Java Stammtisch Carol McDonald

Thursday Jul 30, 2009

Optimistic Concurrency


Optimistic locking lets concurrent transactions process simultaneously, but detects and prevent collisions, this works best for applications where most concurrent transactions do not conflict. JPA Optimistic locking allows anyone to read and update an entity, however a version check is made upon commit and an exception is thrown if the version was updated in the database since the entity was read.  In JPA for Optimistic locking you annotate an attribute with @Version as shown below:

public class Employee {
    @ID int id;
@Version int version;

The Version attribute will be incremented with a successful commit. The Version attribute can be an int, short, long, or timestamp.  This results in SQL like the following:

“UPDATE Employee SET ..., version = version + 1
     WHERE id = ? AND version = readVersion

The advantages of optimistic locking are that no database locks are held which can give better scalability. The disadvantages are that the user or application must refresh and retry failed updates.

Optimistic Locking Example


In the optimistic locking example below, 2 concurrent transactions are updating employee e1. The transaction on the left commits first causing the e1 version attribute to be incremented with the update. The transaction on the right throws an OptimisticLockException because the e1 version attribute is higher than when e1 was read, causing the transaction to roll back.
img60.jpg

Additional Locking with JPA Entity Locking APIs


With JPA it is possible to lock an entity, this allows you to control when, where and which kind of locking to use. JPA 1.0 only supported Optimistic read or Optimistic write locking.  JPA 2.0 supports Optimistic and Pessimistic locking, this is layered on top of @Version checking described above.

JPA 2.0 LockMode values :
  • OPTIMISTIC (JPA 1.0 READ):
    • perform a version check on locked Entity before commit, throw an OptimisticLockException if Entity version mismatch.
  • OPTIMISTIC_FORCE_INCREMENT (JPA 1.0 WRITE)
    • perform a version check on locked Entity before commit, throw an OptimisticLockException if Entity version mismatch, force an increment to the version at the end of the transaction, even if the entity is not modified.
  • PESSIMISTIC:
    • lock the database row when reading
  • PESSIMISTIC_FORCE_INCREMENT
    • lock the database row when reading, force an increment to the version at the end of the transaction, even if the entity is not modified.
There are multiple APIs to specify locking an Entity:
  • EntityManager methods: lock, find, refresh
  • Query methods: setLockMode 
  • NamedQuery annotation: lockMode element

OPTIMISTIC (READ) LockMode Example


In the optimistic locking example below,  transaction1 on the left updates the department name for dep , which causes dep's version attribute to be incremented. Transaction2 on the right gives an employee a raise if he's in the "Eng" department. Version checking on the employee attribute would not throw an exception in this example since it was the dep Version attribute that was updated in transaction1. In this example the employee change should not commit if the department was changed after reading, so an OPTIMISTIC lock is used : em.lock(dep, OPTIMISTIC).  This will cause a version check on the  dep Entity before committing transaction2  which will throw an OptimisticLockException because the dep version attribute is higher than when dep was read, causing the transaction to roll back.
img62.jpg

OPTIMISTIC_FORCE_INCREMENT (write) LockMode Example


In the OPTIMISTIC_FORCE_INCREMENT locking example below,  transaction2 on the right wants to be sure that the dep name does not change during the transaction, so transaction2 locks the dep Entity em.lock(dep, OPTIMISTIC_FORCE_INCREMENT) and then calls em.flush() which causes dep's version attribute to be incremented in the database. This will cause any parallel updates to dep  to throw an OptimisticLockException and roll back. In transaction1 on the left at commit time when the dep version attribute is checked and found to be stale, an OptimisticLockException is thrown
img63.jpg

Pessimistic Concurrency

Pessimistic concurrency locks the database row when data is read, this is the equivalent of a (SELECT . . . FOR UPDATE [NOWAIT]) .  Pessimistic locking ensures that transactions do not update the same entity at the same time, which can simplify application code, but it limits concurrent access to the data which can cause bad scalability and may cause deadlocks. Pessimistic locking is better for applications with a higher risk of contention among concurrent transactions.
The examples below show:
  1. reading an entity and then locking it later
  2. reading an entity with a lock
  3. reading an entity, then later refreshing it with a lock

The Trade-offs are the longer you hold the lock the greater the risks of bad scalability and deadlocks. The later you lock the greater the risk of stale data, which can then cause an optimistic lock exception, if the entity was updated after reading but before locking.
img66.jpg
img672.jpg
The right locking approach depends on your application:
  • what is the risk of risk of contention among concurrent transactions?
  • What are the requirements for scalability?
  • What are the requirements for user re-trying on failure?

