===================================================================
RCS file: /cvs/mandoc/html.c,v
retrieving revision 1.132
retrieving revision 1.138
diff -u -p -r1.132 -r1.138
--- mandoc/html.c	2011/04/09 15:29:40	1.132
+++ mandoc/html.c	2011/05/14 16:28:23	1.138
@@ -1,4 +1,4 @@
-/*	$Id: html.c,v 1.132 2011/04/09 15:29:40 kristaps Exp $ */
+/*	$Id: html.c,v 1.138 2011/05/14 16:28:23 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
@@ -31,6 +31,7 @@
 #include <unistd.h>
 
 #include "mandoc.h"
+#include "libmandoc.h"
 #include "out.h"
 #include "html.h"
 #include "main.h"
@@ -122,7 +123,7 @@ ml_alloc(char *outopts, enum htmltype type)
 
 	h->type = type;
 	h->tags.head = NULL;
-	h->symtab = chars_init(CHARS_HTML);
+	h->symtab = mchars_alloc();
 
 	while (outopts && *outopts)
 		switch (getsubopt(&outopts, UNCONST(toks), &v)) {
@@ -172,7 +173,7 @@ html_free(void *p)
 	}
 	
 	if (h->symtab)
-		chars_free(h->symtab);
+		mchars_free(h->symtab);
 
 	free(h);
 }
@@ -212,11 +213,10 @@ print_gen_head(struct html *h)
 static void
 print_num(struct html *h, const char *p, size_t len)
 {
-	const char	*rhs;
+	char		 c;
 
-	rhs = chars_num2char(p, len);
-	if (rhs)
-		putchar((int)*rhs);
+	if ('\0' != (c = mchars_num2char(p, len)))
+		putchar((int)c);
 }
 
 static void
@@ -226,7 +226,7 @@ print_spec(struct html *h, const char *p, size_t len)
 	const char	*rhs;
 	size_t		 sz;
 
-	if ((cp = chars_spec2cp(h->symtab, p, len)) > 0) {
+	if ((cp = mchars_spec2cp(h->symtab, p, len)) > 0) {
 		printf("&#%d;", cp);
 		return;
 	} else if (-1 == cp && 1 == len) {
@@ -235,7 +235,7 @@ print_spec(struct html *h, const char *p, size_t len)
 	} else if (-1 == cp)
 		return;
 
-	if (NULL != (rhs = chars_spec2str(h->symtab, p, len, &sz)))
+	if (NULL != (rhs = mchars_spec2str(h->symtab, p, len, &sz)))
 		fwrite(rhs, 1, sz, stdout);
 }
 
@@ -247,13 +247,13 @@ print_res(struct html *h, const char *p, size_t len)
 	const char	*rhs;
 	size_t		 sz;
 
-	if ((cp = chars_res2cp(h->symtab, p, len)) > 0) {
+	if ((cp = mchars_res2cp(h->symtab, p, len)) > 0) {
 		printf("&#%d;", cp);
 		return;
 	} else if (-1 == cp)
 		return;
 
-	if (NULL != (rhs = chars_res2str(h->symtab, p, len, &sz)))
+	if (NULL != (rhs = mchars_res2str(h->symtab, p, len, &sz)))
 		fwrite(rhs, 1, sz, stdout);
 }
 
@@ -295,7 +295,42 @@ print_metaf(struct html *h, enum mandoc_esc deco)
 			print_otag(h, TAG_I, 0, NULL);
 }
 
+int
+html_strlen(const char *cp)
+{
+	int		 ssz, sz;
+	const char	*seq, *p;
 
+	/*
+	 * Account for escaped sequences within string length
+	 * calculations.  This follows the logic in term_strlen() as we
+	 * must calculate the width of produced strings.
+	 * Assume that characters are always width of "1".  This is
+	 * hacky, but it gets the job done for approximation of widths.
+	 */
+
+	sz = 0;
+	while (NULL != (p = strchr(cp, '\\'))) {
+		sz += (int)(p - cp);
+		++cp;
+		switch (mandoc_escape(&cp, &seq, &ssz)) {
+		case (ESCAPE_ERROR):
+			return(sz);
+		case (ESCAPE_PREDEF):
+			sz++;
+			break;
+		case (ESCAPE_SPECIAL):
+			sz++;
+			break;
+		default:
+			break;
+		}
+	}
+
+	assert(sz >= 0);
+	return(sz + strlen(cp));
+}
+
 static int
 print_encode(struct html *h, const char *p, int norecurse)
 {
@@ -427,7 +462,7 @@ print_otag(struct html *h, enum htmltag tag, 
 		print_attr(h, "lang", "en");
 	}
 
-	/* Accomodate for XML "well-formed" singleton escaping. */
+	/* Accommodate for XML "well-formed" singleton escaping. */
 
 	if (HTML_AUTOCLOSE & htmltags[tag].flags)
 		switch (h->type) {