home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.shell
- Path: sparky!uunet!munnari.oz.au!metro!physiol.su.OZ.AU!john
- From: john@physiol.su.OZ.AU (John Mackin)
- Subject: Re: recursive mkdir
- Message-ID: <1992Aug20.052006.536@physiol.su.OZ.AU>
- Summary: Sequent busted the shell
- Reply-To: John Mackin <john@civil.su.oz.au>
- Organization: Department of Physiology, University of Sydney, Australia
- References: <glaze.093@glaze> <1992Aug18.033048.7574@physiol.su.OZ.AU> <1992Aug18.161744.27823@u.washington.edu>
- Date: Thu, 20 Aug 1992 05:20:06 GMT
- Lines: 45
-
- In article <1992Aug18.161744.27823@u.washington.edu>,
- donn@carson.u.washington.edu (Donn Cave) writes:
-
- > The script you posted fails under DYNIX 3.1, on a Sequent Symmetry: [...]
- > This is basically a BSD 4.2 system, and the shell is so old that not only
- > does it lack shell functions, it also behaves a bit different with separator
- > characters. My recollection is that it specifically exempts "/" from the
- > separator characters when expanding expressions, for some reason.
- >
- > | IFS="/$IFS"
-
- Donn,
-
- I won't come out and flatly say `I don't believe you,' but I don't
- believe you :).
-
- I don't doubt that the script does fail on the system you mention in
- the way you say. What I _don't_ believe is that any real, un-hacked
- Bourne sh ever worked like that. You see, my native machines these
- days are DECstations running Ultrix. Their /bin/sh is an extremely
- primitive version as well, also straight out of 4.2BSD. It works fine
- with that sh.
-
- Not being the kind of guy who is easily satisfied, I decided I would
- _really_ check this out. You might be surprised to learn that there
- are still PDP-11s running Seventh Edition up and working on this
- campus. Furthermore, I have an account on one (smiles shyly). I
- just logged in there and checked this, and I can assure you that
- a V7 sh on a PDP-11 does NOT treat / in IFS specially. [Boyd:
- it was psych44.]
-
- This allegation caused me some concern, since I make it my business to know
- the history of the commands, especially crucial ones like sh, and know
- what is portable and what isn't. I'm very pleased to have found in this
- case that I did have the history correctly in mind. What I didn't know
- here is that Sequent broke the version of sh that they shipped with DYNIX
- 3.1. I'm not too upset about that, having never logged in to a
- Sequent Symmetry.
-
- Moral: the script _is_ properly portable. Complaints to your vendor. :)
-
- --
- John Mackin <john@civil.su.oz.au>
- `Flood pinging the broadcast address should only be done under very
- controlled conditions.' -Mike Muuss
-