Let's start by looking at the following page definition:


  <page view-id="/auction.xhtml">
    <param name="id" value="#{auctionDetail.selectedAuctionId}"/>
       
    <navigation from-action="#{bidAction.placeBid}">
      <redirect view-id="/bid.xhtml"/>
    </navigation>
  </page>

From here, we can see that invoking the action #{bidAction.placeBid} from our auction view (by the way, all these examples are taken from the seamBay example in Seam), that we will be redirected to /bid.xhtml, which, as we saw previously, is configured with the natural conversation PlaceBid. The declaration for our action method looks like this:

   @Begin(join = true)

   public void placeBid()

When named conversations are specified in the <page/> element, redirection to the named conversation occurs as part of navigation rules, after the action method has already been invoked. This is a problem when redirecting to an existing conversation, as redirection needs to be occur before the action method is invoked. Therefore it is necessary to specify the conversation name when the action is invoked. One way of doing this is by using the s:conversationName tag:


  <h:commandButton id="placeBidWithAmount" styleClass="placeBid" action="#{bidAction.placeBid}">
    <s:conversationName value="PlaceBid"/>
  </h:commandButton>

Another alternative is to specify the conversationName attribute when using either s:link or s:button:


  <s:link value="Place Bid" action="#{bidAction.placeBid}" conversationName="PlaceBid"/>

Workspace management is the ability to "switch" conversations in a single window. Seam makes workspace management completely transparent at the level of the Java code. To enable workspace management, all you need to do is:

The conversation list is very similar to the conversation switcher, except that it is displayed as a table:


<h:dataTable value="#{conversationList}" var="entry"
        rendered="#{not empty conversationList}">
    <h:column>
        <f:facet name="header">Workspace</f:facet>
        <h:commandLink action="#{entry.select}" value="#{entry.description}"/>
        <h:outputText value="[current]" rendered="#{entry.current}"/>
    </h:column>
    <h:column>
        <f:facet name="header">Activity</f:facet>
        <h:outputText value="#{entry.startDatetime}">
            <f:convertDateTime type="time" pattern="hh:mm a"/>
        </h:outputText>
        <h:outputText value=" - "/>
        <h:outputText value="#{entry.lastDatetime}">
            <f:convertDateTime type="time" pattern="hh:mm a"/>
        </h:outputText>
    </h:column>
    <h:column>
        <f:facet name="header">Action</f:facet>
        <h:commandButton action="#{entry.select}" value="#{msg.Switch}"/>
        <h:commandButton action="#{entry.destroy}" value="#{msg.Destroy}"/>
    </h:column>
</h:dataTable>

We imagine that you will want to customize this for your own application.

Only conversations with a description will be included in the list.

Notice that the conversation list lets the user destroy workspaces.

Conversational components have one minor limitation: they cannot be used to hold bindings to JSF components. (We generally prefer not to use this feature of JSF unless absolutely necessary, since it creates a hard dependency from application logic to the view.) On a postback request, component bindings are updated during the Restore View phase, before the Seam conversation context has been restored.

To work around this use an event scoped component to store the component bindings and inject it into the conversation scoped component that requires it.

@Name("grid")

@Scope(ScopeType.EVENT)
public class Grid
{
    private HtmlPanelGrid htmlPanelGrid;
    // getters and setters
    ...
}
@Name("gridEditor")

@Scope(ScopeType.CONVERSATION)
public class GridEditor
{
    @In(required=false)
    private Grid grid;
    
    ...
}

Also, you can't inject a conversation scoped component into an event scoped component which you bind a JSF control to. This includes Seam built in components like facesMessages.

Alternatively, you can access the JSF component tree through the implicit uiComponent handle. The following example accesses getRowIndex()of the UIData component which backs the data table during iteration, it prints the current row number:



<h:dataTable id="lineItemTable" var="lineItem" value="#{orderHome.lineItems}">
   <h:column>
      Row: #{uiComponent['lineItemTable'].rowIndex}
   </h:column>
   ...
</h:dataTable>

JSF UI components are available with their client identifier in this map.

