Fix a regression in term.c rev. 1.229 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, 11 insertions, 12 deletions

diff --git a/chars.c b/chars.c
index 950f9edc..6c217e43 100644
--- a/chars.c
+++ b/chars.c
@@ -104,7 +104,7 @@ 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/html.c b/html.c
index f1d3ad1a..95cd163e 100644
--- a/html.c
+++ b/html.c
@@ -457,11 +457,12 @@ 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/mandocdb.c b/mandocdb.c
index 3c212fc6..d03416b5 100644
--- a/mandocdb.c
+++ b/mandocdb.c
@@ -1903,7 +1903,7 @@ 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/term.c b/term.c
index 8f395a9e..97f6d2bc 100644
--- a/term.c
+++ b/term.c
@@ -468,9 +468,8 @@ 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:
@@ -705,9 +704,8 @@ 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: