Book a Demo

Author Topic: SubMachineState Issues  (Read 3763 times)

Paolo F Cantoni

  • EA Guru
  • *****
  • Posts: 8626
  • Karma: +259/-129
  • Inconsistently correct systems DON'T EXIST!
    • View Profile
SubMachineState Issues
« on: November 06, 2006, 05:58:49 am »
In another topic: [size=13]Suppress display of concurrent substates[/size], I suggest to jjweis that a formal bug report be sent to Sparx regarding the inability to suppress sub-machine entry/exit points and concurrent state regions from the holonym level view of the sub-machine.

However, now that we are starting to use them "in anger" a number of other issues have now made themselves apparent.

The major issue is that there doesn't seem to be any mechanism to formally associate a particular set of sub-states and transitions with a particular concurrent sub-state.

One unfortunate and very annoying side-effect of this is that adjusting the concurrent sub-state regions in either the meronym diagram or the holonym diagram will affect the other!  In the process, any "layout" (since that's all it is at present) of the various states and transitions into one concurrent sub-state versus another is lost...

It's not clear if there is a definitive diagram that holds the definition of the composite state machine.  You seem to be able to add states in all sorts of copies of the diagram.  I would have thought that (unlike most other diagrams) there is an explicit link between the element in the browser and its composite diagram such that all other instances but that are represented solely as the state vertex and with optional entry/exit points and regions (as discussed in the other topic).  All other detail should, by definition, be suppressed.

Thoughts?  Votes?
Paolo
[size=0]©2006 Paolo Cantoni, -Semantica-[/size]
Inconsistently correct systems DON'T EXIST!
... Therefore, aim for consistency; in the expectation of achieving correctness....
-Semantica-
Helsinki Principle Rules!

«Midnight»

  • EA Guru
  • *****
  • Posts: 5651
  • Karma: +0/-0
  • That nice Mister Grey
    • View Profile
Re: SubMachineState Issues
« Reply #1 on: November 06, 2006, 10:45:06 am »
As with boundaries EA appears to disregard - but to a lesser extent - the ownership relationship between a state and its substates.

Layout used to produce extremely weird results with substates, where "weird" should be read as "unacceptable." I have not had the courage to try this for quite a while.

So, at least as far as the ability to recognize and respect (sub)State element ownership, I agree.

As to the rest of your post, I think you are correct, but have not had the opportunity to work through the issue.
No, you can't have it!