A general discussion of concurrent calls to Seam components can be found in Section 4.1.10, “Concurrency model”. Here we will discuss the most common situation in which you will encounter concurrency — accessing conversational components from AJAX requests. We're going to discuss the options that a Ajax client library should provide to control events originating at the client — and we'll look at the options RichFaces gives you.

Conversational components don't allow real concurrent access therefore Seam queues each request to process them serially. This allows each request to be executed in a deterministic fashion. However, a simple queue isn't that great — firstly, if a method is, for some reason, taking a very long time to complete, running it over and over again whenever the client generates a request is bad idea (potential for Denial of Service attacks), and, secondly, AJAX is often to used to provide a quick status update to the user, so continuing to run the action after a long time isn't useful.

Therefore Seam queues the action event for a period of time (the concurrent request timeout); if it can't process the event in time, it creates a temporary conversation and prints out a message to the user to let them know what's going on. It's therefore very important not to flood the server with AJAX events!

We can set a sensible default for the concurrent request timeout (in ms) in components.xml:


<core:manager concurrent-request-timeout="500" />

So far we've discussed "synchronous" AJAX requests - the client tells the server that an event has occur, and then rerenders part of the page based on the result. This approach is great when the AJAX request is lightweight (the methods called are simple e.g. calculating the sum of a column of numbers). But what if we need to do a complex computation?

For heavy computation we should use a truly asynchronous (poll based) approach — the client sends an AJAX request to the server, which causes action to be executed asynchronously on the server (so the the response to the client is immediate); the client then polls the server for updates. This is useful when you have a long-running action for which it is important that every action executes (you don't want some to be dropped as duplicates, or to timeout).

How should we design our conversational AJAX application?

Well first, you need to decide whether you want to use the simpler "synchronous" request or whether you want to add using a poll-style approach.

If you go for a "synchronous" approach, then you need to make an estimate of how long your AJAX request will take to complete - is it much shorter than the concurrent request timeout? If not, you probably want to alter the concurrent request timeout for this method (as discussed above). Next you probably want a queue on the client side to prevent flooding the server with requests. If the event occurs often (e.g. a keypress, onblur of input fields) and immediate update of the client is not a priority you should set a request delay on the client side. When working out your request delay, factor in that the event may also be queued on the server side.

Finally, the client library may provide an option to abort unfinished duplicate requests in favor of the most recent. You need to be careful with this option as it can lead to flooding of the server with requests if the server is not able to abort the unfinished request.

Using a poll-style design requires less fine-tuning. You just mark your action method @Asynchronous and decide on a polling interval:

int total;


// This method is called when an event occurs on the client
// It takes a really long time to execute
@Asynchronous      
public void calculateTotal() {
   total = someReallyComplicatedCalculation();
}
// This method is called as the result of the poll
// It's very quick to execute
public int getTotal() {
   return total;
}

JBoss jBPM is a business process management engine for any Java SE or EE environment. jBPM lets you represent a business process or user interaction as a graph of nodes representing wait states, decisions, tasks, web pages, etc. The graph is defined using a simple, very readable, XML dialect called jPDL, and may be edited and visualised graphically using an eclipse plugin. jPDL is an extensible language, and is suitable for a range of problems, from defining web application page flow, to traditional workflow management, all the way up to orchestration of services in a SOA environment.

Seam applications use jBPM for two different problems:

  • Defining the pageflow involved in complex user interactions. A jPDL process definition defines the page flow for a single conversation. A Seam conversation is considered to be a relatively short-running interaction with a single user.

  • Defining the overarching business process. The business process may span multiple conversations with multiple users. Its state is persistent in the jBPM database, so it is considered long-running. Coordination of the activities of multiple users is a much more complex problem than scripting an interaction with a single user, so jBPM offers sophisticated facilities for task management and dealing with multiple concurrent paths of execution.

Don't get these two things confused ! They operate at very different levels or granularity. Pageflow, conversation and task all refer to a single interaction with a single user. A business process spans many tasks. Futhermore, the two applications of jBPM are totally orthogonal. You can use them together or independently or not at all.

You don't have to know jDPL to use Seam. If you're perfectly happy defining pageflow using JSF or Seam navigation rules, and if your application is more data-driven that process-driven, you probably don't need jBPM. But we're finding that thinking of user interaction in terms of a well-defined graphical representation is helping us build more robust applications.

