sync to 1.9.5: partial rewrite of special character and predefined string

tables and the supporting infrastructure, mostly in preparation for
HTML output support

8 files changed, 552 insertions, 536 deletions

diff --git a/usr.bin/mandoc/Makefile b/usr.bin/mandoc/Makefile
index 8abe16e931c..711a40fd398 100644
--- a/usr.bin/mandoc/Makefile
+++ b/usr.bin/mandoc/Makefile
@@ -1,4 +1,4 @@
-#	$OpenBSD: Makefile,v 1.18 2009/09/18 22:48:29 schwarze Exp $
+#	$OpenBSD: Makefile,v 1.19 2009/10/19 09:16:58 schwarze Exp $
 
 .include <bsd.own.mk>
 
@@ -14,7 +14,7 @@ SRCS=	mandoc.c mdoc_macro.c mdoc.c mdoc_hash.c mdoc_strings.c \
 	arch.c vol.c msec.c st.c
 SRCS+=	man_macro.c man.c man_hash.c man_validate.c \
 	man_action.c man_argv.c
-SRCS+=	main.c mdoc_term.c ascii.c term.c tree.c man_term.c
+SRCS+=	main.c mdoc_term.c chars.c term.c tree.c man_term.c
 
 PROG=	  mandoc
 
diff --git a/usr.bin/mandoc/ascii.c b/usr.bin/mandoc/ascii.c
deleted file mode 100644
index 89816232a9d..00000000000
--- a/usr.bin/mandoc/ascii.c
+++ /dev/null
@@ -1,217 +0,0 @@
-/*	$Id: ascii.c,v 1.3 2009/08/22 17:04:48 schwarze Exp $ */
-/*
- * Copyright (c) 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.
- */
-#include <assert.h>
-#include <err.h>
-#include <stdlib.h>
-#include <string.h>
-
-#include "term.h"
-
-#define	ASCII_PRINT_HI	 126
-#define	ASCII_PRINT_LO	 32
-
-struct	line {
-	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)
-};
-
-struct	linep {
-	const struct line *line;
-	struct linep	 *next;
-};
-
-#define CHAR(w, x, y, z) \
-	{ (w), (y), (x), (z), ASCII_CHAR },
-#define STRING(w, x, y, z) \
-	{ (w), (y), (x), (z), ASCII_STRING },
-#define BOTH(w, x, y, z) \
-	{ (w), (y), (x), (z), ASCII_BOTH },
-static	const struct line lines[] = {
-#include "ascii.in"
-};
-
-struct	asciitab {
-	struct linep	 *lines;
-	void		**htab;
-};
-
-
-static	inline int	  match(const struct line *,
-				const char *, size_t, int);
-static	const char *	  lookup(struct asciitab *, const char *, 
-				size_t, size_t *, int);
-
-
-void
-term_asciifree(void *arg)
-{
-	struct asciitab	*tab;
-
-	tab = (struct asciitab *)arg;
-
-	free(tab->lines);
-	free(tab->htab);
-	free(tab);
-}
-
-
-void *
-term_ascii2htab(void)
-{
-	struct asciitab  *tab;
-	void		**htab;
-	struct linep	 *pp, *p;
-	int		  i, len, hash;
-
-	/*
-	 * Constructs a very basic chaining hashtable.  The hash routine
-	 * is simply the integral value of the first character.
-	 * Subsequent entries are chained in the order they're processed
-	 * (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))))
-		err(1, "malloc");
-
-	htab = calloc(ASCII_PRINT_HI - ASCII_PRINT_LO + 1, 
-			sizeof(struct linep **));
-
-	if (NULL == htab)
-		err(1, "malloc");
-
-	for (i = 0; i < len; 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];
-			continue;
-		}
-
-		for ( ; pp->next; pp = pp->next)
-			/* Scan ahead. */ ;
-
-		pp->next = &p[i];
-	}
-
-	tab->htab = htab;
-	tab->lines = p;
-
-	return(tab);
-}
-
-
-const char *
-term_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz)
-{
-
-	return(lookup((struct asciitab *)arg, p, 
-				sz, rsz, ASCII_CHAR));
-}
-
-
-const char *
-term_a2res(void *arg, const char *p, size_t sz, size_t *rsz)
-{
-
-	return(lookup((struct asciitab *)arg, p, 
-				sz, rsz, ASCII_STRING));
-}
-
-
-static const char *
-lookup(struct asciitab *tab, const char *p, 
-		size_t sz, size_t *rsz, int type)
-{
-	struct linep	 *pp, *prev;
-	void		**htab;
-	int		  hash;
-
-	assert(p);
-	assert(sz > 0);
-
-	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
-	 * to optimise for repeat hits.
