(jcwren) RE: redirection of output
by jcwren (Prior) on Sep 14, 2000 at 21:47 UTC
|
I don't know much about Quake (it makes me ill watching the screen caroming around like that), but rather than all this sockety smickity-smack, take a look at Net::IRC.
It's very easy to use, and fairly capable. The only issue I've found to date is that it doesn't internally implement any form of flood control, so you may have to use some sort of queueing mechanism. And, flood control is only an issue if you're running through an IRC server, as opposed to a DCC connection.
--Chris
e-mail jcwren | [reply] |
RE: redirection of output
by fundflow (Chaplain) on Sep 14, 2000 at 19:39 UTC
|
If you use unix, you should probably do this:
rsh machineA command
This will return the standard output, just as if you ran it on
the local machine.
If you want also to save the output to a file then do
rsh machineA 'command | tee filename'
(it is usually safer to use ssh instead of rsh, although
this requires more proccessing on each side for the encryption/decryption)
Cheers. | [reply] |
Re: redirection of output
by merlyn (Sage) on Sep 14, 2000 at 19:15 UTC
|
Well, if you have two different machines, you have to have a wire protocol to get the data across. It doesn't just magically leap from one place to another.
The simplest would be to run process X as a CGI program, then machine B can
simply request that machine A run process X and the output sent in response.
If that doesn't meet your needs, you'll have to be more specific.
-- Randal L. Schwartz, Perl hacker | [reply] |
|
|
What I am looking to do is this:
Machine A = perl program runs a dedicated quake server and pumps output of console to Machine B using a Perlbot in irc
Machine B = runs a perlbot in irc that can pump out console info and accept and send quake server commands to Machine A.
Basically a client/server admin tool using IRC.
Links:
Perlbot
Perlbot Quake3Plugin
Similar GUI based (Tcl/Tk) program used for Half-Life servers
Not the noblest of causes but its a fun project. =)
Thanks,
<a href="http://www.perlmonks.org/index.pl?node_id=16711&lastnode_id=2437"djw
| [reply] |
|
|
So, you can still use HTTP, but it's probably overkill unless you want some
sort of security.
Short of that, one solution is to set up a TCP server socket on the quake machine,
and have it broadcast whatever's showing up to any and all clients that connect.
Then write a simple TCP client on the perlbot side, and when data is ready, feed
it to the perlbot.
-- Randal L. Schwartz, Perl hacker
| [reply] |
RE (tilly) 1: redirection of output
by tilly (Archbishop) on Sep 14, 2000 at 19:37 UTC
|
I have been happy with ssh for this kind of task. rsh is also possible although it is insecure. No Perl module is needed, just use ssh to run processes remotely. | [reply] |
Re: redirection of output
by Fastolfe (Vicar) on Sep 14, 2000 at 20:37 UTC
|
You may want to see about using Net::Telnet to do this as well. This would let you connect and log in to the remote machine remotely. You could then use object methods to execute the command you want, and retrieve data line by line if necessary.
I personally consider this somewhat of a kludge, but this is probably the solution requiring the least amount of coding on your part. I would go with some of the other recommendations here, involving real TCP client-server interaction, but Net::Telnet should suffice. | [reply] |