There are two ways to define pageflow in Seam:

Very simple applications will only need the stateless navigation model. Very complex applications will use both models in different places. Each model has its strengths and weaknesses!

The stateless model defines a mapping from a set of named, logical outcomes of an event directly to the resulting page of the view. The navigation rules are entirely oblivious to any state held by the application other than what page was the source of the event. This means that your action listener methods must sometimes make decisions about the page flow, since only they have access to the current state of the application.

Here is an example page flow definition using JSF navigation rules:


<navigation-rule>
    <from-view-id>/numberGuess.jsp</from-view-id>
        
    <navigation-case>
        <from-outcome>guess</from-outcome>
        <to-view-id>/numberGuess.jsp</to-view-id>
        <redirect/>
    </navigation-case>

    <navigation-case>
        <from-outcome>win</from-outcome>
        <to-view-id>/win.jsp</to-view-id>
        <redirect/>
    </navigation-case>
        
    <navigation-case>
        <from-outcome>lose</from-outcome>
        <to-view-id>/lose.jsp</to-view-id>
        <redirect/>
    </navigation-case>

</navigation-rule>

Here is the same example page flow definition using Seam navigation rules:


<page view-id="/numberGuess.jsp">
        
    <navigation>
        <rule if-outcome="guess">
            <redirect view-id="/numberGuess.jsp"/>
        </rule>
        <rule if-outcome="win">
            <redirect view-id="/win.jsp"/>
        </rule>
        <rule if-outcome="lose">
            <redirect view-id="/lose.jsp"/>
        </rule>
    </navigation>

</page>

If you find navigation rules overly verbose, you can return view ids directly from your action listener methods:

public String guess() {

    if (guess==randomNumber) return "/win.jsp";
    if (++guessCount==maxGuesses) return "/lose.jsp";
    return null;
}

Note that this results in a redirect. You can even specify parameters to be used in the redirect:

public String search() {

    return "/searchResults.jsp?searchPattern=#{searchAction.searchPattern}";
}

The stateful model defines a set of transitions between a set of named, logical application states. In this model, it is possible to express the flow of any user interaction entirely in the jPDL pageflow definition, and write action listener methods that are completely unaware of the flow of the interaction.

Here is an example page flow definition using jPDL:


<pageflow-definition name="numberGuess">
    
   <start-page name="displayGuess" view-id="/numberGuess.jsp">
      <redirect/>
      <transition name="guess" to="evaluateGuess">
        <action expression="#{numberGuess.guess}" />
      </transition>
   </start-page>
   
   <decision name="evaluateGuess" expression="#{numberGuess.correctGuess}">
      <transition name="true" to="win"/>
      <transition name="false" to="evaluateRemainingGuesses"/>
   </decision>
   
   <decision name="evaluateRemainingGuesses" expression="#{numberGuess.lastGuess}">
      <transition name="true" to="lose"/>
      <transition name="false" to="displayGuess"/>
   </decision>
   
   <page name="win" view-id="/win.jsp">
      <redirect/>
      <end-conversation />
   </page>
   
   <page name="lose" view-id="/lose.jsp">
      <redirect/>
      <end-conversation />
   </page>
   
</pageflow-definition>

There are two things we notice immediately here:

In addition, the stateful model is more constrained. For each logical state (each step in the page flow), there are a constrained set of possible transitions to other states. The stateless model is an ad hoc model which is suitable to relatively unconstrained, freeform navigation where the user decides where he/she wants to go next, not the application.

The stateful/stateless navigation distinction is quite similar to the traditional view of modal/modeless interaction. Now, Seam applications are not usually modal in the simple sense of the word - indeed, avoiding application modal behavior is one of the main reasons for having conversations! However, Seam applications can be, and often are, modal at the level of a particular conversation. It is well-known that modal behavior is something to avoid as much as possible; it is very difficult to predict the order in which your users are going to want to do things! However, there is no doubt that the stateful model has its place.

The biggest contrast between the two models is the back-button behavior.

