Path: utzoo!attcan!uunet!lll-winken!ncis.llnl.gov!ncis!helios.ee.lbl.gov!pasteur!ames!vsi1!daver!lynx!m5
From: m5@lynx.uucp (Mike McNally)
Newsgroups: comp.unix.wizards
Subject: Third party rcp
Message-ID: <5196@lynx.UUCP>
Date: 18 Jan 89 01:30:27 GMT
Reply-To: m5@lynx.UUCP (Mike McNally)
Distribution: na
Organization: Lynx Real-Time Systems Inc, Campbell CA
Lines: 32

Is there a good reason that rcp does not check to see if a third-party
copy is to take place on the same machine, and if so to start a "cp" instead
of an "rcp" there?  Seems to me that that would save some time (not much;
I just wondered why it wasn't done).

The only reason I could think of would be the possibility that the remote
host would interpret its own name (as far as the local system is concerned)
differently.  For example:

	rcp zebra:xyz zebra:abc

The local host has its own idea of what "zebra" means, and will send that
host an rcp command like this:
	
	rsh zebra -n rcp xyz zebra:abc

I suppose that the machine thought to be "zebra" by the local host may call
itself something else, but that seems pretty bad, in the sense that not a
whole lot of stuff would work if that were the case.  Seems like rcp on
the local host could determine that the copy lay entirely within zebra, and
request a simple cp:

	rsh zebra -n cp xyz abc

Maybe I'm wrong, though.


-- 
Mike McNally                                    Lynx Real-Time Systems
uucp: {voder,athsys}!lynx!m5                    phone: 408 370 2233

            Where equal mind and contest equal, go.