References and More Information:

Preventing Non-Repeatable Reads in JPA Using EclipseLink
Java Persistence API 2.0: What's New ?
What's New and Exciting in JPA 2.0
Beginning Java™ EE 6 Platform with GlassFish™ 3
Pro EJB 3: Java Persistence API (JPA 1.0)

Java Persistence API: Best Practices and Tips





Friday Jul 24, 2009

JSF 2.0, JPA, GlassFish and MySQL

JSF 2.0, JPA, GlassFish and MySQL


This Pet Catalog app explains a web application that uses JSF 2.0, JPA, GlassFish and MySQL. I took this example  GlassFish and MySQL, Part 2: Building a CRUD Web Application With Data Persistence and modified it to use some of the new features JSF 2.0. 

Explanation of the usage of JSF 2.0, Java Persistence APIs, Glassfish and MySQL in a sample Store Catalog Application

The image below shows the Catalog Listing page, which allows a user to page through a list of items in a store.



JSF 2.0 Facelets XHTML instead of JSP

For JSF 2.0, Facelets XHTML is the preferred way to declare JSF Web Pages. JSP is supported for backwards compatibility, but not all JSF 2.0 features will be available for views using JSP as their page declaration language.  JSF 2.0 Facelets has some nice features like templating (similar in functionality to Tiles) and composite components, which I'm not going to discuss here but you can read about that in this article: http://www.ibm.com/developerworks/java/library/j-jsf2fu2/index.html and in this Tech Tip Composite UI Components in JSF 2.0.


The Catalog application's resources


JSF 2.0 standardizes how to define web resources. Resources are any artifacts that a component may need in order to be rendered properly -- images, CSS, or JavaScript files.  With JSF 2.0 you put resources in a resources directory or a subdirectory.


In your Facelets pages, you can access css files with the  <h:outputStylesheet>,  javascript files with the <h:outputScript> , and images with the <h:graphicImage> JSF tags. The list.xhtml uses the  <h:outputStylesheet tag to load the styles.css stylesheet , and the <h:graphicImage tag to display images from the resources as shown below:

Code Sample from:  list.xhtml

<h:outputStylesheet name="css/styles.css" target="body"/>

<h:graphicImage library="images" name="banner_logo.gif"  />   




The Catalog application uses a resource bundle to contain the static text and error messages used by the Facelets pages. Putting messages in a resource bundle makes it easier to modify and internationalize your Application text.  The messages are in a properties file in a java package directory.

Code Sample from:  messages.properties 

Title=Pet Catalog
Next=Next
Previous=Prev
Name=Name



The resource bundle is configured in the faces-config.xml File (you don't need any other configuration in the faces-config.xml for JSF 2.0, as explained later you no longer have to configure managed beans and navigation with XML).


Code Sample from:  faces-config.xml

<application>
    <resource-bundle>
        <base-name>web.WebMessages</base-name>
        <var>msgs</var>
    </resource-bundle>
</application>




The List.xhtml facelets page uses a JSF dataTable component to display a list of catalog items in an html table.  The dataTable component is useful when you want to show a set of results in a table. In a JavaServer Faces application, the UIData component (the superclass of dataTable)  supports binding to a collection of data objects. It does the work of iterating over each record in the data source. The HTML dataTable renderer displays the data as an HTML table.

In the list.xhtml web page the dataTable is defined as shown below:  (Note: Red colors are for Java EE tags, annotations code,  and Green is for my code or variables)

Code Sample from:  list.xhtml

<h:dataTable value='#{catalog.items}' var='row' border="1"
      cellpadding="2" cellspacing="0">



The value attribute of a dataTable tag references the data to be included in the table. The var attribute specifies a name that is used by the components within the dataTable tag as an alias to the data referenced in the value attribute of dataTable.  In the dataTable tag from the List.jsp page, the value attribute points to a list of catalog items. The var attribute points to a single item in that list. As the dataTable component iterates through the list, each reference to dataTableItem points to the current item in the list.

