Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Topics - vjr

Pages: [1]
1
General Board / Generated robustness diagrams--no UI objects
« on: February 16, 2011, 10:12:23 am »
Hi all. I'm trying out some of the ICONIX stuff in EA. I've created a domain model and UI model. I create a use case containing context references to these domain/UI classes. When I generate a robustness diagram, it doesn't place any of my UI objects (screens) as Boundary objects on the diagram. The EA help states:

"Each UI element in a step becomes a Boundary element.  A Dependency relationship is created from this Boundary element to the UI element (this connector is not shown on the diagram). "

No Entity elements appear on the diagram either.

Further, if I drag a screen element onto the generated diagram it shows as a UI Screen representation and I can't change the stereotype (e.g. to make it a Boundary stereotype so it will display as a Boundary class).

What am I doing wrong? Thanks.

2
General Board / invoke/extend/include in structured scenario
« on: February 04, 2011, 01:32:34 pm »
  • What's the practical benefit of linking to use cases in this way?
  • What's the difference between extend and invoke, specifically?

Thanks

3
Bugs and Issues / java class name change--synchronization
« on: February 21, 2011, 09:04:49 am »
Hi. Is it possible to change a Java class's name in the model and have it synched with the code (also so it changes the name of the source file ala Java convention)? Currently I get a new class generated in the existing source file as follows:

Pre synch file A.java:
Code: [Select]
public class A { public int a; }

Post synch file A.java (rename class A to B in model and forward engineer):
Code: [Select]
public class A { public int a; }

public class B { public int a; }

I would expect A.java to be removed and B.java in its place with the contents:
Code: [Select]
public class B { public int a; }

Thanks.

4
Uml Process / 'asynchronous' events in use case
« on: February 09, 2011, 10:39:09 am »
Just wondering how people model the situation where the user can perform an action during a step being performed by the system.

For example:
1. User asks system to initiate (ongoing) process A
2. System starts performing process A (which keeps going until explicitly stopped)
3. User asks system to stop process A

During step 2 the user can adjust some parameter of process A, that the system must take into account.

I had something like (simplified):

Alternate Flow : Adjust Inputs
Prior to step 3 of the Main Flow, the user can adjust input values:
1. user changes value XYZ
2. return to main flow

But it's not really an 'alternate' flow for step 3, or any other step.

I guess there are 2 questions here: how to word an ongoing process in a use case; and how to model the situation described above.


5
Uml Process / Structured scenarios
« on: February 03, 2011, 01:46:41 pm »
I've just come back to using EA after a few years away from it.

I have some questions on structured scenarios.

I recently found this feature and immediately jumped at it as a way of writing my use cases (and generating their text for reports). That is, simple textual descriptions (numbered steps) of use cases and extending/included use cases as defined in the UML. (That is, as use-cases are defined; their representation isn't defined). Please tell me if there has already been a similar topic, but I haven't found one.

My concern is mainly regarding use cases vs. the term 'scenarios'. In my understanding of a scenario (not actually defined in the UML) is that it is one possible path through a use case. i.e. at 'run-time'. The use case and extensions define all possible such paths. Whereas a use case step has generic phrases such as "1. user enters password in appropriate area" a scenario step would say "user enters 'XYZ' in "Password:" field", or similar.

A use case basic path can have extension points, where the sequence of use-case steps can 'branch' off to extended/included use cases, after which it will return at the next step of the basic path (unless in error). In EA, a structured scenario has "Entry points" which appear to be extension points. However, we can choose a "Join" point where the basic path will continue after the 'alternate' or 'exception' scenario completes.

There are other differences, such as an Alternate scenario not being allowed to have Alternate scenarios of their own. So, this does not fit the definition of an extension use-case, which is allowed to have extensions.

So, I guess structured scenarios are in fact scenarios as I think of them. Especially as the EA docs state "A scenario is a real-world sequence of operations that you create to describe how an element works in real-time". However, other things confuse me, such as the option to link a step of a scenario to a use-case, either as Include, Extend or Invokes (can someone explain this?).

Having rambled on, I guess I'll ask if people use structured scenarios to document use-cases, or just scenarios (i.e. paths through use-cases) or both. If they are true scenarios, then I'm guessing you have 'actual data/values' in your scenario steps (Actions). I'd like to know how people use this feature, and if in fact people use it in different ways.

Thanks.

Pages: [1]