Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site fortune.UUCP
Path: utzoo!linus!philabs!seismo!harpo!eagle!mhuxl!ihnp4!fortune!olson
From: olson@fortune.UUCP
Newsgroups: net.sources
Subject: a bug in the recently posted curses - (nf)
Message-ID: <2108@fortune.UUCP>
Date: Tue, 27-Dec-83 15:06:11 EST
Article-I.D.: fortune.2108
Posted: Tue Dec 27 15:06:11 1983
Date-Received: Thu, 29-Dec-83 01:15:39 EST
Sender: notes@fortune.UUCP
Organization: Fortune Systems, Redwood City, CA
Lines: 28

#N:fortune:10100009:000:1012
fortune!olson    Dec 27 11:38:00 1983

We discovered a bug in the 'new' libcurses distributed by Ken
Arnold several months ago.  It seems that several of the files
included have either earlier version #'s, or the same version #'s, but
earlier dates.  The only case we have verified as causing a bug so far
is in getstr.c.  The newly distributed version was 1.3, while the one we
have in our sources from Berkeley is 1.4, with a  later date.  The
bug is on the 'while' line.  The line ends with a ';', there should be
none.  As a result, the pointer is never incremented.

(Since the new curses was posted to this group, I thought it better to
post this article here, rather than to net.bugs.4bsd.)

The diff listing follows:
------------------
	6c6
	<  * 4/29/81 (Berkeley) @(#)getstr.c	1.3
	---
	>  * 7/2/81 (Berkeley) @(#)getstr.c	1.4
	12c12
	< 	while ((*str = wgetch(win)) != ERR && *str != '\n');
	---
	> 	while ((*str = wgetch(win)) != ERR && *str != '\n')
------------------
	Dave Olson, Fortune Systems {ihnp4,harpo,ucbvax!amd70}!fortune!olson