JSF 2.0 Annotations instead of XML configuration

The dataTable's value is bound to the items property of the catalog managed bean. With JSF 2.0 managed beans do not have to be configured in the faces-config.xml file, you annotate the managed beans instead as shown below:

Code Sample from: Catalog.java


@ManagedBean
@SessionScoped
public class Catalog implements Serializable {




By convention, the name of a managed bean is the same as the class name, with the first letter of the class name in lowercase. To specify a managed bean name you can use the name attribute of the ManagedBean annotation, like this: @ManagedBean(name = "Catalog").


This Catalog ManagedBean items property is defined as shown below:

Code Sample from: Catalog.java

    private DataModel items = null;

    public DataModel getItems() {
        if (items == null) {
            getPagingInfo();
            items = new ListDataModel(getNextItems(pagingInfo.getBatchSize(), pagingInfo.getFirstItem()));
        }
        return items;
    }





The getItems() method wraps a List of item objects in a DataModel. UIData, the superclass of dataTable, supports data binding to a collection of data objects represented by a DataModel instance.  The data collection underlying a DataModel instance is modeled as a collection of row objects that can be accessed by a row index.  The APIs provide mechanisms to position to a specified row index, and to retrieve an object that represents the data that corresponds to the current row index.   

The Item properties Name, Photo, and price are displayed with the column component:

Code Sample from: list.xhtml

<h:dataTable var="row" value="#{catalog.items}">
  <h:column>
      <f:facet name="header">
          <h:outputText value="#{msgs.Name}"/>
      </f:facet>
      <h:outputText value="#{row.name}"/>
  </h:column>

  <h:column>
      <f:facet name="header">
          <h:outputText value="#{msgs.Photo}"/>
      </f:facet>
<h:graphicImage library="images" name="#{row.imagethumburl}"/>
  </h:column>

  <h:column>
      <f:facet name="header">
          <h:outputText value="#{msgs.Price}"/>
      </f:facet>
      <h:outputText value="#{row.price}"/>
  </h:column>
</h:dataTable>


The column tags represent columns of data in a UIData component. While the UIData component is iterating over the rows of data, it processes the UIColumn component associated with each column tag for each row in the table.

The UIData component  iterates through the list of items (catalog.items)  and displays the row.price. Each time UIData iterates through the list of items, it renders one cell in each column.

The dataTable and column tags use facet to represent parts of the table that are not repeated or updated. These include headers, footers, and captions.

Using the Java Persistence API (JPA) with JSF

The Catalog ManagedBean uses the Java Persistence API EntityManager Query object to return a list of items. The Catalog ManagedBean annotates the field private EntityManager em;  with @PersistenceContext , which causes an entity manager to be injected when the managed bean is instatiated.


Code Sample from: Catalog.java

@ManagedBean
@SessionScoped
public class Catalog implements Serializable {


@PersistenceUnit(unitName = "catalogPU")
    private EntityManagerFactory emf;

    private EntityManager getEntityManager() {
        return emf.createEntityManager();
    }

    public List<Item>  getNextItems(int maxResults, int firstResult) { 
       EntityManager em = getEntityManager();
       try {    
Query q = em.createQuery("select object(o) from Item as o");
         q.setMaxResults(maxResults);
         q.setFirstResult(firstResult);
         return q.getResultList();
        } finally {
            em.close();
        }      
   }


The Java Persistence Query APIs are used to create and execute queries that can return a list of results.  The JPA Query interface provides support for pagination via the setFirstResult() and setMaxResults() methods: q.setMaxResults(int maxResult) sets the maximum number of results to retrieve. q.setFirstResult(int startPosition) sets the position of the first result to retrieve.

In the code below, we show the Item entity class which maps to the  ITEM table that stores the item instances. This is a typical Java Persistence entity object. There are two requirements for an entity:
  1. annotating the class with an @Entity annotation.
  2. annotating   the primary key identifier with @Id
Because the fields name, description.... are basic mappings from the object fields to columns of the same name in the database table, they don't have to be annotated.  The O/R  relationships with Address and Product are also annotated. For more information on defining JPA entities see Pro EJB 3: Java Persistence API book.

