===================================================================
RCS file: /cvs/mandoc/html.c,v
retrieving revision 1.227
retrieving revision 1.236
diff -u -p -r1.227 -r1.236
--- mandoc/html.c	2018/05/21 00:30:23	1.227
+++ mandoc/html.c	2018/06/25 14:00:28	1.236
@@ -1,4 +1,4 @@
-/*	$Id: html.c,v 1.227 2018/05/21 00:30:23 schwarze Exp $ */
+/*	$Id: html.c,v 1.236 2018/06/25 14:00:28 schwarze Exp $ */
 /*
  * Copyright (c) 2008-2011, 2014 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011-2015, 2017, 2018 Ingo Schwarze <schwarze@openbsd.org>
@@ -22,6 +22,7 @@
 #include <assert.h>
 #include <ctype.h>
 #include <stdarg.h>
+#include <stddef.h>
 #include <stdio.h>
 #include <stdint.h>
 #include <stdlib.h>
@@ -29,6 +30,7 @@
 #include <unistd.h>
 
 #include "mandoc_aux.h"
+#include "mandoc_ohash.h"
 #include "mandoc.h"
 #include "roff.h"
 #include "out.h"
@@ -67,8 +69,6 @@ static	const struct htmldata htmltags[TAG_MAX] = {
 	{"br",		HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL},
 	{"a",		0},
 	{"table",	HTML_NLALL | HTML_INDENT},
-	{"colgroup",	HTML_NLALL | HTML_INDENT},
-	{"col",		HTML_NOSTACK | HTML_AUTOCLOSE | HTML_NLALL},
 	{"tr",		HTML_NLALL | HTML_INDENT},
 	{"td",		HTML_NLAROUND},
 	{"li",		HTML_NLAROUND | HTML_INDENT},
@@ -117,7 +117,9 @@ static	const char	*const roffscales[SCALE_MAX] = {
 	"ex", /* SCALE_FS */
 };
 
-static	void	 a2width(const char *, struct roffsu *);
+/* Avoid duplicate HTML id= attributes. */
+static	struct ohash	 id_unique;
+
 static	void	 print_byte(struct html *, char);
 static	void	 print_endword(struct html *);
 static	void	 print_indent(struct html *);
@@ -144,6 +146,8 @@ html_alloc(const struct manoutput *outopts)
 	if (outopts->fragment)
 		h->oflags |= HTML_FRAGMENT;
 
+	mandoc_ohash_init(&id_unique, 4, 0);
+
 	return h;
 }
 
@@ -152,15 +156,22 @@ html_free(void *p)
 {
 	struct tag	*tag;
 	struct html	*h;
+	char		*cp;
+	unsigned int	 slot;
 
 	h = (struct html *)p;
-
 	while ((tag = h->tag) != NULL) {
 		h->tag = tag->next;
 		free(tag);
 	}
-
 	free(h);
+
+	cp = ohash_first(&id_unique, &slot);
+	while (cp != NULL) {
+		free(cp);
+		cp = ohash_next(&id_unique, &slot);
+	}
+	ohash_delete(&id_unique);
 }
 
 void
@@ -192,6 +203,8 @@ print_gen_head(struct html *h)
 	print_endline(h);
 	print_text(h, "span.Pa, span.Ad { font-style: italic; }");
 	print_endline(h);
+	print_text(h, "span.Ms { font-weight: bold; }");
+	print_endline(h);
 	print_text(h, "dl.Bl-diag ");
 	print_byte(h, '>');
 	print_text(h, " dt { font-weight: bold; }");
@@ -255,10 +268,12 @@ print_metaf(struct html *h, enum mandoc_esc deco)
 }
 
 char *
-html_make_id(const struct roff_node *n)
+html_make_id(const struct roff_node *n, int unique)
 {
 	const struct roff_node	*nch;
-	char			*buf, *cp;
+	char			*buf, *bufs, *cp;
+	unsigned int		 slot;
+	int			 suffix;
 
 	for (nch = n->child; nch != NULL; nch = nch->next)
 		if (nch->type != ROFFT_TEXT)
@@ -269,12 +284,42 @@ html_make_id(const struct roff_node *n)
 	if (buf == NULL)
 		return NULL;
 