-	 */
-
-	hash = (int)p[0] - ASCII_PRINT_LO;
-	htab = tab->htab;
-
-	if (NULL == (pp = ((struct linep **)htab)[hash]))
-		return(NULL);
-
-	if (NULL == pp->next) {
-		if ( ! match(pp->line, p, sz, type)) 
-			return(NULL);
-		*rsz = pp->line->outsz;
-		return(pp->line->out);
-	}
-
-	for (prev = NULL; pp; pp = pp->next) {
-		if ( ! match(pp->line, p, sz, type)) {
-			prev = pp;
-			continue;
-		}
-
-		/* Re-order the hash chain. */
-
-		if (prev) {
-			prev->next = pp->next;
-			pp->next = ((struct linep **)htab)[hash];
-			htab[hash] = pp;
-		}
-
-		*rsz = pp->line->outsz;
-		return(pp->line->out);
-	}
-
-	return(NULL);
-}
-
-
-static inline int
-match(const struct line *line, const char *p, size_t sz, int type)
-{
-
-	if ( ! (line->type & type))
-		return(0);
-	if (line->codesz != sz)
-		return(0);
-	return(0 == strncmp(line->code, p, sz));
-}
diff --git a/usr.bin/mandoc/ascii.in b/usr.bin/mandoc/ascii.in
deleted file mode 100644
index 56f72d83bf3..00000000000
--- a/usr.bin/mandoc/ascii.in
+++ /dev/null
@@ -1,299 +0,0 @@
-/*	$Id: ascii.in,v 1.5 2009/08/22 17:04:48 schwarze Exp $ */
-/*
- * Copyright (c) 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.
- */
-
-/*
- * The ASCII translation tables.  STRING corresponds to predefined
- * strings (cf. mdoc_samples.7 and tmac/mdoc/doc-nroff).  CHAR
- * corresponds to special characters (cf. groff_char.7).  BOTH contains
- * sequences that are equivalent in both STRING and CHAR.
- *
- * Either way, the left-hand side corresponds to the input sequence (\x,
- * \(xx, \*(xx and so on) whose length is listed second element.  The
- * right-hand side is what's produced by the front-end, with the fourth
- * element being its length.
- *
- * Be sure to C-escape strings!
- */
-
-STRING("Am",			2,	"&",		1)
-STRING("Ba",			2,	"|",		1)
-STRING("Ge",			2,	">=",		2)
-STRING("Gt",			2,	">",		1)
-STRING("If",			2,	"infinity",	8)
-STRING("Le",			2,	"<=",		2)
-STRING("Lq",			2,	"``",		2)
-STRING("Lt",			2,	"<",		1)
-STRING("Na",			2,	"NaN",		3)
-STRING("Ne",			2,	"!=",		2)
-STRING("Pi",			2,	"pi",		2)
-STRING("Pm",			2,	"+-",		2)
-STRING("R",			1,	"(R)",		3)
-STRING("Rq",			2,	"\'\'",		2)
-STRING("Tm",			2,	"tm",		2)
-STRING("left-bracket",		12,	"[",		1)
-STRING("left-parenthesis",	16,	"(",		1)
-STRING("left-singlequote",	16,	"`",		1)