Code Sample from: Item.java

@Entity
public class Item implements java.io.Serializable {

@Id
    private Integer id;

    private String name;   
    private String description;   
    private String imageurl;   
    private String imagethumburl; 
    private BigDecimal price;
@ManyToOne
    private Address address;
@ManyToOne
    private Product product;


    public Item() { }

    public String getName() {
        return name;
    }

    public void setName(String name) {
        this.name = name;
    }


    ...
}   




The Catalog ManagedBean pages through the list of Items by maintaining the PagingInfo.firstItem and PagingInfo.batchSize attributes and passing these as parameters to the  getNextItems(firstItem, batchSize) method. The catalog's scope  is defined with the annotation @SessionScoped, a JSF Managedbean with session scope will be stored in the session meaning that the bean's properties will stay alive for the life of the Http Session.


A JSF commandButton is  used to provide a button to click on to display the next page of items.  The commandButton tag is used to submit an action event to the application. 

Code Sample from: list.xhtml

 <h:commandButton action="#{catalog.next}" value="#{msgs.Next}" />   


This commandButton action attribute references the catalog Managed bean next() method which calculates the next page's first row number  and returns a logical outcome String, which causes the list.xhtml page to display the next page's list . The catalog next method is defined as shown below:

Code Sample from: catalog.java

   public String next() {
       if (firstItem + batchSize < itemCount()) {
           firstItem += batchSize;
       }
       return "list";
   }


JSF 2.0 Simplified Navigation


The JavaServer Faces 2.0  NavigationHandler convention adds .xhtml to the logical outcome of the action method (in this example list) and loads that file, in this case, it loads the list.xhtml page after this method returns. If the action doesn't begin with a forward slash (/), JSF assumes that it's a relative path.  You can specify an absolute path by adding the slash like this "/items/list".


A JSF commandLink is  used to provide a link to click on to display a page with the item details. This commandLink action attribute  references The catalog getDetail() method:

Code Sample from: list.xhtml

   <h:column>
       <f:facet name="header">
          <h:outputText value="Name"/>
       </f:facet>
       <h:commandLink action="#{catalog.getDetail}" value="#{row.name}"/>   
   </h:column>


The catalog getDetail() method  gets the item data from the current row of the dataModel, and returns a string which causes the detail.xhtml page to display the item details :

Code Sample from: Catalog.java

    public String getDetail() {
        item = (Item) model.getRowData();
        return "detail";
    }


The JavaServer Faces NavigationHandler adds .xhtml to the logical outcome of the action, detail and loads that file. In this case, the JavaServer Faces implementation loads the detail.xhtml page after this method returns.

The detail.xhtml uses the outputText component to display the catalog ManagedBean's item properties:

Code Sample from: detail.xhtml

    <h:outputText value="#{catalog.item.name}" title="Name" />
    <h:outputText value="#{catalog.item.description}" title="Description"/>
    <h:graphicImage library="images" name="#{catalog.item.imageurl}" title="Imageurl" />

    <h:outputText value="#{
catalog.item.price}" title="Price" />
    <h:outputText value="#{
catalog.item.address.city}" title="Address" />
    <h:outputText value="#{
catalog.item.contactinfo.email}" title="Address"/>  





Hot Deployment and Session Retention with JSF 2.0 and Glassfish

  • Incremental compile of all JSF 2.0  artifacts when you save.
  • Auto-deploy of all web or Java EE 6 artifacts
  • Session retention: maintain stateful sessions across re-deployments


Conclusion
This concludes the sample application which demonstrates a pet catalog web application which uses JSF 2.0, JPA, GlassFish and MySQL.

Running the Sample Application

  1. If you haven't already done so, download and install NetBeans IDE , GlassFish , and MySQL Community Server . You can download and install GlassFish with NetBeans as a single bundle.
  2. Follow these instructions to install JSF Mojarra 2.0.0 Beta2 on GlassFish v2 or Glassfish v3
  3. Download the sample code.

Create the Pet Catalog database

In order to run the sample code you first have to create the Pet Catalog database and fill in  the Item table.

  1. Start NetBeans IDE
  2. Ensure that GlassFish is registered in the NetBeans IDE, as follows:
    • Click the Services tab in the NetBeans IDE.
    • Expand the Servers node. You should see GlassFish v2 in the list of servers. If not, register GlassFish v2 as follows:
      • Right-click the Servers node and select Add Server. This opens an Add Server Instance wizard.
      • Select GlassFish v2 in the server list of the wizard and click the Next button.
      • Enter the location information for the server and click the Next button.
      • Enter the admin name and password and click the Finish button.

