| By Javier Paniza |
Article Rating: |
|
| July 21, 2012 01:00 PM EDT |
Reads: |
4,431 |
OpenXava is a framework for Rapid Java Web Development, well-suited for business and database oriented applications.
OpenXava allows you develop applications just by writing simple domain classes with Java or Groovy. The user interface is generated automatically in runtime, without code generation.
OpenXava 4.5 has some new interesting features such as:
- Support for all strategies of JPA inheritance mapping.
- Data for list and collections is obtained via JPA, before JDBC was used.
- Automated Business Logic (ABL) library added.
- All new layout architecture (the code that paints the UI). It still in BETA but can be activated with layout.painter in xava.properties.
And many other useful new features and bug fixes.
Support for all strategies of JPA inheritance mapping
This feature has been demanded for long time by the community. Until now only single table per class hierarchy mapping strategy was supported, that is the case where all classes hierarchy is stored in the same table. Since OpenXava 4.5 you can also use joined and table per class mapping strategies.
That is, you can write some classes in this way:
@Entity
@Inheritance(strategy=InheritanceType.JOINED)
class Computer extends Identifiable {
String name
}
@Entity
class Tablet extends Computer {
int screenSize
}
In this case, you have a table for Computer with ID and NAME columns and a table for Tablet with ID and SCREENSIZE columns. JPA will do a JOIN of the two table when needed.
Also you can write:
@Entity
@Inheritance(strategy=InheritanceType.TABLE_PER_CLASS)
class Shop extends Identifiable {
String name
}
@Entity
class InternetShop extends Shop {
String url
}
In this case, you have a table for Shop with ID and NAME column and another for InternetShop with ID, NAME and URL columns.
From now on OpenXava will generate a correct user interface for the above classes.
Data for list and collections is obtained via JPA
Until now OpenXava has used JDBC to get the data for list and collections. It was in this way for historical reasons, specially because the very first version of OpenXava used EJB 1.0 CMP entities for persistence and EJB CMP is too slow and rigid to implement the OpenXava list behavior, so we used JDBC. But now, JPA is rapid and flexible enough to be used for obtaining the list data, so we have written a new JPATabProvider to read the data of the list, and it is the one used by default. This allows us to support all the inheritance JPA types, but also opens many interesting posibilities for the future, such as using DataNucleus instead of Hibernate as JPA engine to access to non-relational databases.
Automated Business Logic (ABL) library added
The folk from the California startup Automated Business Logic have been working hard to integrate their library, ABL, in OpenXava. ABL allows you to define business logic declaratively using annotations. The wiki explains how to use ABL from OpenXava.
All new layout architecture (the code that paints the UI)
The people from SicFlex, a Dominican company that has developed a complete ERP with OpenXava, have developed a new render engine for OpenXava. This is the more important and complicated part of OpenXava. This new render engine will generate even more complex and beautiful user interfaces for OpenXava applications. It is still in beta and it's disabled by default, but you can activate it with the next entries in xava.properties:
# Layout parser calculate and prepare properties to be renderer.
# This is optional, by default OX uses org.openxava.render.DefaultLayoutParser
layout.parser=org.openxava.web.layout.DefaultLayoutParser
# Layout painter, render all properties in cell forms
# If a valid painter is defined here, view rendering is performed by it.
layout.painter=org.openxava.web.layout.DefaultLayoutPainter
And more
See changelog for the complete list of new features and fixes.
Download OpenXava 4.5 and try it
Subscribe to Open Source Magazine Email News Alerts
Subscribe via RSS