-STRING("lp",			2,	"(",		1)
-STRING("q",			1,	"\"",		1)
-STRING("quote-left",		10,	"`",		1)
-STRING("quote-right",		11,	"\'",		1)
-STRING("right-bracket",		13,	"]",		1)
-STRING("right-parenthesis",	17,	")",		1)
-STRING("right-singlequote",	17,	"\'",		1)
-STRING("rp",			2,	")",		1)
-
-BOTH("<=",			2,	"<=",		2)
-BOTH(">=",			2,	">=",		2)
-BOTH("aa",			2,	"\'",		1)
-BOTH("ga",			2,	"`",		1)
-BOTH("lq",			2, 	"``",		2)
-BOTH("rq",			2, 	"\'\'",		2)
-BOTH("ua",			2,	"^",		1)
-
-CHAR(" ",			1,	" ",		1)
-CHAR("!=",			2, 	"!=",		2)
-CHAR("%",			1,	"",		0)
-CHAR("&",			1,	"",		0)
-CHAR("(=",			2,	"(=",		2)
-CHAR("**",			2, 	"*",		1)
-CHAR("*A",			2,	"A",		1)
-CHAR("*B",			2,	"B",		1)
-CHAR("*C",			2,	"H",		1)
-CHAR("*D",			2,	"/\\",		2)
-CHAR("*E",			2,	"E",		1)
-CHAR("*F",			2,	"O_",		1)
-CHAR("*G",			2,	"|",		1)
-CHAR("*H",			2,	"O",		1)
-CHAR("*I",			2,	"I",		1)
-CHAR("*K",			2,	"K",		1)
-CHAR("*L",			2,	"/\\",		2)
-CHAR("*M",			2,	"M",		1)
-CHAR("*N",			2,	"N",		1)
-CHAR("*O",			2,	"O",		1)
-CHAR("*P",			2,	"TT",		2)
-CHAR("*Q",			2,	"Y",		1)
-CHAR("*R",			2,	"P",		1)
-CHAR("*S",			2,	">",		1)
-CHAR("*T",			2,	"T",		1)
-CHAR("*U",			2,	"Y",		1)
-CHAR("*W",			2,	"O",		1)
-CHAR("*X",			2,	"X",		1)
-CHAR("*Y",			2,	"H",		1)
-CHAR("*Z",			2,	"Z",		1)
-CHAR("*a",			2,	"a",		1)
-CHAR("*b",			2,	"B",		1)
-CHAR("*c",			2,	"E",		1)
-CHAR("*d",			2,	"d",		1)
-CHAR("*e",			2,	"e",		1)
-CHAR("*f",			2,	"o",		1)
-CHAR("*g",			2,	"y",		1)
-CHAR("*h",			2,	"0",		1)
-CHAR("*i",			2,	"i",		1)
-CHAR("*k",			2,	"k",		1)
-CHAR("*l",			2,	"\\",		1)
-CHAR("*m",			2,	"u",		1)
-CHAR("*n",			2,	"v",		1)
-CHAR("*o",			2,	"o",		1)
-CHAR("*p",			2,	"n",		1)
-CHAR("*q",			2,	"u",		1)
-CHAR("*r",			2,	"p",		1)
-CHAR("*s",			2,	"o",		1)
-CHAR("*t",			2,	"t",		1)
-CHAR("*u",			2,	"u",		1)
-CHAR("*w",			2,	"w",		1)
-CHAR("*x",			2,	"x",		1)
-CHAR("*y",			2,	"n",		1)
-CHAR("*z",			2,	"C",		1)
-CHAR("+-",			2, 	"+-",		2)
-CHAR("+f",			2,	"o",		1)
-CHAR("+h",			2,	"0",		1)
-CHAR("+p",			2,	"w",		1)
-CHAR(",C",			2,	"C",		1)
-CHAR(",c",			2,	"c",		1)
-CHAR("-",			1,	"-",		1)
-CHAR("->",			2, 	"->",		2)
-CHAR("-D",			2,	"D",		1)
-CHAR(".",			1,	".",		1)
-CHAR("/L",			2,	"L",		1)
-CHAR("/O",			2,	"O",		1)
-CHAR("/l",			2,	"l",		1)
-CHAR("/o",			2,	"o",		1)
-CHAR("0",			1,	" ",		1)
-CHAR(":A",			2,	"A",		1)
-CHAR(":E",			2,	"E",		1)
-CHAR(":I",			2,	"I",		1)
-CHAR(":O",			2,	"O",		1)
-CHAR(":U",			2,	"U",		1)
-CHAR(":a",			2,	"a",		1)
-CHAR(":e",			2,	"e",		1)
-CHAR(":i",			2,	"i",		1)
-CHAR(":o",			2,	"o",		1)
-CHAR(":u",			2,	"u",		1)
-CHAR(":y",			2,	"y",		1)
-CHAR("<-",			2, 	"<-",		2)
-CHAR("<=",			2, 	"<=",		2)
-CHAR("<>",			2, 	"<>",		2)
-CHAR("=)",			2,	"=)",		2)
-CHAR("==",			2, 	"==",		2)
-CHAR("=~",			2,	"=~",		2)
-CHAR(">=",			2, 	">=",		2)
-CHAR("AE",			2,	"AE",		2)
-CHAR("AN",			2,	"^",		1)
-CHAR("Ah",			2,	"N",		1)
-CHAR("Bq",			2, 	",,",		2)
-CHAR("Cs",			2,	"x",		1)
-CHAR("Do",			2,	"$",		1)
-CHAR("Eu",			2,	"EUR",		3)
-CHAR("Fc",			2,	">>",		2)
-CHAR("Fi",			2,	"ffi",		3)
-CHAR("Fl",			2,	"ffl",		3)
-CHAR("Fn",			2,	"f",		1)
-CHAR("Fo",			2,	"<<",		2)
-CHAR("Im",			2,	"I",		1)
-CHAR("OE",			2,	"OE",		2)
-CHAR("OR",			2,	"v",		1)
-CHAR("Po",			2,	"L",		1)
-CHAR("Re",			2,	"R",		1)
-CHAR("Sd",			2,	"o",		1)
-CHAR("TP",			2,	"b",		1)
-CHAR("Tp",			2,	"b",		1)
-CHAR("Ye",			2,	"Y",		1)
-CHAR("\'",			1,	"\'",		1)
-CHAR("\'A",			2,	"A",		1)
-CHAR("\'E",			2,	"E",		1)
-CHAR("\'I",			2,	"I",		1)
-CHAR("\'O",			2,	"O",		1)
-CHAR("\'U",			2,	"U",		1)
-CHAR("\'a",			2,	"a",		1)
-CHAR("\'e",			2,	"e",		1)
-CHAR("\'i",			2,	"i",		1)
-CHAR("\'o",			2,	"o",		1)
-CHAR("\'u",			2,	"u",		1)
-CHAR("\\",			1,	"\\",		1)
-CHAR("^",			1,	"",		0)
-CHAR("^A",			2,	"A",		1)
-CHAR("^E",			2,	"E",		1)
-CHAR("^I",			2,	"I",		1)
-CHAR("^O",			2,	"O",		1)
-CHAR("^U",			2,	"U",		1)
-CHAR("^a",			2,	"a",		1)
-CHAR("^e",			2,	"e",		1)
-CHAR("^i",			2,	"i",		1)
-CHAR("^o",			2,	"o",		1)
-CHAR("^u",			2,	"u",		1)
-CHAR("`",			1,	"`",		1)
-CHAR("`A",			2,	"A",		1)
-CHAR("`E",			2,	"E",		1)
-CHAR("`I",			2,	"I",		1)
-CHAR("`O",			2,	"O",		1)
-CHAR("`U",			2,	"U",		1)
-CHAR("`a",			2,	"a",		1)
-CHAR("`e",			2,	"e",		1)
-CHAR("`i",			2,	"i",		1)
-CHAR("`o",			2,	"o",		1)
-CHAR("`u",			2,	"u",		1)
-CHAR("a-",			2,	"-",		1)
-CHAR("a\"",			2,	"\"",		1)
-CHAR("a^",			2,	"^",		1)
-CHAR("aa",			2, 	"\'",		1)
-CHAR("ab",			2,	"`",		1)
-CHAR("ac",			2,	",",		1)
-CHAR("ad",			2,	"\"",		1)
-CHAR("ae",			2,	"ae",		2)
-CHAR("ah",			2,	"v",		1)
-CHAR("ao",			2,	"o",		1)
-CHAR("ap",			2,	"~",		1)
-CHAR("aq",			2, 	"\'",		1)
-CHAR("a~",			2, 	"~",		1)
-CHAR("ba",			2, 	"|",		1)
-CHAR("bb",			2, 	"|",		1)
-CHAR("bq",			2, 	",",		1)
