Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84 SMI; site sun.uucp
Path: utzoo!watmath!clyde!burl!ulysses!gamma!epsilon!zeta!sabre!bellcore!decvax!decwrl!sun!guy
From: guy@sun.uucp (Guy Harris)
Newsgroups: net.sources.games
Subject: Re: hack.tty.c for SysVR2
Message-ID: <2159@sun.uucp>
Date: Sat, 4-May-85 06:08:11 EDT
Article-I.D.: sun.2159
Posted: Sat May  4 06:08:11 1985
Date-Received: Mon, 6-May-85 00:55:53 EDT
References: <399@ho95b.UUCP>
Organization: Sun Microsystems, Inc.
Lines: 31

> #ifdef BSD

This should be "#ifdef V7"; it's not the BSD driver that's being used, it's
the V7 driver (the BSD driver is a V7 superset).

> #define	CBREAK		ICANON

Nope.  Turning CBREAK *on* is equivalent to turning ICANON off.  However,
the later code doesn't use CBREAK for the USDL driver, and does recognize
that the sense of the bit is reversed, so this may not be a problem.

> #ifdef BSD
> register int cf = (flags.cbreak == ON) ? CBREAK : 0;
> #else
> register int cf = (flags.cbreak == ON) ? 0 : ICANON;
> #endif
> ...
> #ifdef BSD
> 	if((curttyb.sg_flags & CBREAK) != cf){
> 		curttyb.sg_flags &= ~CBREAK;
> #else
> 	if((curttyb.sg_flags & ICANON || curttyb.c_cc[VMIN] > 1 ) != cf){

I'd remove the test of c_cc[VMIN] if I were you - the appearance of the
"||" means that the LHS of the comparison will be 1 or 0, but the RHS is
"cf" which is either ICANON or 0 - ICANON is, if I remember correctly, 2 not
1.

Other than that, it looks sane.

	Guy Harris