When JSF or Seam navigation rules are used, Seam lets the user freely navigate via the back, forward and refresh buttons. It is the responsibility of the application to ensure that conversational state remains internally consistent when this occurs. Experience with the combination of web application frameworks like Struts or WebWork - that do not support a conversational model - and stateless component models like EJB stateless session beans or the Spring framework has taught many developers that this is close to impossible to do! However, our experience is that in the context of Seam, where there is a well-defined conversational model, backed by stateful session beans, it is actually quite straightforward. Usually it is as simple as combining the use of no-conversation-view-id with null checks at the beginning of action listener methods. We consider support for freeform navigation to be almost always desirable.

In this case, the no-conversation-view-id declaration goes in pages.xml. It tells Seam to redirect to a different page if a request originates from a page rendered during a conversation, and that conversation no longer exists:


<page view-id="/checkout.xhtml" 
        no-conversation-view-id="/main.xhtml"/>

On the other hand, in the stateful model, backbuttoning is interpreted as an undefined transition back to a previous state. Since the stateful model enforces a defined set of transitions from the current state, back buttoning is by default disallowed in the stateful model! Seam transparently detects the use of the back button, and blocks any attempt to perform an action from a previous, "stale" page, and simply redirects the user to the "current" page (and displays a faces message). Whether you consider this a feature or a limitation of the stateful model depends upon your point of view: as an application developer, it is a feature; as a user, it might be frustrating! You can enable backbutton navigation from a particular page node by setting back="enabled".


<page name="checkout" 
        view-id="/checkout.xhtml" 
        back="enabled">
    <redirect/>
    <transition to="checkout"/>
    <transition name="complete" to="complete"/>
</page>

This allows backbuttoning from the checkout state to any previous state!

Of course, we still need to define what happens if a request originates from a page rendered during a pageflow, and the conversation with the pageflow no longer exists. In this case, the no-conversation-view-id declaration goes into the pageflow definition:


<page name="checkout" 
        view-id="/checkout.xhtml" 
        back="enabled" 
        no-conversation-view-id="/main.xhtml">
    <redirect/>
    <transition to="checkout"/>
    <transition name="complete" to="complete"/>
</page>

In practice, both navigation models have their place, and you'll quickly learn to recognize when to prefer one model over the other.

We "start" a jPDL-based pageflow by specifying the name of the process definition using a @Begin, @BeginTask or @StartTask annotation:

@Begin(pageflow="numberguess")

public void begin() { ... }

Alternatively we can start a pageflow using pages.xml:


<page>
        <begin-conversation pageflow="numberguess"/>
    </page>

If we are beginning the pageflow during the RENDER_RESPONSE phase—during a @Factory or @Create method, for example—we consider ourselves to be already at the page being rendered, and use a <start-page> node as the first node in the pageflow, as in the example above.

But if the pageflow is begun as the result of an action listener invocation, the outcome of the action listener determines which is the first page to be rendered. In this case, we use a <start-state> as the first node in the pageflow, and declare a transition for each possible outcome:


<pageflow-definition name="viewEditDocument">

    <start-state name="start">
        <transition name="documentFound" to="displayDocument"/>
        <transition name="documentNotFound" to="notFound"/>
    </start-state>
    
    <page name="displayDocument" view-id="/document.jsp">
        <transition name="edit" to="editDocument"/>
        <transition name="done" to="main"/>
    </page>
    
    ...
    
    <page name="notFound" view-id="/404.jsp">
        <end-conversation/>
    </page>
    
</pageflow-definition>

Each <page> node represents a state where the system is waiting for user input:


<page name="displayGuess" view-id="/numberGuess.jsp">
    <redirect/>
    <transition name="guess" to="evaluateGuess">
        <action expression="#{numberGuess.guess}" />
    </transition>
</page>

The view-id is the JSF view id. The <redirect/> element has the same effect as <redirect/> in a JSF navigation rule: namely, a post-then-redirect behavior, to overcome problems with the browser's refresh button. (Note that Seam propagates conversation contexts over these browser redirects. So there is no need for a Ruby on Rails style "flash" construct in Seam!)

The transition name is the name of a JSF outcome triggered by clicking a command button or command link in numberGuess.jsp.


<h:commandButton type="submit" value="Guess" action="guess"/>

When the transition is triggered by clicking this button, jBPM will activate the transition action by calling the guess() method of the numberGuess component. Notice that the syntax used for specifying actions in the jPDL is just a familiar JSF EL expression, and that the transition action handler is just a method of a Seam component in the current Seam contexts. So we have exactly the same event model for jBPM events that we already have for JSF events! (The One Kind of Stuff principle.)

