Hi Jake, Thanks for your reply. It seems that we should re-define transition EMF data model for this tool, rebuild the tool from the base. I will read SCXML transition specification carefully and carry on the rebuild job
在 2010年8月9日 上午9:22，Jacob Beard <jbea...@cs.mcgill.ca>写道： > Hi, Long, > > Replies below: > > On 10-08-08 09:06 PM, Xun Long Gui wrote: > > I do not think we can create a transition without a target, target is a > > necessary element for a transition. > > > The spec says that the target attribute on transition is not required. > In that case, I think the semantics are similar to a Static Reaction, as > defined in the paper describing the Rhapsody semantics of statecharts. > From the spec: > > The identifier(s) of the state or parallel region to transition to. If > it is omitted, the transition will not cause a change in the state > configuration when it is executed. The executable content contained in > the transition will still be executed, so the transition will function > as a simple event handler. If the target is present and equal to the > state containing the transition element, the transition will cause the > state machine to leave and then re-enter this state. > > From: http://www.w3.org/TR/scxml/#transition > > To feature 1 and 3 in Chris's commends, we can achieve them, not a big > > thing. What do you mean about "showstoppers" bug ? > > > What I mean is that I think these things are critical to usability. > > Jake > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org > For additional commands, e-mail: dev-h...@commons.apache.org > > -- Best Regards Gui Xun Long (桂训龙)
