1. Pete's RDC
Has anyone other than me been experimenting with Pete Eberlein's RDC.e (Remote
Dynamic Calls) routines?
There appear to be a LOT of uses for something like this. So far, I have a
prototype of a multi-player word game which operates over the network. Once I
add some graphics to it, I'll send a copy to Rob to post.
At the moment, I am wondering how difficult it would be to monitor multiple
ports, so that each client could have its own connection with the server.
Anyone know? Pete, are you still around?
Regards,
Irv
2. Re: Pete's RDC
----- Original Message -----
From: Irv Mullins <irv at ELLIJAY.COM>
To: <EUPHORIA at LISTSERV.MUOHIO.EDU>
Sent: Friday, January 28, 2000 12:46 PM
Subject: Pete's RDC
> Has anyone other than me been experimenting with Pete Eberlein's RDC.e
(Remote
> Dynamic Calls) routines?
> There appear to be a LOT of uses for something like this. So far, I have a
> prototype of a multi-player word game which operates over the network.
Once I
> add some graphics to it, I'll send a copy to Rob to post.
>
> At the moment, I am wondering how difficult it would be to monitor
multiple
> ports, so that each client could have its own connection with the server.
> Anyone know? Pete, are you still around?
I have found that several users can connect to the same port. It's rather
common on irc, with most people connecting to port 6667 or such on the irc
server. My httpd will serve several files to several places at once on port
80, as will commercial web servers. It does get a little more complicated,
each connection has a different id, and each variable associated with a
specific id has to have that id too,,, in Eu i'd guess that all vars would
then be sequences, and you'd use the id to pick the subseq for that id,,
like: connection_time[12] for id 12.
Greg coded the robobot server (in the mirc code i posted) so that this could
*not* happen, so that other people on the net couldn't connect while i was
running the code on my puter for my own use. (Actually, he did it for
himself, people can't get to my code thru the firewall unless i am sitting
here to authorise it.)
For connecting to the Altavista Babelfish site, i have this code:
set %Bable.room [ $+ [ %Bable.num ] ] $1
/sockopen Bable [ $+ [ %Bable.num ] ] %Bable.ip 80
which tags the socket with "Bable" & Bable.num. I can then open another
socket on the same port with the id of "Bable" & Bable.num+1 if i want to.
And i know that a report coming back on Bable & Bable.num is to be sent to
Bable.room & Bable.num when i detect data on that port, and i can tell the
socket name by using $socket, or in Eu- Socket=WsockAccept(Server). Now i
wonder just how many clients can be connected to the same port......
In short,, i wouldn't use the different ports for each connection,, you
never know when you'll run out of free ports to open.
Kat
3. Re: Pete's RDC
----- Original Message -----
From: Kat <gertie at ZEBRA.NET>
To: <EUPHORIA at LISTSERV.MUOHIO.EDU>
Sent: Friday, January 28, 2000 2:36 PM
Subject: Re: Pete's RDC
> > Has anyone other than me been experimenting with Pete Eberlein's RDC.e
> (Remote
> > Dynamic Calls) routines?...
> I have found that several users can connect to the same port. It's rather
> common on irc, with most people connecting to port 6667 or such on the irc
> server. My httpd will serve several files to several places at once on
port
> 80, as will commercial web servers. It does get a little more complicated,
> each connection has a different id, and each variable associated with a
> specific id has to have that id too,,, in Eu i'd guess that all vars would
> then be sequences, and you'd use the id to pick the subseq for that id,,
> like: connection_time[12] for id 12.
Thanks, KAT. I had set up a user list on the server, and wrote client
routines
so they send a "packet" containing the user id, the remote routine_id, and
the arguments. That seems to work fine as long as each client connects,
calls the
remote function, and immediately disconnects, freeing the socket for other
clients.
One thing: in order to update the clients' screens, each client has to send
a
request_for_update periodically - I'm using a 1 second timer to trigger the
request. Is
there a better way to handle this? Can the server "broadcast" to all
clients?
Thanks,
Irv
4. Re: Pete's RDC
I was under the impression that Pete had updated RDC to cope with multiple
connections...possibly other ports as well?
Monty King
>Has anyone other than me been experimenting with Pete Eberlein's RDC.e
(Remote
>Dynamic Calls) routines?
>There appear to be a LOT of uses for something like this. So far, I have a
>prototype of a multi-player word game which operates over the network. Once
I
>add some graphics to it, I'll send a copy to Rob to post.
>
>At the moment, I am wondering how difficult it would be to monitor multiple
>ports, so that each client could have its own connection with the server.
>Anyone know? Pete, are you still around?
>
>Regards,
>Irv
