Janus Dam Nielsen <[EMAIL PROTECTED]> writes:

Hi Janus, thanks for the reply.

> I will take a look at it when I get home tonight. My initial though,
> without having read your post, is that: Yes you can do it in Java but
> it won't be pretty.
>
> --
> Janus

Right, that's the conclussion I arrived at too. Without function
pointers, callbacks are no fun. Python also has the advantage of
anonymous functions (lambda expressions).

I discussed it with Mikkel, and he commented that the comparison was
biased since you wouldn't even start like that in Java to begin with.

One would probably represent the expression tree explicitly and evaluate
it starting with the leaf nodes.

If all nodes are associated with a program counter (as in VIFF), then
that could be used as a key to lookup the right node to evaluate when
data arrives from the network. If a add-node has both inputs ready, then
it is removed and its parent then has one less input to wait for. Yeah,
that could definitely be programmed... :-)

Using a thread to receive listed for input, and another thread to work
on reducing the tree, then this should work nicely. Perhaps a third
thread to send out messages produced by the tree reducer...

Tomas, I guess this design is similar to the one you though of when we
discussed alternatives for the SIMAP runtime?

-- 
Martin Geisler

Attachment: pgpoJebFaQlns.pgp
Description: PGP signature

_______________________________________________
viff-devel mailing list (http://viff.dk/)
[email protected]
http://lists.viff.dk/listinfo.cgi/viff-devel-viff.dk

Reply via email to