1
General Board / Concept of "Is Return" message in sequence diagram
« on: April 01, 2009, 08:50:15 pm »
Hello everyone,
I have a problem to understand the concept of the checkbox "Is Return" of dialog "Message Property" on sequence diagrams.
In my opinion I should be able to select an operation from the called object as message if I check "Is Return". But I can only select an operation from the caller.
Another possibility would be that EA reverts the arrow if I select "Is Return".
But neither is the case.
Am I missing the point here?
In more detail:
- I have two Objects: A and B
- Object B has an operation: calculate(float, float): float
- I draw a message from A to B and select this operation as message
- EA shows the signature of this message.
Now I want to model the return value explicitly
- I draw a message from B to A
- and tick "Is Return".
But I can't select an operation from Object B.
So I would have to add the same operation to A
in order to show the correct signature.
But this is semantically wrong.
A doesn't have this operation at all - it uses it.
or
- I draw a message from A to B
- and select the operation from B
- EA shows the correct signature -
but the arrow points in the wrong direction
and the complete diagram semantics is wrong now.
I have a problem to understand the concept of the checkbox "Is Return" of dialog "Message Property" on sequence diagrams.
In my opinion I should be able to select an operation from the called object as message if I check "Is Return". But I can only select an operation from the caller.
Another possibility would be that EA reverts the arrow if I select "Is Return".
But neither is the case.
Am I missing the point here?
In more detail:
- I have two Objects: A and B
- Object B has an operation: calculate(float, float): float
- I draw a message from A to B and select this operation as message
- EA shows the signature of this message.
Now I want to model the return value explicitly
- I draw a message from B to A
- and tick "Is Return".
But I can't select an operation from Object B.
So I would have to add the same operation to A
in order to show the correct signature.
But this is semantically wrong.
A doesn't have this operation at all - it uses it.
or
- I draw a message from A to B
- and select the operation from B
- EA shows the correct signature -
but the arrow points in the wrong direction
and the complete diagram semantics is wrong now.