[16896] in Athena Bugs
Re: early-release, bindkey
daemon@ATHENA.MIT.EDU (Alicia Allen)
Wed Jun 16 06:49:12 1999
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Message-Id: <v04020a1ab38d3043520f@[18.162.0.220]>
In-Reply-To: <199906160233.WAA24299@contents-vnder-pressvre.mit.edu>
Date: Wed, 16 Jun 1999 06:48:46 -0400
To: John Hawkinson <jhawk@mit.edu>
From: Alicia Allen <iggy@MIT.EDU>
Cc: bugs@mit.edu
Eudora doesn't seem to handle "In-Reply-To:" lines, sorry...
>Didn't your [authority figure] tell you not to use csh scripts? ;-)
>
nope.
out of curiosity, and the desire to have abetter clue about athena&UNIX,
why not?
>A not-too-well-known fact is that csh scripts execute dotfiles
>unless you use -f. I.e. your script should be using:
>
> #!/bin/csh -f
>
>you'll find it is faster that way, too.
>
Why should one? I mean, I can see perfectly well that:
man csh
....
-f Fast start. Read neither the .cshrc file, nor the
.login file (if a login shell) upon startup.
but why would you want to or not want to read that file? As I think of it,
I can't see why it has you read that, if you assume that the .cshrc is
already read on startup, in the first place.
>On the other hand, this is certainly still a bug, yes.
>/bin/csh should be a supported shell and /usr/athena/lib/init/cshrc
>shouldn't use tcsh-isms without verifying that the shell
>one is using is tcsh.
>
>(The problem here is that you run csh, which runs your .cshrc,
>which runs /usr/athena/lib/init/cshrc, which has tcsh-specific
>stuff (bindkey); you could also change the script to #!/bin/tcsh -f
>(or even omit the -f, but that would be silly).)
>
Is csh not the default? I'm not clear that I purposely change my shell,
then again, I've had so much advice and so many customizations over the
years, that you never know...
-alicia
----------------------------------------------
Alicia Allen iggy@mit.edu
MIT Resnet Coordinator (617) 253-0191
MIT Network Security N42-140
----------------------------------------------
