ascii_xxx -> chars_xxx (intended to hold more than just ascii encoding).

More html work.

9 files changed, 136 insertions, 93 deletions

diff --git a/Makefile b/Makefile
index 24cad136..4b2da44c 100644
--- a/Makefile
+++ b/Makefile
@@ -35,11 +35,11 @@ MANOBJS	   = man_macro.o man.o man_hash.o man_validate.o \
 MANSRCS	   = man_macro.c man.c man_hash.c man_validate.c \
 	     man_action.c mandoc.c man_argv.c
 
-MAINLNS	   = main.ln mdoc_term.ln ascii.ln term.ln tree.ln \
+MAINLNS	   = main.ln mdoc_term.ln chars.ln term.ln tree.ln \
 	     compat.ln man_term.ln html.ln
-MAINOBJS   = main.o mdoc_term.o ascii.o term.o tree.o compat.o \
+MAINOBJS   = main.o mdoc_term.o chars.o term.o tree.o compat.o \
 	     man_term.o html.o
-MAINSRCS   = main.c mdoc_term.c ascii.c term.c tree.c compat.c \
+MAINSRCS   = main.c mdoc_term.c chars.c term.c tree.c compat.c \
 	     man_term.c html.c
 
 LLNS	   = llib-llibmdoc.ln llib-llibman.ln llib-lmandoc.ln
@@ -47,7 +47,7 @@ LNS	   = $(MAINLNS) $(MDOCLNS) $(MANLNS)
 LIBS	   = libmdoc.a libman.a
 OBJS	   = $(MDOCOBJS) $(MAINOBJS) $(MANOBJS)
 SRCS	   = $(MDOCSRCS) $(MAINSRCS) $(MANSRCS)
-DATAS	   = arch.in att.in lib.in msec.in st.in vol.in ascii.in
+DATAS	   = arch.in att.in lib.in msec.in st.in vol.in chars.in
 HEADS	   = mdoc.h libmdoc.h man.h libman.h term.h libmandoc.h
 SGMLS	   = index.sgml 
 XSLS	   = ChangeLog.xsl
@@ -117,8 +117,8 @@ arch.o: arch.c arch.in libmdoc.h
 vol.ln: vol.c vol.in libmdoc.h
 vol.o: vol.c vol.in libmdoc.h
 
-ascii.ln: ascii.c ascii.in term.h
-ascii.o: ascii.c ascii.in term.h
+chars.ln: chars.c chars.in chars.h
+chars.o: chars.c chars.in chars.h
 
 msec.ln: msec.c msec.in libmdoc.h
 msec.o: msec.c msec.in libmdoc.h
diff --git a/ascii.c b/chars.c
index 7bd50693..eaf80689 100644
--- a/ascii.c
+++ b/chars.c
@@ -19,69 +19,66 @@
 #include <stdlib.h>
 #include <string.h>
 
-#include "term.h"
+#include "chars.h"
 
 #define	ASCII_PRINT_HI	 126
 #define	ASCII_PRINT_LO	 32
 
-struct	line {
+struct	ln {
+	struct ln	 *next;
 	const char	 *code;
 	const char	 *out;
 	size_t		  codesz;
 	size_t		  outsz;
 	int		  type;
-#define	ASCII_CHAR	 (1 << 0)
-#define	ASCII_STRING	 (1 << 1)
-#define ASCII_BOTH	 (0x03)
+#define	CHARS_CHAR	 (1 << 0)
+#define	CHARS_STRING	 (1 << 1)
+#define CHARS_BOTH	 (0x03)
 };
 
-struct	linep {
-	const struct line *line;
-	struct linep	 *next;
-};
+#define	LINES_MAX	  266
 
 #define CHAR(w, x, y, z) \
-	{ (w), (y), (x), (z), ASCII_CHAR },
+	{ NULL, (w), (y), (x), (z), CHARS_CHAR },
 #define STRING(w, x, y, z) \
-	{ (w), (y), (x), (z), ASCII_STRING },
+	{ NULL, (w), (y), (x), (z), CHARS_STRING },
 #define BOTH(w, x, y, z) \
-	{ (w), (y), (x), (z), ASCII_BOTH },
-static	const struct line lines[] = {
-#include "ascii.in"
-};
+	{ NULL, (w), (y), (x), (z), CHARS_BOTH },
 
-struct	asciitab {
-	struct linep	 *lines;
-	void		**htab;
+static	struct ln lines[LINES_MAX] = {
+#include "chars.in"
 };
 
+struct	tbl {
+	struct ln	**htab;
+};
 
