Re: GetLastEventData() new elementary action

Emmanuel BEUQUE (ebeuque@pratique.fr)
Tue, 3 Feb 1998 11:52:32 +0100

À (At) 15:25 +0100 2/02/98, Andreas Kraft écrivait (wrote) :
>> I understand that this may be a problem for an engine. While an event is
>> only generated at a time and checked immediately, several link effects
>> could have been added to the execution queue, E.G. a link effect action can
>> fire another link.
>>
>> I think that programmers of engine would consequently ask for some
>> restrictions, for instance, the following one :
>>
>> "GetLastEventData only returns the eventData associated with the last
>> generated event. It should generally be called as the first elementary
>> action of the linkEffect in order to prevent other link to be fired before
>> its evaluation. The author should also be aware of possible side effects if
>> several links may fire on the same event."
>
>I think it should read:
>
>"GetLastEventData returns the associated eventData for the currently
> processed event."
>
>There is the possibility that the currently processed event is _not_ the last
>one generated. The GetLastEventData action does only make sense if I can get
>the eventData for the current event.
>I don't think that it complicates implementations. You have to store the
>eventData for the comparison of the eventData and the associated eventData
>anyway.

I fully agree with you Andreas. I would really like the GetLastEventData to
behave like that. In fact, I was just playing the devil's advocate because
I've never seen any discussion on this point in the MUG mailing list.

I think it's important that this point could be discussed before your
meeting in Berlin. I wish I could be there, but as an independent
developer, I definitely cannot affort the trip. But if some points could be
discussed during a phone conference before the resolutions of the MTF are
stated, this could be great for me.

Regarding the GetLastEventData action to be added, it would be perfect for
me if the standard ensure that the eventData corresponds to the event that
fired the link that contains this elementary action.
But, IMHO, again playing the devil's advocate, I imagine that this could be
hard for some implementors if they put all actions of the link effect in a
common execution stack, without knowing from where an abitrary elementary
action come from when it is executed.
Again, if it acts like you said, that's marvellous, perfect, from an author
point of view, and I fully subscribe to your suggestion. I was just trying
to have the opinion of some implementors.

If they don't react to this discussion, then we could assume that they
agree with what you propose :-)

Best regards,
Emmanuel

__________________________________________________________________
Emmanuel BEUQUE mailto:ebeuque@pratique.fr
MediaServ tel: +33 (0) 2 99 64 35 64
Multimedia consultant & developer fax: +33 (0) 2 99 64 36 65