Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uwm.edu!uakari.primate.wisc.edu!ames!amdahl!altos!altos86!ti From: ti@altos86.Altos.COM (Ti Kan) Newsgroups: comp.unix.wizards Subject: Re: SYS V Bourne Shell .shrc file Message-ID: <154@altos86.Altos.COM> Date: 11 Jan 90 19:20:38 GMT References: <1792@ncrcce.StPaul.NCR.COM> <1864@sauron.Columbia.NCR.COM> <11888@smoke.BRL.MIL> <1990Jan7.175807.13054@NCoast.ORG> Reply-To: ti@altos86.UUCP (Ti Kan) Organization: Altos Computer Systems, San Jose, CA Lines: 32 In article <1990Jan7.175807.13054@NCoast.ORG> allbery@ncoast.ORG (Brandon S. Allbery) writes: >I think the csh behavior cited above is more correct. I've gotten sick of >kluging around Altos's csh, which invokes .cshrc before /etc/cshrc I don't know what version of our System V on what hardware platform you have, but I have checked our csh under Altos System V on the Altos 80386 series 600, 1000 and 2000 as well as the Altos 80486 series 1000, and all of it invokes $HOME/.cshrc *after* /etc/cshrc. >a login shell, because /etc/cshrc usually contains a line like "stty kill ^u" >(Altos's default is ^X, ugh). Boo, hiss. Ctrl-X is a semi-standard kill character, as is Ctrl-U. The standard kill character dating back to who-knows-when is @ (would you use that?). Stty allows you to set it to anything you want, so this shouldn't be a problem. >Besides which, the user shouldn't >be given the opportunity to alter what /etc/cshrc does; if .cshrc is run >forst, the user can drop an alias in place of one of the commands in >/etc/cshrc. (It's not a security hole, but it *can* violate the Principle of >Least Surprise.) >Brandon S. Allbery allbery@NCoast.ORG, BALLBERY (MCI Mail), ALLBERY (Delphi) I agree with you. But, like I said, I found our csh to behave correctly in this regard. -Ti -- Ti Kan \\\ vorsprung durch technik! \\\ Internet: ti@altos.com /// \\\ UUCP: ...!uunet!altos!ti ////////\