-	/* http://www.w3.org/TR/html5/dom.html#the-id-attribute */
+	/*
+	 * In ID attributes, only use ASCII characters that are
+	 * permitted in URL-fragment strings according to the
+	 * explicit list at:
+	 * https://url.spec.whatwg.org/#url-fragment-string
+	 */
 
 	for (cp = buf; *cp != '\0'; cp++)
-		if (*cp == ' ')
+		if (isalnum((unsigned char)*cp) == 0 &&
+		    strchr("!$&'()*+,-./:;=?@_~", *cp) == NULL)
 			*cp = '_';
 
+	if (unique == 0)
+		return buf;
+
+	/* Avoid duplicate HTML id= attributes. */
+
+	bufs = NULL;
+	suffix = 1;
+	slot = ohash_qlookup(&id_unique, buf);
+	cp = ohash_find(&id_unique, slot);
+	if (cp != NULL) {
+		while (cp != NULL) {
+			free(bufs);
+			if (++suffix > 127) {
+				free(buf);
+				return NULL;
+			}
+			mandoc_asprintf(&bufs, "%s_%d", buf, suffix);
+			slot = ohash_qlookup(&id_unique, bufs);
+			cp = ohash_find(&id_unique, slot);
+		}
+		free(buf);
+		buf = bufs;
+	}
+	ohash_insert(&id_unique, slot, buf);
 	return buf;
 }
 
@@ -641,32 +686,6 @@ print_otag(struct html *h, enum htmltag tag, const cha
 		case 'u':
 			su = va_arg(ap, struct roffsu *);
 			break;
-		case 'w':
-			if ((arg2 = va_arg(ap, char *)) != NULL) {
-				su = &mysu;
-				a2width(arg2, su);
-			}
-			if (*fmt == '*') {
-				if (su != NULL && su->unit == SCALE_EN &&
-				    su->scale > 5.9 && su->scale < 6.1)
-					su = NULL;
-				fmt++;
-			}
-			if (*fmt == '+') {
-				if (su != NULL) {
-					/* Make even bold text fit. */
-					su->scale *= 1.2;
-					/* Add padding. */
-					su->scale += 3.0;
-				}
-				fmt++;
-			}
-			if (*fmt == '-') {
-				if (su != NULL)
-					su->scale *= -1.0;
-				fmt++;
-			}
-			break;
 		default:
 			abort();
 		}
@@ -677,18 +696,9 @@ print_otag(struct html *h, enum htmltag tag, const cha
 		case 'h':
 			attr = "height";
 			break;
-		case 'i':
-			attr = "text-indent";
-			break;
 		case 'l':
 			attr = "margin-left";
 			break;
-		case 'w':
-			attr = "width";
-			break;
-		case 'W':
-			attr = "min-width";
-			break;
 		case '?':
 			attr = arg1;
 			arg1 = va_arg(ap, char *);
@@ -1002,22 +1012,4 @@ print_word(struct html *h, const char *cp)
 {
 	while (*cp != '\0')
 		print_byte(h, *cp++);
-}
-
-/*
- * Calculate the scaling unit passed in a `-width' argument.  This uses
- * either a native scaling unit (e.g., 1i, 2m) or the string length of
- * the value.
- */
-static void
-a2width(const char *p, struct roffsu *su)
-{
-	const char	*end;
-
-	end = a2roffsu(p, su, SCALE_MAX);
-	if (end == NULL || *end != '\0') {
-		su->unit = SCALE_EN;
-		su->scale = html_strlen(p);
-	} else if (su->scale < 0.0)
-		su->scale = 0.0;
 }