Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site sdcsvax.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxl!houxm!houxz!vax135!cornell!uw-beaver!tektronix!hplabs!sdcrdcf!sdcsvax!laman
From: laman@sdcsvax.UUCP (Mike Laman)
Newsgroups: net.bugs.usg
Subject: System V.2 Curses #4
Message-ID: <1115@sdcsvax.UUCP>
Date: Fri, 10-Aug-84 19:45:49 EDT
Article-I.D.: sdcsvax.1115
Posted: Fri Aug 10 19:45:49 1984
Date-Received: Mon, 13-Aug-84 00:16:51 EDT
Organization: EECS Dept. U.C. San Diego
Lines: 33

The "FILE" type variable "outf" is used throughout the curses library for
debugging output.  There are several causes where it is assumed to NOT be zero;
i.e. the value is not checked before using it.  It is possible that "outf" is
zero, and since the library is very careful to "always" check it, I feel the
following are oversights that should be fixed.

The following routines do not properly check "outf" before using it:

	1. "draino()" (screen/draino.c) should have an "if (outf)" just before
	   the "fprintf()" that is approximately in the middle of the for loop.
	   [ Sorry, I don't have the sources on this system so I can't give you
	     any diffs. ]
	2. "mvcur()" (screen/mvcur.c) should have an "if (outf)" just before
	   the first "fprintf()" call in "_loc_right()".
	3. "_ll_refresh()" (screen/ll_refresh.c) should have an "if (outf)" just
	   before each "putc('\'', outf);" around line #118 and #142.  I made
	   the following sample change, since it is the simplest and to
	   minimize the changes to the source (shorter diffs!).

	   for (j=0; j<SP->std_body[i]->length; j++)
	   {
		n = SP->std_body[i]->body[j];
		if (n & A_ATTRIBUTES)
		{
			if (outf) /* Add this line <------------- */
			putc('\'', outf);
		}
		:
		:
	   }

		Mike Laman, NCR @ Torrey Pines
		UUCP: {ucbvax,philabs,sdcsla}!sdcsvax!laman