In the case of a null outcome (for example, a command button with no action defined), Seam will signal the transition with no name if one exists, or else simply redisplay the page if all transitions have names. So we could slightly simplify our example pageflow and this button:


<h:commandButton type="submit" value="Guess"/>

Would fire the following un-named transition:


<page name="displayGuess" view-id="/numberGuess.jsp">
    <redirect/>
    <transition to="evaluateGuess">
        <action expression="#{numberGuess.guess}" />
    </transition>
</page>

It is even possible to have the button call an action method, in which case the action outcome will determine the transition to be taken:


<h:commandButton type="submit" value="Guess" action="#{numberGuess.guess}"/>

<page name="displayGuess" view-id="/numberGuess.jsp">
    <transition name="correctGuess" to="win"/>
    <transition name="incorrectGuess" to="evaluateGuess"/>
</page>

However, this is considered an inferior style, since it moves responsibility for controlling the flow out of the pageflow definition and back into the other components. It is much better to centralize this concern in the pageflow itself.

A business process is a well-defined set of tasks that must be performed by users or software systems according to well-defined rules about who can perform a task, and when it should be performed. Seam's jBPM integration makes it easy to display lists of tasks to users and let them manage their tasks. Seam also lets the application store state associated with the business process in the BUSINESS_PROCESS context, and have that state made persistent via jBPM variables.

A simple business process definition looks much the same as a page flow definition (One Kind of Stuff), except that instead of <page> nodes, we have <task-node> nodes. In a long-running business process, the wait states are where the system is waiting for some user to log in and perform a task.


<process-definition name="todo">
   
   <start-state name="start">
      <transition to="todo"/>
   </start-state>
   
   <task-node name="todo">
      <task name="todo" description="#{todoList.description}">
         <assignment actor-id="#{actor.id}"/>
      </task>
      <transition to="done"/>
   </task-node>
   
   <end-state name="done"/>
   
</process-definition>

It is perfectly possible that we might have both jPDL business process definitions and jPDL pageflow definitions in the same project. If so, the relationship between the two is that a single <task> in a business process corresponds to a whole pageflow <pageflow-definition>

Seam provides extensive support for the two most popular persistence architectures for Java: Hibernate3, and the Java Persistence API introduced with EJB 3.0. Seam's unique state-management architecture allows the most sophisticated ORM integration of any web application framework.

Seam grew out of the frustration of the Hibernate team with the statelessness typical of the previous generation of Java application architectures. The state management architecture of Seam was originally designed to solve problems relating to persistence—in particular problems associated with optimistic transaction processing. Scalable online applications always use optimistic transactions. An atomic (database/JTA) level transaction should not span a user interaction unless the application is designed to support only a very small number of concurrent clients. But almost all interesting work involves first displaying data to a user, and then, slightly later, updating the same data. So Hibernate was designed to support the idea of a persistence context which spanned an optimistic transaction.

Unfortunately, the so-called "stateless" architectures that preceded Seam and EJB 3.0 had no construct for representing an optimistic transaction. So, instead, these architectures provided persistence contexts scoped to the atomic transaction. Of course, this resulted in many problems for users, and is the cause of the number one user complaint about Hibernate: the dreaded LazyInitializationException. What we need is a construct for representing an optimistic transaction in the application tier.

EJB 3.0 recognizes this problem, and introduces the idea of a stateful component (a stateful session bean) with an extended persistence context scoped to the lifetime of the component. This is a partial solution to the problem (and is a useful construct in and of itself) however there are two problems:

Seam solves the first problem by providing conversations, and stateful session bean components scoped to the conversation. (Most conversations actually represent optimistic transactions in the data layer.) This is sufficient for many simple applications (such as the Seam booking demo) where persistence context propagation is not needed. For more complex applications, with many loosly-interacting components in each conversation, propagation of the persistence context across components becomes an important issue. So Seam extends the persistence context management model of EJB 3.0, to provide conversation-scoped extended persistence contexts.