-static	inline int	  match(const struct line *,
+static	inline int	  match(const struct ln *,
 				const char *, size_t, int);
-static	const char *	  lookup(struct asciitab *, const char *, 
+static	const char	 *find(struct tbl *, const char *, 
 				size_t, size_t *, int);
 
 
 void
-term_asciifree(void *arg)
+chars_free(void *arg)
 {
-	struct asciitab	*tab;
+	struct tbl	*tab;
 
-	tab = (struct asciitab *)arg;
+	tab = (struct tbl *)arg;
 
-	free(tab->lines);
 	free(tab->htab);
 	free(tab);
 }
 
 
+/* ARGSUSED */
 void *
-term_ascii2htab(void)
+chars_init(enum chars type)
 {
-	struct asciitab  *tab;
-	void		**htab;
-	struct linep	 *pp, *p;
-	int		  i, len, hash;
+	struct tbl	 *tab;
+	struct ln	**htab;
+	struct ln	 *pp;
+	int		  i, hash;
 
 	/*
 	 * Constructs a very basic chaining hashtable.  The hash routine
@@ -90,71 +87,59 @@ term_ascii2htab(void)
 	 * (they're in-line re-ordered during lookup).
 	 */
 
-	if (NULL == (tab = malloc(sizeof(struct asciitab))))
-		err(1, "malloc");
-
-	len = sizeof(lines) / sizeof(struct line);
-
-	if (NULL == (p = calloc((size_t)len, sizeof(struct linep))))
+	if (NULL == (tab = malloc(sizeof(struct tbl))))
 		err(1, "malloc");
 
 	htab = calloc(ASCII_PRINT_HI - ASCII_PRINT_LO + 1, 
-			sizeof(struct linep **));
+			sizeof(struct ln **));
 
 	if (NULL == htab)
 		err(1, "malloc");
 
-	for (i = 0; i < len; i++) {
+	for (i = 0; i < LINES_MAX; i++) {
 		assert(lines[i].codesz > 0);
 		assert(lines[i].code);
 		assert(lines[i].out);
 
-		p[i].line = &lines[i];
-
 		hash = (int)lines[i].code[0] - ASCII_PRINT_LO;
 
-		if (NULL == (pp = ((struct linep **)htab)[hash])) {
-			htab[hash] = &p[i];
+		if (NULL == (pp = htab[hash])) {
+			htab[hash] = &lines[i];
 			continue;
 		}
 
 		for ( ; pp->next; pp = pp->next)
 			/* Scan ahead. */ ;
 
-		pp->next = &p[i];
+		pp->next = &lines[i];
 	}
 
 	tab->htab = htab;
-	tab->lines = p;
-
 	return(tab);
 }
 
 
 const char *
-term_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz)
+chars_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz)
 {
 
-	return(lookup((struct asciitab *)arg, p, 
-				sz, rsz, ASCII_CHAR));
+	return(find((struct tbl *)arg, p, sz, rsz, CHARS_CHAR));
 }
 
 
 const char *
-term_a2res(void *arg, const char *p, size_t sz, size_t *rsz)
+chars_a2res(void *arg, const char *p, size_t sz, size_t *rsz)
 {
 
-	return(lookup((struct asciitab *)arg, p, 
-				sz, rsz, ASCII_STRING));
+	return(find((struct tbl *)arg, p, sz, rsz, CHARS_STRING));
 }
 
 
 static const char *
-lookup(struct asciitab *tab, const char *p, 
-		size_t sz, size_t *rsz, int type)
+find(struct tbl *tab, const char *p, size_t sz, size_t *rsz, int type)
 {
-	struct linep	 *pp, *prev;
-	void		**htab;
+	struct ln	 *pp, *prev;
+	struct ln	**htab;
 	int		  hash;
 
 	assert(p);
@@ -163,7 +148,6 @@ lookup(struct asciitab *tab, const char *p,
 	if (p[0] < ASCII_PRINT_LO || p[0] > ASCII_PRINT_HI)
 		return(NULL);
 
-
 	/*
 	 * Lookup the symbol in the symbol hash.  See ascii2htab for the
 	 * hashtable specs.  This dynamically re-orders the hash chain
@@ -173,18 +157,18 @@ lookup(struct asciitab *tab, const char *p,
 	hash = (int)p[0] - ASCII_PRINT_LO;
 	htab = tab->htab;
 
-	if (NULL == (pp = ((struct linep **)htab)[hash]))
+	if (NULL == (pp = htab[hash]))
 		return(NULL);
 
 	if (NULL == pp->next) {
-		if ( ! match(pp->line, p, sz, type)) 
+		if ( ! match(pp, p, sz, type)) 
 			return(NULL);
-		*rsz = pp->line->outsz;
-		return(pp->line->out);
+		*rsz = pp->outsz;
+		return(pp->out);
 	}
 
 	for (prev = NULL; pp; pp = pp->next) {
-		if ( ! match(pp->line, p, sz, type)) {
+		if ( ! match(pp, p, sz, type)) {
 			prev = pp;
 			continue;
 		}
@@ -193,12 +177,12 @@ lookup(struct asciitab *tab, const char *p,
 
 		if (prev) {
 			prev->next = pp->next;
-			pp->next = ((struct linep **)htab)[hash];
+			pp->next = htab[hash];
 			htab[hash] = pp;
 		}
 
-		*rsz = pp->line->outsz;
-		return(pp->line->out);
+		*rsz = pp->outsz;
+		return(pp->out);
 	}
 
 	return(NULL);
@@ -206,12 +190,12 @@ lookup(struct asciitab *tab, const char *p,
 
 
 static inline int
-match(const struct line *line, const char *p, size_t sz, int type)
+match(const struct ln *ln, const char *p, size_t sz, int type)
 {
 
-	if ( ! (line->type & type))
+	if ( ! (ln->type & type))
 		return(0);
-	if (line->codesz != sz)
+	if (ln->codesz != sz)
 		return(0);
-	return(0 == strncmp(line->code, p, sz));
+	return(0 == strncmp(ln->code, p, sz));
 }
diff --git a/chars.h b/chars.h
new file mode 100644
index 00000000..a18c2880
--- /dev/null
+++ b/chars.h
@@ -0,0 +1,34 @@
+/*	$Id$ */
+/*
+ * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
+ *
+ * Permission to use, copy, modify, and distribute this software for any
+ * purpose with or without fee is hereby granted, provided that the above
+ * copyright notice and this permission notice appear in all copies.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
+ * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
+ * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
+ * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
+ * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
+ * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
+ * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
+ */
+#ifndef CHARS_H
+#define CHARS_H
+
+__BEGIN_DECLS
+
+enum	chars {
+	CHARS_ASCII,
+	CHARS_HTML
+};
+
+void		 *chars_init(enum chars);
+const char	 *chars_a2ascii(void *, const char *, size_t, size_t *);
+const char	 *chars_a2res(void *, const char *, size_t, size_t *);
+void		  chars_free(void *);
+
+__END_DECLS
+
+#endif /*!CHARS_H*/
diff --git a/ascii.in b/chars.in
index 60f2de00..57f2dee7 100644
--- a/ascii.in
+++ b/chars.in
@@ -26,7 +26,8 @@
  * right-hand side is what's produced by the front-end, with the fourth
  * element being its length.
  *
- * Be sure to C-escape strings!
+ * XXX - C-escape strings!
+ * XXX - update LINES_MAX if adding more!
  */
 
 STRING("Am",			2,	"&",		1)
diff --git a/html.c b/html.c
index 5fa25665..4be3d55d 100644
--- a/html.c
+++ b/html.c
@@ -138,9 +138,11 @@ static	void		  print_encode(const char *);
 static	void		  print_text(struct html *, const char *);
 static	int		  mdoc_root_pre(MDOC_ARGS);
 
+static	int		  mdoc_ar_pre(MDOC_ARGS);
 static	int		  mdoc_fl_pre(MDOC_ARGS);
 static	int		  mdoc_nd_pre(MDOC_ARGS);
 static	int		  mdoc_nm_pre(MDOC_ARGS);
+static	int		  mdoc_ns_pre(MDOC_ARGS);
 static	int		  mdoc_op_pre(MDOC_ARGS);
 static	void		  mdoc_op_post(MDOC_ARGS);
 static	int		  mdoc_pp_pre(MDOC_ARGS);
@@ -165,7 +167,7 @@ static	const struct htmlmdoc mdocs[MDOC_MAX] = {
 	{NULL, NULL}, /* It */
 	{NULL, NULL}, /* Ad */ 
 	{NULL, NULL}, /* An */
-	{NULL, NULL}, /* Ar */
+	{mdoc_ar_pre, NULL}, /* Ar */
 	{NULL, NULL}, /* Cd */
 	{NULL, NULL}, /* Cm */
 	{NULL, NULL}, /* Dv */ 
@@ -222,7 +224,7 @@ static	const struct htmlmdoc mdocs[MDOC_MAX] = {
 	{NULL, NULL}, /* Fx */
 	{NULL, NULL}, /* Ms */
 	{NULL, NULL}, /* No */
-	{NULL, NULL}, /* Ns */
+	{mdoc_ns_pre, NULL}, /* Ns */
 	{NULL, NULL}, /* Nx */
 	{NULL, NULL}, /* Ox */
 	{NULL, NULL}, /* Pc */
@@ -697,7 +699,7 @@ mdoc_nd_pre(MDOC_ARGS)
 {
 
 	if (MDOC_BODY == n->type)
-		print_text(h, "--");
+		print_text(h, "\\(en");
 	return(1);
 }
 
