Re: [boost] [msm] on_entry must be implemented for every initial events for every states in substatemachines.

Subject: Re: [boost] [msm] on_entry must be implemented for every initial events for every states in substatemachines.
From: Richard Szabo (sz.richard_at_[hidden])
Date: 2010-09-02 04:10:15

Hi Christophe

Now this is the right attachment.

Br.
Richie

On 1 September 2010 21:24, Richard Szabo <sz.richard_at_[hidden]> wrote:
> Hi Christophe
>
> Oh yes wrong file attached I have no luck with attachment recently :(.
>
> I will resend the post tomorrow morning in CET time. ( I have it in
> the office only and I have left it already )
>
> Sorry about that. :(.
>
> Br.
> Richie
>
> On 1 September 2010 19:34, Christophe Henry
> <christophe.j.henry_at_[hidden]> wrote:
>>> Hi All
>>>
>>> In the following little example see attachment,
>>> I have a problem that in the RunningStateMachine::State2 the on_entry
>>> function must be implemented with event1 and with event3 as well
>>> but from the logic point of view there is no possibility to call
>>> on_entry in RunningStateMachine::State2 with event1.
>>>
>>> actually at runtime there is no call is happening to to the function :
>>> void on_entry(event1 const&,FSM& ) {std::cout << "entering:
>>> State2:Event1" << std::endl;}
>>>
>>> but it must be defined and even implemented ( if no implementation
>>> provided linker error occurs).
>>>
>>> Is it a restriction/bug or I'm missing something ?
>>> It is only happening with initial events for substatemachines ?
>>> Or is there a kind of pattern to know what events can enter exit a
>>> state in a top of the ones present in the transition tables ?.
>>>
>>> Br.
>>> Richie
>>
>> Hi,
>>
>> I don't find this entry function or a State2. Am I missing something
>> or did you attach the wrong file?
>>
>> Regards,
>> Christophe
>> _______________________________________________
>> Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
>>
>