Our existing diagram implementation allows the user to draw a message
from a lifeline back to itself.

It's not very obvious how to do this. Try the following.

1. Start drawing a message horizontally out from the lifeline
2. Without releasing the mouse move the mouse down, the line won't
follow but carry on all the same.
3. Move the mouse back to the lifeline and release.

I assume this is a part implementation of the "more()" message
demonstrate in Fig 89 of UML Spec 1.4.2 -
http://argouml-sequence.tigris.org/fig89.png

I'm wondering about how to implement this better from the users
perspective in the new implementation.

For normal messages at the moment the user has to draw a call message
and then draw a delete message. This would seem painful for a self
referencing message (and could possibly even be automated more for non
self-referencing).

For self reference messages I'm think that as soon as the user draw a
message back onto the same lifeline then the stacked activation should
get created and positioned from where the draw started to where the
draw ended and both the call and return message are created and
position at either end.

Another way might be by a new tool where rather than drawing an edge
the user clicks a "Stacked Activation" tool and then simply clicks
anywhere on another existing activation.

Could I have some feedback on any ideas for this before I create an
issue for Christian.

Regards

Bob.

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to