Don't silently skip non-ASCII characters, but replace them with ``?''.

This is less likely to break the syntax of macros.
Patch provided by joerg@.

1 files changed, 8 insertions, 5 deletions

diff --git a/usr.bin/mandoc/read.c b/usr.bin/mandoc/read.c
index 11862ac6120..d5d36ea11fe 100644
--- a/usr.bin/mandoc/read.c
+++ b/usr.bin/mandoc/read.c
@@ -1,4 +1,4 @@
-/*	$Id: read.c,v 1.5 2011/11/05 16:02:18 schwarze Exp $ */
+/*	$Id: read.c,v 1.6 2012/02/26 21:01:43 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2010, 2011 Ingo Schwarze <schwarze@openbsd.org>
@@ -315,9 +315,9 @@ mparse_buf_r(struct mparse *curp, struct buf blk, int start)
 			 * Warn about bogus characters.  If you're using
 			 * non-ASCII encoding, you're screwing your
 			 * readers.  Since I'd rather this not happen,
-			 * I'll be helpful and drop these characters so
-			 * we don't display gibberish.  Note to manual
-			 * writers: use special characters.
+			 * I'll be helpful and replace these characters
+			 * with "?", so we don't display gibberish.
+			 * Note to manual writers: use special characters.
 			 */
 
 			c = (unsigned char) blk.buf[i];
@@ -325,8 +325,11 @@ mparse_buf_r(struct mparse *curp, struct buf blk, int start)
 			if ( ! (isascii(c) && 
 					(isgraph(c) || isblank(c)))) {
 				mandoc_msg(MANDOCERR_BADCHAR, curp,
-						curp->line, pos, "ignoring byte");
+						curp->line, pos, NULL);
 				i++;
+				if (pos >= (int)ln.sz)
+					resize_buf(&ln, 256);
+				ln.buf[pos++] = '?';
 				continue;
 			}