@@ -785,3 +787,27 @@ mdoc_xr_pre(MDOC_ARGS)
 
 	return(0);
 }
+
+
+/* ARGSUSED */
+static int
+mdoc_ns_pre(MDOC_ARGS)
+{
+
+	h->flags |= HTML_NOSPACE;
+	return(1);
+}
+
+
+/* ARGSUSED */
+static int
+mdoc_ar_pre(MDOC_ARGS)
+{
+	struct htmlpair tag;
+
+	tag.key = ATTR_CLASS;
+	tag.val = "arg";
+
+	print_otag(h, TAG_SPAN, 1, &tag);
+	return(1);
+}
diff --git a/man_hash.c b/man_hash.c
index 4d839d7d..6e9f2afb 100644
--- a/man_hash.c
+++ b/man_hash.c
@@ -21,10 +21,12 @@
 
 #include "libman.h"
 
-
 static	u_char		table[26 * 6];
 
-
+/*
+ * XXX - this hash has global scope, so if intended for use as a library
+ * with multiple callers, it will need re-invocation protection.
+ */
 void
 man_hash_init(void)
 {
@@ -49,7 +51,6 @@ man_hash_init(void)
 	}
 }
 
-
 int
 man_hash_find(const char *tmp)
 {
diff --git a/mdoc_hash.c b/mdoc_hash.c
index 4088e79a..be2d457a 100644
--- a/mdoc_hash.c
+++ b/mdoc_hash.c
@@ -27,7 +27,10 @@
 
 static	u_char		 table[27 * 12];
 
-
+/*
+ * XXX - this hash has global scope, so if intended for use as a library
+ * with multiple callers, it will need re-invocation protection.
+ */
 void
 mdoc_hash_init(void)
 {
@@ -54,7 +57,6 @@ mdoc_hash_init(void)
 	}
 }
 
-
 int
 mdoc_hash_find(const char *p)
 {
@@ -84,4 +86,3 @@ mdoc_hash_find(const char *p)
 
 	return(MDOC_MAX);
 }
-
diff --git a/term.c b/term.c
index f6fe8f47..6b3a56d3 100644
--- a/term.c
+++ b/term.c
@@ -20,6 +20,7 @@
 #include <stdlib.h>
 #include <string.h>
 
+#include "chars.h"
 #include "term.h"
 #include "man.h"
 #include "mdoc.h"
@@ -56,7 +57,7 @@ terminal_man(void *arg, const struct man *man)
 
 	p = (struct termp *)arg;
 	if (NULL == p->symtab)
-		p->symtab = term_ascii2htab();
+		p->symtab = chars_init(CHARS_ASCII);
 
 	man_run(p, man);
 }
@@ -69,7 +70,7 @@ terminal_mdoc(void *arg, const struct mdoc *mdoc)
 
 	p = (struct termp *)arg;
 	if (NULL == p->symtab)
-		p->symtab = term_ascii2htab();
+		p->symtab = chars_init(CHARS_ASCII);
 
 	mdoc_run(p, mdoc);
 }
@@ -90,7 +91,7 @@ term_free(struct termp *p)
 	if (p->buf)
 		free(p->buf);
 	if (TERMENC_ASCII == p->enc && p->symtab)
-		term_asciifree(p->symtab);
+		chars_free(p->symtab);
 
 	free(p);
 }
@@ -331,7 +332,7 @@ do_special(struct termp *p, const char *word, size_t len)
 	size_t		 sz;
 	int		 i;
 
-	rhs = term_a2ascii(p->symtab, word, len, &sz);
+	rhs = chars_a2ascii(p->symtab, word, len, &sz);
 
 	if (NULL == rhs) {
 #if 0
@@ -354,7 +355,7 @@ do_reserved(struct termp *p, const char *word, size_t len)
 	size_t		 sz;
 	int		 i;
 
-	rhs = term_a2res(p->symtab, word, len, &sz);
+	rhs = chars_a2res(p->symtab, word, len, &sz);
 
 	if (NULL == rhs) {
 #if 0
diff --git a/term.h b/term.h
index 06570633..c99035d8 100644
--- a/term.h
+++ b/term.h
@@ -48,11 +48,6 @@ struct	termp {
 	void		 *symtab;	/* Encoded-symbol table. */
 };
 
-void		 *term_ascii2htab(void);
-const char	 *term_a2ascii(void *, const char *, size_t, size_t *);
-const char	 *term_a2res(void *, const char *, size_t, size_t *);
-void		  term_asciifree(void *);
-
 void		  term_newln(struct termp *);
 void		  term_vspace(struct termp *);
 void		  term_word(struct termp *, const char *);