See also in Java

Draw PDF 417 in Java See also

2.1 Generating home, remote, local, and local home interfaces 2.2 Adding and customizing the JNDI name for the home interface 2.5 Generating a primary key class

As you develop more and more entity beans, you find yourself also having to create the primary key class. As we emphasized in this chapter, having to code one more class just adds to the time it takes to develop a bean and increases your chances for having source files out of synch.

To have XDoclet generate a primary key class, use the @ejb.pk tag in your bean source file, use the @ejb.pk-field tag to denote an accessor for the primary key, and modify your <ejbdoclet/> task to include the <entitypk/> subtask. For instance, examine the ItemBean class shown in listing 2.10. The XDoclet tags applicable to this recipe are shown in bold; the others can be found in earlier recipes.

package ch2; import javax.ejb.*; /** * @ejb.bean type="CMP" * name="ItemBean" * jndi-name="ejb/ItemBean" Identifies the * view-type="both" primary key field * primkey-field="ID"; * @ejb.pk Adds the primary key tag */ public abstract class ItemBean implements EntityBean { public void setEntityContext( EntityContext context) {} public void unsetEntityContext( ) {} public public public public public void void void void void ejbRemove() {} ejbLoad() {} ejbStore() {} ejbActivate() {} ejbPassivate() {}

/** * @ejb.create-method */

public void ejbCreate( String id ) { setID( id ); } /** * @ejb.interface-method Identifies the primary * @ejb.persistence key getter method * @ejb.pk-field */ public abstract String getID(); }

Notice in addition to the placement of the specified tags that we included the primkey-field attribute in the @ejb.bean tag at the class declaration level. Note that you must also use the @ejb.persistence tag in combination with the @ejb.pk-field tag.

The result of using these tags is a source file named ItemBeanPK.java (containing the ItemBeanPK class). Listing 2.11 shows the generated code for this class.

/* * Generated by XDoclet - Do not edit! */ package ch2; /** * Primary key for ItemBean. */ public class ItemBeanPK extends java.lang.Object implements java.io.Serializable { private int _hashCode = Integer.MIN_VALUE; private StringBuffer _toStringValue = null; public java.lang.String iD; public ItemBeanPK() { } public ItemBeanPK( java.lang.String iD ) { this.iD = iD; }

public java.lang.String getID() { return iD; } public void setID(java.lang.String iD) { this.iD = iD; _hashCode = Integer.MIN_VALUE; } public int hashCode() { if( _hashCode == Integer.MIN_VALUE ) { if (this.iD != null) _hashCode += this.iD.hashCode(); } return _hashCode; } public boolean equals(Object obj) { if( !(obj instanceof ch2.ItemBeanPK) ) return false; ch2.ItemBeanPK pk = (ch2.ItemBeanPK)obj; boolean eq = true; if( obj == null ) { eq = false; } else { if( this.iD == null && ((ch2.ItemBeanPK)obj).getID() == null ) { eq = true; } else { if( this.iD == null || ((ch2.ItemBeanPK)obj).getID() == null ) { eq = false; } else { eq = eq && this.iD.equals( pk.iD ); } } }

return eq; } /** @return String representation of this pk in the form of [.field1.field2.field3]. */ public String toString() { if( _toStringValue == null ) { _toStringValue = new StringBuffer("[."); _toStringValue.append(this.iD).append('.'); _toStringValue.append(']'); } return _toStringValue.toString(); } }

The generated primary key class contains a default constructor, an initialization constructor that accepts a String ID parameter, a getter method, a setter method, hashcode() and equals() methods, and a toString() method. If you use the @ejb.pk tag without using the @ejb.pk-field tag, you generate a primary key file without the getter, setter, and initialization constructor.

You would like to centralize values in one place and not have to modify source files in order to update the values.

XDoclet is a great tool for generating necessary EJB files. In addition, it lets you specify values for the XML deployment descriptor and JNDI names for your beans. Using XDoclet with your development lets you automate and generate almost everything you need. However, as you add more XDoclet JavaDoc tags to your source files, you are specifying more values in code for things like JNDI names and bean names. Now you have many values spread out across many bean source files.