===================================================================
RCS file: /cvs/mandoc/mandoc.c,v
retrieving revision 1.45
retrieving revision 1.53
diff -u -p -r1.45 -r1.53
--- mandoc/mandoc.c	2011/04/09 15:29:40	1.45
+++ mandoc/mandoc.c	2011/05/24 21:31:23	1.53
@@ -1,4 +1,4 @@
-/*	$Id: mandoc.c,v 1.45 2011/04/09 15:29:40 kristaps Exp $ */
+/*	$Id: mandoc.c,v 1.53 2011/05/24 21:31:23 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
@@ -23,6 +23,8 @@
 
 #include <assert.h>
 #include <ctype.h>
+#include <errno.h>
+#include <limits.h>
 #include <stdlib.h>
 #include <stdio.h>
 #include <string.h>
@@ -95,15 +97,6 @@ numescape(const char *start)
 	return(++i);
 }
 
-/*
- * Handle an escaped sequeence.  This should be called with any
- * string subsequent a `\'.  Pass a pointer to this substring as "end";
- * it will be set to the supremum of the parsed escape sequence.  If
- * this returns ESCAPE_ERROR, the string is bogus and should be thrown
- * away.  If not ESCAPE_ERROR or ESCAPE_IGNORE, "start" is set to the
- * first relevant character of the substring (font, glyph, whatever) of
- * length sz.  Both "start" and "sz" may be NULL.
- */
 enum mandoc_esc
 mandoc_escape(const char **end, const char **start, int *sz)
 {
@@ -116,10 +109,9 @@ mandoc_escape(const char **end, const char **start, in
 	rstart = cp;
 	if (start)
 		*start = rstart;
-	i = 0;
+	i = lim = 0;
 	gly = ESCAPE_ERROR;
-	term = '\0';
-	numeric = 0;
+	term = numeric = '\0';
 
 	switch ((c = cp[i++])) {
 	/*
@@ -133,6 +125,14 @@ mandoc_escape(const char **end, const char **start, in
 		break;
 	case ('['):
 		gly = ESCAPE_SPECIAL;
+		/*
+		 * Unicode escapes are defined in groff as \[uXXXX] to
+		 * \[u10FFFF], where the contained value must be a valid
+		 * Unicode codepoint.  Here, however, only check whether
+		 * it's not a zero-width escape.
+		 */
+		if ('u' == cp[i] && ']' != cp[i + 1])
+			gly = ESCAPE_UNICODE;
 		term = ']';
 		break;
 	case ('C'):
@@ -164,10 +164,6 @@ mandoc_escape(const char **end, const char **start, in
 		if (ESCAPE_ERROR == gly)
 			gly = ESCAPE_IGNORE;
 		/* FALLTHROUGH */
-	case ('*'):
-		if (ESCAPE_ERROR == gly)
-			gly = ESCAPE_PREDEF;
-		/* FALLTHROUGH */
 	case ('f'):
 		if (ESCAPE_ERROR == gly)
 			gly = ESCAPE_FONT;
@@ -379,6 +375,7 @@ out:
 			gly = ESCAPE_FONTROMAN;
 			break;
 		}
+		break;
 	case (ESCAPE_SPECIAL):
 		if (1 != rlim)
 			break;
@@ -467,11 +464,11 @@ mandoc_getarg(struct mparse *parse, char **cpp, int ln
 
 	/* Quoting can only start with a new word. */
 	start = *cpp;
+	quoted = 0;
 	if ('"' == *start) {
 		quoted = 1;
 		start++;
-	} else
-		quoted = 0;
+	} 
 
 	pairs = 0;
 	white = 0;
@@ -612,7 +609,7 @@ mandoc_eos(const char *p, size_t sz, int enclosed)
 	/*
 	 * End-of-sentence recognition must include situations where
 	 * some symbols, such as `)', allow prior EOS punctuation to
-	 * propogate outward.
+	 * propagate outward.
 	 */
 
 	found = 0;
@@ -695,3 +692,35 @@ mandoc_getcontrol(const char *cp, int *ppos)
 	*ppos = pos;
 	return(1);
 }
+
+/*
+ * Convert a string to a long that may not be <0.
+ * If the string is invalid, or is less than 0, return -1.
+ */
+int
+mandoc_strntou(const char *p, size_t sz, int base)
+{
+	char		 buf[32];
+	char		*ep;
+	long		 v;
+
+	if (sz > 31)
+		return(-1);
+
+	memcpy(buf, p, sz);
+	buf[(int)sz] = '\0';
+
+	errno = 0;
+	v = strtol(buf, &ep, base);
+
+	if (buf[0] == '\0' || *ep != '\0')
+		return(-1);
+
+	if ((errno == ERANGE && 
+			(v == LONG_MAX || v == LONG_MIN)) ||
+			(v > INT_MAX || v < 0))
+		return(-1);
+
+	return((int)v);
+}
+