Jeffrey Juliano <[EMAIL PROTECTED]> writes:
> I think the problem lies in a text/binary mode thing. ssh requires
> cygwin be in the mode where it expects not to see ^M's at the end of
> lines. ssh (as compiled) doesn't run if you don't put cygwin in this
> mode. However, I suspect that emacs is piping into it dos-style EOL
> characters.
Hm. There is a variable rcp-rsh-end-of-line which is used. The
default value is "\n" which led me into thinking that Unix-style EOLs
would be used at all times.
But of course, that was a foolish thing to do: I had completely
forgotten about the automatic eol conversion done by Emacs. Does it
help if you frob the process coding system of the SSH process? The
function set-process-coding-system does this.
If you find that changing the process coding system helps, I suggest
that you set process-coding-system-alist such that the right coding
system is used for SSH.
However, I must admit that I had been restricting myself to Unixish
systems when developing rcp.el -- I thought it wouldn't be possible to
connect any other system via rsh (or telnet, or the like) anyway. But
I had completely forgotten about the client side!
So, if there is someone out there who groks this kind of stuff, I'm
happy to accept suggestion on how to deal with the coding system. (I
also have the vague feeling that rcp.el might now work well in CJKV
locales, but I know too little of i18n issues to be sure.)
kai
--
Life is hard and then you die.
