Re: Looking for real example of building guard expressions on conditions

David M. Karr wrote:
> As I'm having trouble getting a useful reply on my continuing question
> about "junction points", I'll rephrase my question.
>
> I'm looking for a real example of how to build an activity diagram
> containing decision or junction points, where I can label the outgoing
> transitions with either guard conditions, or labels, or something, so I
> can show the logic for taking that transition.  It may seem like a
> simple thing to do, but I still can't see how to do it.  This is my only
> real alternative to Enterprise Architect.
>
> ------------------------------------------------------
> http://argouml.tigris.org/ds/viewMessage.do?dsForumId=449&dsMessageId=2371132
>
> To unsubscribe from this discussion, e-mail: [[email protected]].
Hello David,

When you select a transition, a blue square indicates the locations of a
text area.
Double click on the blue square, and use "Notation" to enter all the
details of the transition. I.e. start typing text on the diagram, at the
location of this blue box by double clicking it.

The syntax is indicated in the status bar of ArgoUML (as long as you
keep the mouse pointer where it is!! - which is a known bug):

    trigger [guard] / effect

This is the standard UML Notation syntax for a transition.
The trigger is the event that causes the transition to be taken.
The guard is a condition, if false then the transition is NOT taken. One
predefined condition is [else] for use with choices.
The effect is an action that is performed while the transition is taken.
All these parts are optional, so valid are:
    [guard]
    trigger /effect
... or any other combination.

All things you type on the diagram, are interpreted. If there is a
syntax error, the input is rejected, and an error message appears in the
status bar at the bottom of the ArgoUML window (hardly visible, I know).
If the syntax is OK, then the result becomes also visible in the
property panels. Here you can see how e.g. an effect is modeled: the
text shown on the diagram is the body of the script expression of the
call action that is the effect of the transition.
This is so complex, since triggers and actions are reusable entities in UML.

If you know how it is structured, then you can as well enter transition
details via the property panels, but it is far easier to try things out
first by editing on the diagram - using UML notation.

Kind regards,
Michiel

------------------------------------------------------
http://argouml.tigris.org/ds/viewMessage.do?dsForumId=449&dsMessageId=2371173

To unsubscribe from this discussion, e-mail: [[email protected]].