===================================================================
RCS file: /cvs/mandoc/html.c,v
retrieving revision 1.141
retrieving revision 1.150
diff -u -p -r1.141 -r1.150
--- mandoc/html.c	2011/05/17 11:19:45	1.141
+++ mandoc/html.c	2011/10/05 21:35:17	1.150
@@ -1,4 +1,4 @@
-/*	$Id: html.c,v 1.141 2011/05/17 11:19:45 kristaps Exp $ */
+/*	$Id: html.c,v 1.150 2011/10/05 21:35:17 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010, 2011 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011 Ingo Schwarze <schwarze@openbsd.org>
@@ -108,8 +108,6 @@ static	const char	*const roffscales[SCALE_MAX] = {
 };
 
 static	void	 bufncat(struct html *, const char *, size_t);
-static	void	 print_spec(struct html *, const char *, size_t);
-static	void	 print_res(struct html *, const char *, size_t);
 static	void	 print_ctag(struct html *, enum htmltag);
 static	int	 print_encode(struct html *, const char *, int);
 static	void	 print_metaf(struct html *, enum mandoc_esc);
@@ -120,13 +118,14 @@ static void *
 ml_alloc(char *outopts, enum htmltype type)
 {
 	struct html	*h;
-	const char	*toks[4];
+	const char	*toks[5];
 	char		*v;
 
 	toks[0] = "style";
 	toks[1] = "man";
 	toks[2] = "includes";
-	toks[3] = NULL;
+	toks[3] = "fragment";
+	toks[4] = NULL;
 
 	h = mandoc_calloc(1, sizeof(struct html));
 
@@ -145,6 +144,9 @@ ml_alloc(char *outopts, enum htmltype type)
 		case (2):
 			h->base_includes = v;
 			break;
+		case (3):
+			h->oflags |= HTML_FRAGMENT;
+			break;
 		default:
 			break;
 		}
@@ -219,45 +221,6 @@ print_gen_head(struct html *h)
 }
 
 static void
-print_spec(struct html *h, const char *p, size_t len)
-{
-	int		 cp;
-	const char	*rhs;
-	size_t		 sz;
-
-	if ((cp = mchars_spec2cp(h->symtab, p, len)) > 0) {
-		printf("&#%d;", cp);
-		return;
-	} else if (-1 == cp && 1 == len) {
-		fwrite(p, 1, len, stdout);
-		return;
-	} else if (-1 == cp)
-		return;
-
-	if (NULL != (rhs = mchars_spec2str(h->symtab, p, len, &sz)))
-		fwrite(rhs, 1, sz, stdout);
-}
-
-
-static void
-print_res(struct html *h, const char *p, size_t len)
-{
-	int		 cp;
-	const char	*rhs;
-	size_t		 sz;
-
-	if ((cp = mchars_res2cp(h->symtab, p, len)) > 0) {
-		printf("&#%d;", cp);
-		return;
-	} else if (-1 == cp)
-		return;
-
-	if (NULL != (rhs = mchars_res2str(h->symtab, p, len, &sz)))
-		fwrite(rhs, 1, sz, stdout);
-}
-
-
-static void
 print_metaf(struct html *h, enum mandoc_esc deco)
 {
 	enum htmlfont	 font;
@@ -272,6 +235,8 @@ print_metaf(struct html *h, enum mandoc_esc deco)
 	case (ESCAPE_FONTBOLD):
 		font = HTMLFONT_BOLD;
 		break;
+	case (ESCAPE_FONT):
+		/* FALLTHROUGH */
 	case (ESCAPE_FONTROMAN):
 		font = HTMLFONT_NONE;
 		break;
@@ -315,10 +280,10 @@ html_strlen(const char *cp)
 		switch (mandoc_escape(&cp, &seq, &ssz)) {
 		case (ESCAPE_ERROR):
 			return(sz);
+		case (ESCAPE_UNICODE):
+			/* FALLTHROUGH */
 		case (ESCAPE_NUMBERED):
 			/* FALLTHROUGH */
-		case (ESCAPE_PREDEF):
-			/* FALLTHROUGH */
 		case (ESCAPE_SPECIAL):
 			sz++;
 			break;
@@ -373,17 +338,26 @@ print_encode(struct html *h, const char *p, int norecu
 			break;
 
