Fix a regression in term.c rev. 1.89 reported by bentley@:

In UTF-8 output, do not print anything if mchars_spec2cp() returns 0.
In particular, this repairs handling of zero-width spaces (\&).

While here, let mchars_spec2cp() return 0xFFFD instead of -1
if the character is not found, simplifying the using code.
In HTML output, do not print obfuscated ASCII characters and
do not test for one-char escapes, mchars_spec2cp() already does that.

4 files changed, 15 insertions, 16 deletions

diff --git a/usr.bin/mandoc/chars.c b/usr.bin/mandoc/chars.c
index 961769c80f0..00713a9cd30 100644
--- a/usr.bin/mandoc/chars.c
+++ b/usr.bin/mandoc/chars.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: chars.c,v 1.31 2014/10/26 18:06:28 schwarze Exp $ */
+/*	$OpenBSD: chars.c,v 1.32 2014/10/27 13:29:30 schwarze Exp $ */
 /*
  * Copyright (c) 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011, 2014 Ingo Schwarze <schwarze@openbsd.org>
@@ -102,7 +102,7 @@ mchars_spec2cp(const struct mchars *arg, const char *p, size_t sz)
 	const struct ln	*ln;
 
 	ln = find(arg, p, sz);
-	return(ln != NULL ? ln->unicode : sz == 1 ? *p : -1);
+	return(ln != NULL ? ln->unicode : sz == 1 ? *p : 0xFFFD);
 }
 
 char
diff --git a/usr.bin/mandoc/html.c b/usr.bin/mandoc/html.c
index 219671ec597..47c59297012 100644
--- a/usr.bin/mandoc/html.c
+++ b/usr.bin/mandoc/html.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: html.c,v 1.47 2014/10/26 17:11:18 schwarze Exp $ */
+/*	$OpenBSD: html.c,v 1.48 2014/10/27 13:29:30 schwarze Exp $ */
 /*
  * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org>
@@ -455,11 +455,12 @@ print_encode(struct html *h, const char *p, int norecurse)
 			break;
 		case ESCAPE_SPECIAL:
 			c = mchars_spec2cp(h->symtab, seq, len);
-			if (c > 0)
+			if (c <= 0)
+				break;
+			if (c < 0x20 || c > 0x7e)
 				printf("&#%d;", c);
-			else if (-1 == c && 1 == len &&
-			    !print_escape(*seq))
-				putchar((int)*seq);
+			else if ( ! print_escape(c))
+				putchar(c);
 			break;
 		case ESCAPE_NOSPACE:
 			if ('\0' == *p)
diff --git a/usr.bin/mandoc/mandocdb.c b/usr.bin/mandoc/mandocdb.c
index fb1e881826b..e4e0983ef39 100644
--- a/usr.bin/mandoc/mandocdb.c
+++ b/usr.bin/mandoc/mandocdb.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: mandocdb.c,v 1.121 2014/10/12 20:32:16 schwarze Exp $ */
+/*	$OpenBSD: mandocdb.c,v 1.122 2014/10/27 13:29:30 schwarze Exp $ */
 /*
  * Copyright (c) 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org>
@@ -1893,7 +1893,7 @@ render_key(struct mchars *mc, struct str *key)
 		 */
 
 		if (write_utf8) {
-			if (0 == (u = mchars_spec2cp(mc, seq, len)))
+			if ((u = mchars_spec2cp(mc, seq, len)) <= 0)
 				continue;
 			cpp = utfbuf;
 			if (0 == (sz = utf8(u, utfbuf)))
diff --git a/usr.bin/mandoc/term.c b/usr.bin/mandoc/term.c
index b64b49095f3..33ed30422c4 100644
--- a/usr.bin/mandoc/term.c
+++ b/usr.bin/mandoc/term.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: term.c,v 1.89 2014/10/26 17:11:18 schwarze Exp $ */
+/*	$OpenBSD: term.c,v 1.90 2014/10/27 13:29:30 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2010-2014 Ingo Schwarze <schwarze@openbsd.org>
@@ -466,9 +466,8 @@ term_word(struct termp *p, const char *word)
 					encode(p, cp, ssz);
 			} else {
 				uc = mchars_spec2cp(p->symtab, seq, sz);
-				if (uc <= 0)
-					uc = 0xFFFD;
-				encode1(p, uc);
+				if (uc > 0)
+					encode1(p, uc);
 			}
 			break;
 		case ESCAPE_FONTBOLD:
@@ -703,9 +702,8 @@ term_strlen(const struct termp *p, const char *cp)
 				} else {
 					c = mchars_spec2cp(p->symtab,
 					    seq, ssz);
-					if (c <= 0)
-						c = 0xFFFD;
-					sz += cond_width(p, c, &skip);
+					if (c > 0)
+						sz += cond_width(p, c, &skip);
 				}
 				break;
 			case ESCAPE_SKIPCHAR: