Ameeta,
The
Use Case View is essentially the requirements capture. It should have pretty
abstract / high level use cases that indicate each 'piece' of functionality that
the system must provide.
It
primarily includes Actors and use cases and, only ocassionally,
classes.
The
Logical view is basically the detailed description of the system. It should be
broken into two major parts - Analysis View and Design View.
A good
way of proceeding from Use Case View (Requirements) to Analysis View is to take
each high level use case and realise (use the 'realises' relationship for
tracability) it in sequence diagrams.
In so
doing you will need to think about objects that will take responsibility
for the tasks (operations) in the use case sequence. Thus, you will generate
high level classes in your Analysis (the standard stereotypes are quite
suggestive here, e.g. boundary class to interact with an
Actor).
Then
to proceed to the Design view you can either realise all of your Analysis
classes with detailed Design classes or realise your Analysis sequences with
detailed Design sequences (usually a mixture of both).
Finally, may I suggest that you go another level down -
into the component view - which should be treated as the Implementation model.
At this level import any components (DLLs, OCXs, etc.) that you will be using
and explicitly use the classes, methods, etc brought in. In other words, focus
here on the system and language specific aspects of a given implementation
environment. If you do this well then you can use the code generation tools and
hopefully you'll find that the code only needs fleshing out and tidying
up.
Regards,
Stephen.
-----Original Message-----
From: Ameeta [mailto:[EMAIL PROTECTED]]
Sent: 03 October 2001 10:50
To: Rose_Forum (E-mail)
Subject: (ROSE) UseCase View To Logical View MappingHi,What is the mapping or how is the mapping fro the Use Case View to the Logical View done?Rgds,Ameeta
