Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!cs.utexas.edu!swrinde!ucsd!ucbvax!van-bc!ubc-cs!alberta!bode!eric
From: eric@bode.ee.ualberta.ca (Eric Norum,EB 547,4626)
Newsgroups: comp.sys.next
Subject: Re^2: printf bug
Message-ID: <70@bode.ee.ualberta.ca>
Date: 5 Jul 90 20:44:13 GMT
Reply-To: eric@bode.ee.ualberta.ca (Eric Norum)
Organization: University of Alberta Electrical Engineering
Lines: 19

Looks like there were a couple of problems with my article describing
what I believe to be a bug in printf.

1. I don't know how the reply field got trashed.  The reply field
   should have been: eric@bode.ee.ualberta.ca

2. I didn't state my problem clearly enough.  The bug is in the way
   the `%g' format of the C printf function handles floating point
   numbers with exponents between -1 and -4, inclusive.  My comments
   about gnuplot were confusing.  It was when working with gnuplot that
   I discovered the problem with printf.

Summary: on the NeXT:
	printf ("%.3g\n", 0.0001234);
prints 0, and I think it should print 0.000123
Eric Norum
Dept. of Electrical Engineering      eric@bode.ee.ualberta.ca
University of Alberta
Edmonton, Canada.                   phone: (403) 492-4626