-CHAR("bu",			2, 	"o",		1)
-CHAR("c",			1,	"",		0)
-CHAR("ca",			2,	"(^)",		3)
-CHAR("ci",			2, 	"O",		1)
-CHAR("co",			2, 	"(C)",		3)
-CHAR("ct",			2,	"c",		1)
-CHAR("cu",			2,	"U",		1)
-CHAR("dA",			2, 	"v",		1)
-CHAR("da",			2, 	"v",		1)
-CHAR("dd",			2,	"=",		1)
-CHAR("de",			2,	"o",		1)
-CHAR("dg",			2,	"-",		1)
-CHAR("di",			2,	"-:-",		3)
-CHAR("e",			1,	"\\",		1)
-CHAR("em",			2, 	"--",		2)
-CHAR("en",			2, 	"-",		1)
-CHAR("eq",			2,	"=",		1)
-CHAR("es",			2,	"{}",		2)
-CHAR("eu",			2,	"EUR",		3)
-CHAR("fa",			2,	"V",		1)
-CHAR("fc",			2,	">",		1)
-CHAR("ff",			2,	"ff",		2)
-CHAR("fi",			2,	"fi",		2)
-CHAR("fl",			2,	"fl",		2)
-CHAR("fo",			2,	"<",		1)
-CHAR("ga",			2, 	"`",		1)
-CHAR("gr",			2,	"V",		1)
-CHAR("hA",			2, 	"<=>",		3)
-CHAR("ho",			2,	",",		1)
-CHAR("hy",			2, 	"-",		1)
-CHAR("if",			2, 	"oo",		2)
-CHAR("lA",			2, 	"<=",		2)
-CHAR("lB",			2, 	"[",		1)
-CHAR("lC",			2, 	"{",		1)
-CHAR("la",			2, 	"<",		1)
-CHAR("lh",			2,	"<=",		2)
-CHAR("mo",			2,	"E",		1)
-CHAR("mu",			2,	"x",		1)
-CHAR("na",			2, 	"NaN",		3)
-CHAR("nm",			2,	"E",		1)
-CHAR("no",			2,	"~",		1)
-CHAR("oA",			2,	"A",		1)
-CHAR("oa",			2,	"a",		1)
-CHAR("oe",			2,	"oe",		2)
-CHAR("oq",			2, 	"`",		1)
-CHAR("pd",			2,	"a",		1)
-CHAR("pl",			2,	"+",		1)
-CHAR("ps",			2,	"9|",		2)
-CHAR("r!",			2,	"i",		1)
-CHAR("r?",			2,	"c",		1)
-CHAR("rA",			2, 	"=>",		2)
-CHAR("rB",			2, 	"]",		1)
-CHAR("rC",			2, 	"}",		1)
-CHAR("ra",			2, 	">",		1)
-CHAR("rg",			2, 	"(R)",		3)
-CHAR("rh",			2,	"=>",		2)
-CHAR("sc",			2,	"S",		1)
-CHAR("ss",			2,	"ss",		2)
-CHAR("st",			2,	"-)",		2)
-CHAR("te",			2,	"3",		1)
-CHAR("tf",			2,	".:.",		3)
-CHAR("tm",			2, 	"tm",		2)
-CHAR("ts",			2,	"s",		1)
-CHAR("uA",			2, 	"^",		1)
-CHAR("ua",			2, 	"^",		1)
-CHAR("|",			1,	"",		0)
-CHAR("~",			1,	" ",		1)
-CHAR("~=",			2,	"~=",		2)
-CHAR("~A",			2,	"A",		1)
-CHAR("~N",			2,	"N",		1)
-CHAR("~O",			2,	"O",		1)
-CHAR("~a",			2,	"a",		1)
-CHAR("~n",			2,	"n",		1)
-CHAR("~o",			2,	"o",		1)
-CHAR("~~",			2,	"~~",		2)
diff --git a/usr.bin/mandoc/chars.c b/usr.bin/mandoc/chars.c
new file mode 100644
index 00000000000..587da515789
--- /dev/null
+++ b/usr.bin/mandoc/chars.c
@@ -0,0 +1,204 @@
+/*	$Id: chars.c,v 1.1 2009/10/19 09:16:58 schwarze Exp $ */
+/*
+ * Copyright (c) 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.