 		switch (esc) {
+		case (ESCAPE_UNICODE):
+			/* Skip passed "u" header. */
+			c = mchars_num2uc(seq + 1, len - 1);
+			if ('\0' != c)
+				printf("&#x%x;", c);
+			break;
 		case (ESCAPE_NUMBERED):
 			c = mchars_num2char(seq, len);
 			if ('\0' != c)
 				putchar(c);
 			break;
-		case (ESCAPE_PREDEF):
-			print_res(h, seq, len);
-			break;
 		case (ESCAPE_SPECIAL):
-			print_spec(h, seq, len);
+			c = mchars_spec2cp(h->symtab, seq, len);
+			if (c > 0)
+				printf("&#%d;", c);
+			else if (-1 == c && 1 == len)
+				putchar((int)*seq);
 			break;
+		case (ESCAPE_FONT):
+			/* FALLTHROUGH */
 		case (ESCAPE_FONTPREV):
 			/* FALLTHROUGH */
 		case (ESCAPE_FONTBOLD):
@@ -543,9 +517,11 @@ print_text(struct html *h, const char *word)
 			print_otag(h, TAG_I, 0, NULL);
 
 	assert(word);
-	if ( ! print_encode(h, word, 0))
+	if ( ! print_encode(h, word, 0)) {
 		if ( ! (h->flags & HTML_NONOSPACE))
 			h->flags &= ~HTML_NOSPACE;
+	} else
+		h->flags |= HTML_NOSPACE;
 
 	if (h->metaf) {
 		print_tagq(h, h->metaf);
@@ -625,7 +601,6 @@ bufcat(struct html *h, const char *p)
 
 	h->buflen = strlcat(h->buf, p, BUFSIZ);
 	assert(h->buflen < BUFSIZ);
-	h->buflen--;
 }
 
 void
@@ -656,6 +631,7 @@ buffmt_includes(struct html *h, const char *name)
 
 	pp = h->base_includes;
 	
+	bufinit(h);
 	while (NULL != (p = strchr(pp, '%'))) {
 		bufncat(h, pp, (size_t)(p - pp));
 		switch (*(p + 1)) {
@@ -680,7 +656,7 @@ buffmt_man(struct html *h, 
 
 	pp = h->base_man;
 	
-	/* LINTED */
+	bufinit(h);
 	while (NULL != (p = strchr(pp, '%'))) {
 		bufncat(h, pp, (size_t)(p - pp));
 		switch (*(p + 1)) {
@@ -713,32 +689,11 @@ bufcat_su(struct html *h, const char *p, const struct 
 }
 
 void
-html_idcat(char *dst, const char *src, int sz)
+bufcat_id(struct html *h, const char *src)
 {
-	int		 ssz;
 
-	assert(sz > 2);
-
 	/* Cf. <http://www.w3.org/TR/html4/types.html#h-6.2>. */
 
-	/* We can't start with a number (bah). */
-
-	if ('#' == *dst) {
-		dst++;
-		sz--;
-	}
-	if ('\0' == *dst) {
-		*dst++ = 'x';
-		*dst = '\0';
-		sz--;
-	}
-
-	for ( ; *dst != '\0' && sz; dst++, sz--)
-		/* Jump to end. */ ;
-
-	for ( ; *src != '\0' && sz > 1; src++) {
-		ssz = snprintf(dst, (size_t)sz, "%.2x", *src);
-		sz -= ssz;
-		dst += ssz;
-	}
+	while ('\0' != *src)
+		bufcat_fmt(h, "%.2x", *src++);
 }