EJB session beans feature declarative transaction management. The EJB container is able to start a transaction transparently when the bean is invoked, and end it when the invocation ends. If we write a session bean method that acts as a JSF action listener, we can do all the work associated with that action in one transaction, and be sure that it is committed or rolled back when we finish processing the action. This is a great feature, and all that is needed by some Seam applications.

However, there is a problem with this approach. A Seam application may not perform all data access for a request from a single method call to a session bean.

The more transactions per request, the more likely we are to encounter atomicity and isolation problems when our application is processing many concurrent requests. Certainly, all write operations should occur in the same transaction!

Hibernate users developed the "open session in view" pattern to work around this problem. In the Hibernate community, "open session in view" was historically even more important because frameworks like Spring use transaction-scoped persistence contexts. So rendering the view would cause LazyInitializationExceptions when unfetched associations were accessed.

This pattern is usually implemented as a single transaction which spans the entire request. There are several problems with this implementation, the most serious being that we can never be sure that a transaction is successful until we commit it—but by the time the "open session in view" transaction is committed, the view is fully rendered, and the rendered response may already have been flushed to the client. How can we notify the user that their transaction was unsuccessful?

Seam solves both the transaction isolation problem and the association fetching problem, while working around the problems with "open session in view". The solution comes in two parts:

In the next section, we'll tell you how to set up a conversation-scope persistence context. But first we need to tell you how to enable Seam transaction management. Note that you can use conversation-scoped persistence contexts without Seam transaction management, and there are good reasons to use Seam transaction management even when you're not using Seam-managed persistence contexts. However, the two facilities were designed to work together, and work best when used together.

Seam transaction management is useful even if you're using EJB 3.0 container-managed persistence contexts. But it is especially useful if you use Seam outside a Java EE 5 environment, or in any other case where you would use a Seam-managed persistence context.

Seam provides a transaction management abstraction for beginning, committing, rolling back, and synchronizing with a transaction. By default Seam uses a JTA transaction component that integrates with Container Managed and programmatic EJB transactions. If you are working in a Java EE 5 environment, you should install the EJB synchronization component in components.xml:


<transaction:ejb-transaction />

However, if you are working in a non EE 5 container, Seam will try auto detect the transaction synchronization mechanism to use. However, if Seam is unable to detect the correct transaction synchronization to use, you may find you need configure one of the following:

Configure JPA RESOURCE_LOCAL transaction management by adding the following to your components.xml where #{em} is the name of the persistence:managed-persistence-context component. If your managed persistence context is named entityManager, you can opt to leave out the entity-manager attribute. (see Seam-managed persistence contexts )


<transaction:entity-transaction entity-manager="#{em}"/>

To configure Hibernate managed transactions declare the following in your components.xml where #{hibernateSession} is the name of the project's persistence:managed-hibernate-session component. If your managed hibernate session is named session, you can opt to leave out the session attribute. (see Seam-managed persistence contexts )


<transaction:hibernate-transaction session="#{hibernateSession}"/>

To explicitly disable Seam managed transactions declare the following in your components.xml:


<transaction:no-transaction />

For configuring Spring managed transactions see using Spring PlatformTransactionManagement .

If you're using Seam outside of a Java EE 5 environment, you can't rely upon the container to manage the persistence context lifecycle for you. Even if you are in an EE 5 environment, you might have a complex application with many loosly coupled components that collaborate together in the scope of a single conversation, and in this case you might find that propagation of the persistence context between component is tricky and error-prone.

In either case, you'll need to use a managed persistence context (for JPA) or a managed session (for Hibernate) in your components. A Seam-managed persistence context is just a built-in Seam component that manages an instance of EntityManager or Session in the conversation context. You can inject it with @In.

Seam-managed persistence contexts are extremely efficient in a clustered environment. Seam is able to perform an optimization that EJB 3.0 specification does not allow containers to use for container-managed extended persistence contexts. Seam supports transparent failover of extended persisence contexts, without the need to replicate any persistence context state between nodes. (We hope to fix this oversight in the next revision of the EJB spec.)