+ */
+#include <assert.h>
+#include <err.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "chars.h"
+
+#define	PRINT_HI	 126
+#define	PRINT_LO	 32
+
+struct	ln {
+	struct ln	 *next;
+	const char	 *code;
+	const char	 *ascii;
+	const char	 *html;
+	size_t		  codesz;
+	size_t		  asciisz;
+	size_t		  htmlsz;
+	int		  type;
+#define	CHARS_CHAR	 (1 << 0)
+#define	CHARS_STRING	 (1 << 1)
+#define CHARS_BOTH	 (0x03)
+};
+
+#define	LINES_MAX	  266
+
+#define CHAR(w, x, y, z, a, b) \
+	{ NULL, (w), (y), (a), (x), (z), (b), CHARS_CHAR },
+#define STRING(w, x, y, z, a, b) \
+	{ NULL, (w), (y), (a), (x), (z), (b), CHARS_STRING },
+#define BOTH(w, x, y, z, a, b) \
+	{ NULL, (w), (y), (a), (x), (z), (b), CHARS_BOTH },
+
+static	struct ln lines[LINES_MAX] = {
+#include "chars.in"
+};
+
+struct	tbl {
+	enum chars	  type;
+	struct ln	**htab;
+};
+
+static	inline int	  match(const struct ln *,
+				const char *, size_t, int);
+static	const char	 *find(struct tbl *, const char *, 
+				size_t, size_t *, int);
+
+
+void
+chars_free(void *arg)
+{
+	struct tbl	*tab;
+
+	tab = (struct tbl *)arg;
+
+	free(tab->htab);
+	free(tab);
+}
+
+
+void *
+chars_init(enum chars type)
+{
+	struct tbl	 *tab;
+	struct ln	**htab;
+	struct ln	 *pp;
+	int		  i, hash;
+
+	/*
+	 * Constructs a very basic chaining hashtable.  The hash routine
+	 * is simply the integral value of the first character.
+	 * Subsequent entries are chained in the order they're processed
+	 * (they're in-line re-ordered during lookup).
+	 */
+
+	if (NULL == (tab = malloc(sizeof(struct tbl))))
+		err(1, "malloc");
+	tab->type = type;
+
+	htab = calloc(PRINT_HI - PRINT_LO + 1, sizeof(struct ln **));
+	if (NULL == htab)
+		err(1, "malloc");
+
+	for (i = 0; i < LINES_MAX; i++) {
+		hash = (int)lines[i].code[0] - PRINT_LO;
+
+		if (NULL == (pp = htab[hash])) {
+			htab[hash] = &lines[i];
+			continue;
+		}
+
+		for ( ; pp->next; pp = pp->next)
+			/* Scan ahead. */ ;
+		pp->next = &lines[i];
+	}
+
+	tab->htab = htab;
+	return(tab);
+}
+
+
+const char *
+chars_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz)
+{
+
+	return(find((struct tbl *)arg, p, sz, rsz, CHARS_CHAR));
+}
+
+
+const char *
+chars_a2res(void *arg, const char *p, size_t sz, size_t *rsz)
+{
+
+	return(find((struct tbl *)arg, p, sz, rsz, CHARS_STRING));
+}
+
+
+static const char *
+find(struct tbl *tab, const char *p, size_t sz, size_t *rsz, int type)
+{
+	struct ln	 *pp, *prev;
+	struct ln	**htab;
+	int		  hash;
+
+	assert(p);
+	assert(sz > 0);
+
+	if (p[0] < PRINT_LO || p[0] > PRINT_HI)
+		return(NULL);
+
+	/*
+	 * Lookup the symbol in the symbol hash.  See ascii2htab for the
+	 * hashtable specs.  This dynamically re-orders the hash chain
+	 * to optimise for repeat hits.
