DatabaseToUML - Roadmap

M0 – Simple reverse from a java class main using a static JDBC driver (PostGres 8.x)

Limited conversion of datatypes, using standard UML datatypes (String, Integer, and Boolean).
Limited profile (Entity for classes & Identifier for primary keys)
Limited understanding of constraints.
Translation of tables composed only of foreign keys into many to many associations.

M1 – Simple reverse using a wizard.

Wizard to specify the JDBC URL, the catalog and schema + the name of the model.

M2 – Moving to the Datatools Platform.

Wizard using the connection profile, and feed the raw connection.

M3 – Datatools model instead of a proprietary model (representing the database schema)

Use of the internal model from Datatools

M4 – Extended dataype

Use of JavaPrimitive types (from standard UML2 datatypes)

M5 – Parametrized model : V0.

This model will describe how the reverse engineering should be done.

Reverse Normalization parameters.

In example : Convert Table name and field name to pascal and lower camel case convention.

Association preference:

Naming (using class names for association ends)
Default navigation (both ends are navigable)

M6 – Parametrized model : V1

Profiles mapping (using external UML 2.x profiles)
Datatypes mapping (using external UML2.x dataypes)