Deployment descriptor in Java

Encode PDF-417 2d barcode in Java Deployment descriptor

<ejb-jar> <enterprise-beans> <!-- =========[ ContainerManaged Bean ]======== --> <entity> Declares the EJB that <ejb-name>containerManaged</ejb-name> will use container<home>containerManaged.EquityHome</home> managed persistence <remote>containerManaged.Equity</remote> <ejb-class>containerManaged.EquityBean</ejb-class> <persistence-type>Container</persistence-type> <prim-key-class>java.lang.String</prim-key-class> <reentrant>False</reentrant> <cmp-version>2.x</cmp-version> <abstract-schema-name>EquityBean</abstract-schema-name> <cmp-field> Declares the persistent <field-name>symbol</field-name> fields of the bean </cmp-field> <cmp-field> <field-name>description</field-name> Indicates the use of the </cmp-field> 2.0 specification and <cmp-field> declares the schema name <field-name>lastTrade</field-name> </cmp-field> <cmp-field> <field-name>change</field-name> </cmp-field> <cmp-field> <field-name>volume</field-name> </cmp-field> <cmp-field> <field-name>marketCap</field-name> </cmp-field> <cmp-field> <field-name>pe</field-name>

</cmp-field> <cmp-field> <field-name>avgVolume</field-name> Declares the primary </cmp-field> key field of the bean <primkey-field>symbol</primkey-field> </entity> <!-- =========[ ContainerManaged Bean ]======== --> </enterprise-beans> <assembly-descriptor> </assembly-descriptor> </ejb-jar>

As you can see in the EquityBean example, the source for a CMP bean is noticeably different from that of a BMP bean. First of all, the methods will not contain any persistence code, like JDBC connections or SQL statements. Also, a CMP entity bean has a much more descriptive deployment descriptor than a BMP entity bean. However, like a BMP bean, it must declare its persistence type. In this case, it should be declared Container. Similarly, the deployment descriptor should also tell the container to use the EJB 2.0 specification (which is what this recipe describes). After declaring the bean to be a 2.0 CMP entity bean, you should describe the persistent fields and schema used by the bean. Indicate each field that will be persisted with a <cmp-field/> tag. The EJB container will generate a concrete subclass of your abstract bean that implements the abstract methods to actually perform the persistence and loading of the bean data as needed. The schema and persistent fields are mapped to an actual database table in a vendor-specific manner, usually with a vendor deployment descriptor for the bean. Check your vendor s documentation for more information on what you need to do. Typically, application servers require you to provide an additional XML descriptor file. For example, Weblogic requires you to build two files: weblogic-ejb-jar.xml and weblogic-cmp-rdbms-jar.xml. Lastly, you should declare which persistent field is the primary key for this entity bean. When using entity beans in your application, you should always encapsulate access to them through a session bean. Not only does this separate your clients from the data layer, but it lets you provide better security, transaction management, and performance. For instance, when you access your entity beans through a session facade (see recipe 3.15), your entity beans need only implement local

interfaces. This prevents any remote client from finding your entity data without first passing through the session bean layer and being validated.

2.3 Keeping your EJB deployment descriptor current 3.4 Using a database sequence to generate primary key values for entity beans 3.5 Using a compound primary key for your entity beans

In your application, you need the ability to provide a different database view (or database) depending on the current user. This is one way to prevent unauthorized users from updating restricted data. Since the EJB security model provides for the discovery of the invoking client, you can easily determine a user s identity and assigned roles. Programmatically restricting data access to specific users can make your code less flexible and maintainable.

In order to switch data sources for a particular user, you must first determine which user is calling an EJB method. Inside the UserSpecificDBBean session bean shown in listing 3.5, the getConnection() method determines the invoking user and returns a JDBC connection based on the retrieved value.

public class UserSpecificDBBean implements SessionBean { private SessionContext ctx; public void setSessionContext(SessionContext ctx) { this.ctx = ctx; } public Connection getConnection() throws SQLException{ String srcName = null;

InitialContext initCtx = null; try { Principal p = this.ctx.getCallerPrincipal(); String username = p.getName();