RE: [Swarm-Support] Time efficiency of at$createActionTo$message

[Marshall, James A R]

Ah, right, OK. I think I remember a discussion on this issue some time
back... I'll have to look at the archives to see if that's also relevant to
my problem, although I think it's not the root cause as the Selector side of
at$createActionTo$Message shouldn't be dependent on where in the schedule
the message it's inserted?
        James

---
Dr James A R Marshall
Complex Systems Modelling Group (COSMIC)
Department of Earth Science and Engineering
Imperial College London
Tel: +44 (0)20 7594 7493
Fax: +44 (0)20 7594 7444
Container World Project - http://www.ese.ic.ac.uk/research/containerworld/



-----Original Message-----
From: Marcus G. Daniels [mailto:address@hidden
Sent: 18 February 2003 17:49
To: address@hidden
Subject: Re: [Swarm-Support] Time efficiency of
at$createActionTo$message


Marshall, James A R wrote:

>  sorry to be dense but what are the ObjectiveC style constructors in
>Java... are those the interfaces/classes postfixed with S, C, SImpl, CImpl,
>etc.?
>
Oh, I see wrote that e-mail too fast.  I meant to use the word 
"Selectors".  The object you pass to createActionTo$message.   The 
`messaging' aspect of Objective C (by itself, but very much when 
combined with Java) is costly in Swarm.  FAction calls don't have that 
overhead.  In the example a Selector is used to setup the FAction call, 
but it is only used as a convenience. The Selector in the example uses 
Java reflection to identify the method -- once identified the method is 
fast to call, unlike when the Selector is used directly in 
createActionTo$message.


_______________________________________________
Support mailing list
address@hidden
http://www.swarm.org/mailman/listinfo/support