Re: [boost] [review] FSM Review Starts Today August 11th

From: Andrey Semashev (andrey.semashev_at_[hidden])
Date: 2008-08-12 11:55:12

Phil Endecott wrote:
> Martin Vuille wrote:
>> The formal review for Andrey Semashev's Finite State Machines (FSM) 
>> library begins today and will run until the 20th.
>>
>> The documentation (overview and reference) is available online:
>> http://boost-extension.redshoelace.com/docs/boost/fsm/doc/state_machine.html 
>>
>> http://boost-extension.redshoelace.com/docs/boost/fsm/doc/reference.html
> 
> I have a question about the lifetime of the state objects.
> 
> I have the impression that the state objects are created when then FSM 
> is created and destroyed when it is destroyed.  Is this correct?

That's right.

> Thinking about my own FSM code, implemented ad-hoc, I have variables 
> whose lifetime should correspond to the period spent in a particular 
> state.  (Or, more generally, the period spent in a hierarchical group of 
> states.)  It looks like I can get this only by using the enter/leave 
> hooks and a pointer.  Is this correct?

Yes, you can use enter/leave handlers. However, I would consider using 
boost::optional in order to avoid dynamic memory allocation.