When you rebuild parts of your model, it is important to understand the impact on the LANSA Repository. For example, what happens when you delete an entity? Or what happens if you change an element definition?
When you remove entities from your model, the corresponding file in the LANSA Repository is NOT removed. For example, if you delete a Customer entity, the related fields and file will not be removed from the repository.
When an element is rebuilt, it will replace the existing definition. For example, any manual changes to help text would be lost when an element is rebuilt. It is important that you implement proper procedures during the modeling stage of the project to provide the maximum benefit from the model builds.
Therefore, it is recommended that you do not perform any manual modification of your database until you have completed all possible work with the Logical Modeler.
Once you reach a point where you are completely satisfied with the modeler implemented physical database, you can start to prototype your applications and enhance the database with LANSA features such as virtual fields, predetermined joined fields, etc.
Once you start to develop your database and applications using the LANSA development environment, you should effectively discard your model. Any changes to the model and its subsequent rebuilding will overwrite any changes you have made manually.
The Logical Modeler supports a one-way build process. Model changes can be made and the database can be rebuilt. The modeler is not designed for a two-way build process where changes to the physical model are read back into the logical model. Because the build is a one-way process, you must carefully consider when to end your logical modeling in order to start your physical development.
Remember, minor database changes can be made directly to the physical model. You do not have to make all changes in the logical model and then rebuild. Determining the point at which you should start full development of the database and application should be based on your project plan. Take advantage of the productivity of model process but do not overall analyze or over model your database.
