Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!cs.utexas.edu!rutgers!njin!princeton!njsmu!mccc!jonlab!jon
From: jon@jonlab.UUCP (Jon H. LaBadie)
Newsgroups: comp.unix.questions
Subject: Re: simple question about mv
Summary: too many forks
Message-ID: <626@jonlab.UUCP>
Date: 1 Feb 89 03:23:51 GMT
References: <18231@adm.BRL.MIL> <4093@omepd.UUCP>
Organization: 4455 Province Line Rd., Princeton, NJ 08540
Lines: 33

In article <4093@omepd.UUCP>, merlyn@intelob.intel.com (Randal L. Schwartz @ Stonehenge) writes:
In article <18231@adm.BRL.MIL>, moss@BRL (VLD/VMB) writes:
| Well, this should work under a Bourne shell:
| $ for i in *.flip
| > do	mv $i `basename $i .flip` .flop
| > done
| $

Randal and others (this is not intended to pick on Randal's suggestion)
have proposed work-arounds to UNIX's command line interpreters (or to
mv's) deficiency in wild card matching.  Each have proposed a separate
mv command for each file.  This is unfortunate, but probably necessary.

However, each suggestion has also necessitated a separate process such
as cut, sed, or in Randal's case, basename, for each file to be moved.
An alternative, is one command line to create a list of "root names"
and loop through them.  For a large number of files, this could
greatly enhance the efficiency of the processing.  Ksh users could
improve upon this even more using the special substitution facilities
of that shell.

	for root in `ls *.flip | cut -d. -f1`
	do
		mv ${root}.flip ${root}.flop
	done

Potential problems still exist if a) two "."s are int the file names,
or b) not ".flip" files exist.

Isn't UNIX great?  :-)  for guru's!  ;-)
-- 
Jon LaBadie
{att, ulysses, princeton, bcr}!jonlab!jon