Configuring a managed persistence context is easy. In components.xml, we can write:


<persistence:managed-persistence-context name="bookingDatabase" 
                                  auto-create="true"
                   persistence-unit-jndi-name="java:/EntityManagerFactories/bookingData"/>

This configuration creates a conversation-scoped Seam component named bookingDatabase that manages the lifecycle of EntityManager instances for the persistence unit (EntityManagerFactory instance) with JNDI name java:/EntityManagerFactories/bookingData.

Of course, you need to make sure that you have bound the EntityManagerFactory into JNDI. In JBoss, you can do this by adding the following property setting to persistence.xml.


<property name="jboss.entity.manager.factory.jndi.name" 
          value="java:/EntityManagerFactories/bookingData"/>

Now we can have our EntityManager injected using:

@In EntityManager bookingDatabase;

If you are using EJB3 and mark your class or method @TransactionAttribute(REQUIRES_NEW) then the transaction and persistence context shouldn't be propagated to method calls on this object. However as the Seam-managed persistence context is propagated to any component within the conversation, it will be propagated to methods marked REQUIRES_NEW. Therefore, if you mark a method REQUIRES_NEW then you should access the entity manager using @PersistenceContext.

Persistence contexts scoped to the conversation allows you to program optimistic transactions that span multiple requests to the server without the need to use the merge() operation , without the need to re-load data at the beginning of each request, and without the need to wrestle with the LazyInitializationException or NonUniqueObjectException.

As with any optimistic transaction management, transaction isolation and consistency can be achieved via use of optimistic locking. Fortunately, both Hibernate and EJB 3.0 make it very easy to use optimistic locking, by providing the @Version annotation.

By default, the persistence context is flushed (synchronized with the database) at the end of each transaction. This is sometimes the desired behavior. But very often, we would prefer that all changes are held in memory and only written to the database when the conversation ends successfully. This allows for truly atomic conversations. As the result of a truly stupid and shortsighted decision by certain non-JBoss, non-Sun and non-Sybase members of the EJB 3.0 expert group, there is currently no simple, usable and portable way to implement atomic conversations using EJB 3.0 persistence. However, Hibernate provides this feature as a vendor extension to the FlushModeTypes defined by the specification, and it is our expectation that other vendors will soon provide a similar extension.

Seam lets you specify FlushModeType.MANUAL when beginning a conversation. Currently, this works only when Hibernate is the underlying persistence provider, but we plan to support other equivalent vendor extensions.

@In EntityManager em; //Seam-managed persistence context


@Begin(flushMode=MANUAL)
public void beginClaimWizard() {
    claim = em.find(Claim.class, claimId);
}

Now, the claim object remains managed by the persistence context for the rest ot the conversation. We can make changes to the claim:

public void addPartyToClaim() {

    Party party = ....;
    claim.addParty(party);
}

But these changes will not be flushed to the database until we explicitly force the flush to occur:

@End

public void commitClaim() {
    em.flush();
}

Of course, you could set the flushMode to MANUAL from pages.xml, for example in a navigation rule:


<begin-conversation flush-mode="MANUAL" />

The EntityManager interface lets you access a vendor-specific API via the getDelegate() method. Naturally, the most interesting vendor is Hibernate, and the most powerful delegate interface is org.hibernate.Session. You'd be nuts to use anything else. Trust me, I'm not biased at all. If you must use a different JPA provider see Using Alternate JPA Providers.

But regardless of whether you're using Hibernate (genius!) or something else (masochist, or just not very bright), you'll almost certainly want to use the delegate in your Seam components from time to time. One approach would be the following:

@In EntityManager entityManager;


@Create
public void init() {
    ( (Session) entityManager.getDelegate() ).enableFilter("currentVersions");
}

But typecasts are unquestionably the ugliest syntax in the Java language, so most people avoid them whenever possible. Here's a different way to get at the delegate. First, add the following line to components.xml:


<factory name="session" 
         scope="STATELESS" 
         auto-create="true" 
         value="#{entityManager.delegate}"/>

Now we can inject the session directly:

@In Session session;


@Create
public void init() {
    session.enableFilter("currentVersions");
}

In plain JSF, validation is defined in the view:


<h:form>
    <h:messages/>

    <div>
        Country:
        <h:inputText value="#{location.country}" required="true">
            <my:validateCountry/>
        </h:inputText>
    </div>
    
    <div>
        Zip code:
        <h