[OT] tree structure design question

[David Bolsover <da...@bolsover.com>]

Design Question - this is not strictly a struts question but relates to the
data structure underpinning the application and the mechanisms for accessing
that data so I hope I will be forgiven.

First, an explanation of (some of) the issues:

The project has both GUI (Swing) and Browser (Struts) interfaces and is
accessible by both local intranet and through the internet. In the case of
the swing application, the delivery of relatively large volumes of data is
not too much of a problem.  However, in the case of the Struts application,
large volumes of data are (in the present design) avoided by using a custom
data provided and optimised SQL queries.

Part of the application is a tree structure that shows client organisations
divided into national regions (North, South, East etc.).

Within each region the client contact organisations are detailed
(Laboratory#1, Laboratory#2, Laboratory#3 etc.)

Each client organisation may be divided into several divisions (Thermal
analysis, Chemical analysis etc.)

Regions, organisations and divisions may each have associated contact
personnel (Regional administrator, Laboratory head, Divisional manager etc.)

Users of the system are restricted to viewing only that data associated with
regions to which they have been granted access; some users can only view the
data associated with one region whilst others can see all regions.

To date, managing the data structure has been relatively simple - other than
the fact that regional boundaries sometimes shift with the resultant move of
Laboratories and personnel.

The underlying data structure is a simple table containing all the data for
Regions, Laboratories, Divisions and Personnel each child record having a
link to its parent 

My client in his wisdom has now decided that he wants to reorganise his
sales force into Thermal and Chemical specialists and effectively duplicate
the existing regions (Thermal North, Chemical North, Thermal South etc.) but
retain the flexibility for regional boundaries to move.

The reorganisation now means that Laboratory#1 may be in both Thermal North
and Chemical North regions rather than having a single unique parent as
before.

Now the question

Given that I am constrained to using MySql as the underlying storage
mechanism, what data retrieval and persistence mechanisms are suggested and
how should the data be organised to permit the access restrictions
indicated?


All sugestions, links to resources etc appreciated.

db



---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@struts.apache.org
For additional commands, e-mail: user-help@struts.apache.org