I've been toying with the idea of using gkermit
over an ssh connection. As it stands, gkermit is essentially
featureless, a GPL kermit implementation that's enough to
get the Free software advocated off their backs. However,
since they want to continue to sell C-Kermit, it has all
the `useful' features, such as scripting, but more importantly
the ability to create a connection over which to transfer files.
That's where ssh comes in. Since gkermit has no interactive
mode, only a command line interface, the only way to use it
with ssh would be in a command like:
gkermit -s hello.c |ssh user at host gkermit -r
But what if I'm connecting to a telnet (or, more precisly,
ssh) BBS? What if I want to use ZModem instead of kermit?
I think a useful feature of ssh would to be able to replace
the current input/output stream with that of a spawned
programs, returning i/o control to the tty once the program
exits.
I envision a new escape sequence, like ~| or something,
that would send all input from the remote machine to a
program, and spit all output from the program back to the
remote machine.
I am going to try and code this, 'cause I think it would
be fun. I'm looking for any suggestions, or pointers
on how to go about implementing this (or reasons why
I shouldn't -- yes, I already know about scp...).
Thanks,
Damian
--
Damian Gryski ==> dgryski at uwaterloo.ca | Linux, the choice of a GNU
generation
512 pt Hacker Test score = 37% | 500 pt Nerd Test score = 56%
geek / linux zealot / coder / juggler / sysadmin
