On Tue, Sep 13, 2011 at 1:43 PM, Dario D <[email protected]> wrote:
> 2011/9/13 Rahul Akolkar [email protected]
>
>>
>> When the state machine is in "state2", there are three candidate
>> transitions that match event "action.next" -- two with guards in
>> "state2" and one unguarded or condition-less in state "_processing".
>> The two in "state2" are not followed, but subsequently the unguarded
>> one is selected and followed which leads to state "_end". Given state
>> "_end" is a final state, no further events triggered on the state
>> machine will have any effect.
>>
>>
>
> Thank you Rahul. There is no way to "catch" the parsing error if it happens
> in such conditions, and consequently trigger an error event?
<snip/>

If thats what is desired, the state machine can be changed to reflect
that. There isn't any inherent limitation precluding those kinds of
patterns. There are a number of ways in which the state machine can be
changed here, such as:

 * Fix the root cause that causes ill-formed expressions to appear
 * Transition to a non-final state that can still process "error.*" events
 * Use better ontology for events, rather than one "action.next" that
is shared at different "path fragment depths"

-Rahul

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to