Deciding your data model strategy

First you should decide which classes in the model package are appropriate, depending on the characteristics of your existing business data and your requirements.

There are five levels of data model implementation, each appropriate for certain situations:

There are three possible situations:

You do not have existing data that you want to display and you will create your diagram(s) from scratch.

You have existing data, but you want to duplicate it before displaying it, and it does not matter how the data will be represented in memory.

You have exported your data to an XML file, and you want to read in the XML data.

In these cases, use the default SDM model implementation, IlvDefaultSDMModel.

The nodes and the links of the default model are instances of the classes IlvDefaultSDMNode and IlvDefaultSDMLink. These objects have an arbitrary set of user-defined properties, so the default model can represent all kinds of objects.

Your data is stored in a database, and you want to display it but not modify it directly in the diagram component.

The model package provides a specific SDM model for this case: IlvJDBCSDMModel. This model uses the JDBC™ API to access the database. All you need to do is configure the model with the database URL, the names of the tables containing the nodes and links, and a few other parameters.

You have existing data stored in classes which conform to the JavaBeans standard: there are accessor methods ( get and set ) for each bean property.

In this case, use the class IlvJavaBeanSDMModel. This model uses your JavaBeans as the nodes and links of a graph, so there is no duplication or additional memory consumption. The properties of nodes and links are read and written using the JavaBeans API. All you have to do is supply the set of JavaBeans that will make up your graph. For an example, see JavaBeans example.

4. Existing Java™ classes that are not JavaBeans or existing data but no Java classes.

For example, you already have an implementation of a Swing JTree model that represents your data and you want to display the contents of the tree without duplicating data.

One of the following cases applies:

You have existing data represented by Java classes that are not JavaBeans.

You want to display data that is not already represented in memory by Java classes.

In such cases, there are two possible approaches:

Use the class IlvAbstractSDMModel, and wrap your objects into new objects that implement the IlvSDMNode and IlvSDMLink interfaces. This requires an extra object allocation because of the indirection, but allows you to make use of the predefined interfaces for nodes and links. For an example, see NonJavaBeans example: abstract model variant.

Subclass the basic model, IlvBasicSDMModel, and implement the methods that list the objects and retrieve the properties of each object. This requires more work but is more open, and may be preferable if all objects already exist. For an example, see NonJavaBeans example: basic model variant.

Summary of data model strategies
Data Characteristics	Recommended Class
New or from XML	IlvDefaultSDMModel
Database (display only)	IlvJDBCSDMModel
JavaBeans	IlvJavaBeanSDMModel
Java classes or other existing objects (for example, Swing)	IlvAbstractSDMModel
Java classes or other existing objects (for example, Swing)	IlvBasicSDMModel

If the graphic objects are to be moved interactively, SDM tries to store the new object locations as x,y properties in the model. However, the model can refuse to set the x,y properties, for example when it is a read-only model. To cope with such a case, the SDM library provides a metadata system that manages new properties as if they were stored in the true model.

To enable metadata, call ilog.views.sdm.IlvSDMEngine.setMetadataEnabled(true). Metadata is saved only in XML files. Other persistent mechanisms are responsible for saving metadata if they need to.