  3. Start the MySQL or Java DB database as follows:
    • Click the Services tab in the NetBeans IDE.
    • Expand the databases node. You should see the Java DB database in the list of databases. If you have installed the MySQL server database, you should also see the MySQL database in the list of databases.. Note:  Java DB  comes bundled with Netbeans, you can  download MySQL separately.

    • Right-mouse click on the Java DB or MySQL server database and select Start.
  4. If you installed MySQL, set the properties of the MySQL server database as follows:
    • Right-click on the MySQL server database and select Properties. This opens the MySQL Server Properties dialog box, as shown in Figure 8.

      MySQL Server Basic Properties
      Figure 8. MySQL Server Basic Properties

    • In the Basic Properties tab, enter the server host name and port number. The IDE specifies localhost as the default server host name and 3306 as the default server port number.
    • Enter the administrator user name, if not displayed, and the administrator password -- the default administrator password is blank.
    • Click the Admin Properties tab.
    • Enter an appropriate path in the Path/URL to admin tool field. You can find the path by browsing to the location of a MySQL Administration application such as the MySQL Admin Tool.
    • Enter an appropriate path in the Path to start command. You can find the path by browsing to the location of the MySQL start command. To find the start command, look for mysqld in the bin folder of the MySQL installation directory.
    • Enter an appropriate path in the Path to stop command field. You can find the path by browsing to the location of the MySQL stop command. This is usually the path to mysqladmin in the bin folder of the MySQL installation directory. If the command is mysqladmin, in the Arguments field, type -u root stop to grant root permissions for stopping the server. The Admin Properties tab should look similar to Figure 9.

      MySQL Server Administration Properties
      Figure 9. MySQL Server Administration Properties

    • Click the OK button.

  5. Right-click on the MySQL server or Java DB database and select Start.
  6. Create the petcatalog database as follows:
    • Right-mouse click on the Java DB or MySQL server database and select Create Database. This will open a create Database window.
    • Enter the database name catalog for Java DB or petcatalog for MySQL.


      For Java DB enter userid password app app as shown below:


       Click O.K. to accept the displayed settings.
  7. Create the tables in the catalog database as follows:
    • Underneath Databases you should see a database connection for the petcatalog database. For example MySQL:

      or Java DB:

    • Right-mouse click on the petcatalog connection and select Connect.
    • Right-mouse click on the petcatalog connection and select Execute Command. This will open up a SQL command window.
    • Copy the contents of the catalog.sql file in the riapetcatalog\exercises\exercise0 directory and paste the contents into the SQL command window, as shown in below:

      Creating Tables in the Database
    • Click the Run SQL icon Run SQL icon (Ctrl+Shift+E) above the SQL command window.
    • Note: It is ok to see this: "Error code -1, SQL state 42Y55: 'DROP TABLE' cannot be performed on 'ITEM' because it does not exist. Line 2, column 1" . This just means you are deleting a table that does not exist.  If you need to delete and recreate the tables you will not  see this message the second time.
  8. View the data in the Pet Catalog database Item table as follows:
    • Underneath Databases you should see a database connection for the petcatalog database. For example MySQL:

      or Java DB:

    • If the database connection is broken like in the following diagram:

      • Right-mouse click on the petcatalog connection and select Connect. as shown below:

      • if prompted for a password, for MySQL leave it blank, for JavaDB enter user app password app.
    • Expand the Tables node below the petcatalog database in the Services window. You should see the item table under the Tables node. You can expand the item table node to see the table columns, indexes, and any foreign keys, as shown in below :
      An Expanded Table Node
      Figure 12. An Expanded Table Node

      You can view the contents of a table or column by right-clicking the table or column and selecting View Data as shown  below:

      Viewing the Contents of a Table
      Figure 13. Viewing the Contents of a Table


  9. Follow these instructions to Create a JDBC Connection pool and JDBC resource.Name the  pool mysql_petcatalog_rootPool and the jndi resource jdbc/petcatalog. Note: you do not have to create a JDBC connection pool and resource if you use the Netbeans wizard to generate JPA entities from database tables as described in this article GlassFish and MySQL, Part 2: Building a CRUD Web Application With Data Persistence.

Running the Sample solution:

If you want to run the sample solution, you have to create the catalog database tables first as described above.

  1. If you haven't already download the sample code and start the NetBeans IDE. Unzip the catalog.zip file which you downloaded, this will create a catalog directory with the project code.
  2. Open the catalog/setup/sun-resources.xml file and verify that the property values it specifies match those of the petcatalog database and jdbc resources you created. Edit the property values as necessary.

  3. Open the catalog project as follows:
    • In NetBeans IDE, click Open Project in the File menu. This opens the Open Project dialog.
    • Navigate in the Open Project dialog to the catalog  directory and click the Open Project button.

    In response, the IDE opens the catalog project.  You can view the logical structure of the project in the Projects window (Ctrl-1).
  4. Run the catalog by right-clicking on the catalog project in the Projects window and selecting Run Project. The NetBeans IDE compiles the application, deploys it on Glassfish, and brings up the default page in your browser.  (at http://localhost:8080/catalog/).

For more information see the following resources:



Wednesday Jul 01, 2009

jax

June: 2 JavaOne Hands On Labs , Sun Technology Exchange, Java Technology Day Israel, and Java Day Turkey  

I had a very busy June, I gave two Hands on Labs at JavaOne, two sessions at the Sun Technology Exchange, three sessions at Java Technology Day in Tel Aviv Israel, and one session at Java Day in Istanbul Turkey.

JavaOne Hands On Labs:

iguana.jpg iguana.jpg

I co-developed and delivered 2 Hands On Labs for JavaOne this year:

You can download these 2 HOLs documentation and code below:


Sun Technology Exchange:

iguana.jpg iguana.jpg
In Fort Lauderdale as part of the Sun Technology Exchange I gave two educational sessions to learn how:
  • JavaFX can help you build rich internet applications (RIAs) and includes the tools and platform SDK for developers, web developers, and designers to create dynamic applications.
  • GlassFish, an enterprise-quality Java EE 5 application server, offers advanced clustering, centralized administration, and best-in-class performance.
  • download the slides


Java Technology Day Israel

iguana.jpg iguana.jpgiguana.jpg
At the Java Technology Day in Israel I gave the following sessions:
  • WSIT Reliability Security and Transactions in Web Services
    • Metro is a high-performance, extensible, easy-to-use web service stack. You can use it for every type of web service, from simple to reliable, secured, and transacted web services that interoperate with .NET services. Metro bundles stable versions of the JAX-WS (Java API for XML Web Services) reference implementation and WSIT (Web Services Interoperability Technology). JAX-WS is a fundamental technology for developing SOAP-based and RESTful Java technology-based web services. WSIT enables secure, reliable interoperability between Java technology-based web services and Microsoft's Windows Communication Foundation.
    • you can download and try out WSIT in this JavaOne HOL: Metro: Try Out Simple and Interoperable Web Services and with these lab instructions.
    • You can read more about some of the example code for this session at
      GlassFish and MySQL, Part 3: Creating a Pet Catalog Web Service
  • MySQL for Developers
    • If you are a developer using MySQL, you should learn enough to take advantage of its strengths, because having an understanding of the database can help you develop better-performing applications. This session talks about MySQL database design and SQL tuning for developers.
    • download or view a screencast of this presentation
  • OpenESB and Connecting Enterprises
    • This session  explains and demonstrates several concrete technologies that make SOA architecture possible - BPEL (Business Process Execution Language), JBI (Java Business Integration) and OpenESB. The part of of BPEL starts with an explanation of the requirements of standardized business process language. The BPEL language is then described using an example. The relationship between BPEL and WSDL is also explained. Finally, BPEL designer and runtime that comes with NetBeans IDE is demonstrated using Travel reservation sample BPEL project. It also explains the motivation of the JBI and OpenESB as a standardized application integration framework in the same way J2EE architecture standardized how enterprise applications are built and deployed. Finally Sun's solution in SOA and application integration space is discussed. Whenever possible, concrete steps of building, deploying and testing SOA applications will be demonstrated step by step.
  • download the slides for all 3

Java Day Turkey

iguana.jpg iguana.jpg iguana.jpg
At the Java Day in Istanbul Turkey I gave the MySQL for Developers session again, see above for more information.