+	 */
+
+	hash = (int)p[0] - PRINT_LO;
+	htab = tab->htab;
+
+	if (NULL == (pp = htab[hash]))
+		return(NULL);
+
+	if (NULL == pp->next) {
+		if ( ! match(pp, p, sz, type)) 
+			return(NULL);
+
+		if (CHARS_HTML == tab->type) {
+			*rsz = pp->htmlsz;
+			return(pp->html);
+		}
+		*rsz = pp->asciisz;
+		return(pp->ascii);
+	}
+
+	for (prev = NULL; pp; pp = pp->next) {
+		if ( ! match(pp, p, sz, type)) {
+			prev = pp;
+			continue;
+		}
+
+		if (prev) {
+			prev->next = pp->next;
+			pp->next = htab[hash];
+			htab[hash] = pp;
+		}
+
+		if (CHARS_HTML == tab->type) {
+			*rsz = pp->htmlsz;
+			return(pp->html);
+		}
+		*rsz = pp->asciisz;
+		return(pp->ascii);
+	}
+
+	return(NULL);
+}
+
+
+static inline int
+match(const struct ln *ln, const char *p, size_t sz, int type)
+{
+
+	if ( ! (ln->type & type))
+		return(0);
+	if (ln->codesz != sz)
+		return(0);
+	return(0 == strncmp(ln->code, p, sz));
+}
diff --git a/usr.bin/mandoc/chars.h b/usr.bin/mandoc/chars.h
new file mode 100644
index 00000000000..fa1608a3b58
--- /dev/null
+++ b/usr.bin/mandoc/chars.h
@@ -0,0 +1,34 @@
+/*	$Id: chars.h,v 1.1 2009/10/19 09:16:58 schwarze Exp $ */
+/*
+ * 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/usr.bin/mandoc/chars.in b/usr.bin/mandoc/chars.in
new file mode 100644
index 00000000000..83ffaf28bfb
--- /dev/null
+++ b/usr.bin/mandoc/chars.in
@@ -0,0 +1,300 @@
+/*	$Id: chars.in,v 1.1 2009/10/19 09:16:58 schwarze Exp $ */
+/*
+ * Copyright (c) 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.
+ */
+
+/*
+ * The ASCII translation tables.  STRING corresponds to predefined
+ * strings (cf. mdoc_samples.7 and tmac/mdoc/doc-nroff).  CHAR
+ * corresponds to special characters (cf. groff_char.7).  BOTH contains
+ * sequences that are equivalent in both STRING and CHAR.
+ *
+ * Either way, the left-hand side corresponds to the input sequence (\x,
+ * \(xx, \*(xx and so on) whose length is listed second element.  The
+ * right-hand side is what's produced by the front-end, with the fourth
+ * element being its length.
+ *
+ * XXX - C-escape strings!
+ * XXX - update LINES_MAX if adding more!
+ */
+
+STRING("Am",			2,	"&",		1,	"&amp;",	5)
+STRING("Ba",			2,	"|",		1,	"|",		0)
+STRING("Ge",			2,	">=",		2,	"&#8805;",	7)
+STRING("Gt",			2,	">",		1,	"&gt;",		4)
+STRING("If",			2,	"infinity",	8,	"infinity",	8)
+STRING("Le",			2,	"<=",		2,	"&#8804;",	7)
+STRING("Lq",			2,	"``",		2,	"&#8220;",	7)
+STRING("Lt",			2,	"<",		1,	"&lt;",		4)
+STRING("Na",			2,	"NaN",		3,	"NaN",		3)
+STRING("Ne",			2,	"!=",		2,	"&#8800;",	7)
+STRING("Pi",			2,	"pi",		2,	"&#960;",	6)
+STRING("Pm",			2,	"+-",		2,	"&#177;",	6)
+STRING("R",			1,	"(R)",		3,	"&#174;",	6)
+STRING("Rq",			2,	"\'\'",		2,	"&#8221;",	7)
+STRING("Tm",			2,	"tm",		2,	"&#8424;",	7)
+STRING("left-bracket",		12,	"[",		1,	"[",		1)
+STRING("left-parenthesis",	16,	"(",		1,	"(",		1)
+STRING("left-singlequote",	16,	"`",		1,	"&#8216;",	7)
+STRING("lp",			2,	"(",		1,	"(",		1)
+STRING("q",			1,	"\"",		1,	"&quot;",	6)
+STRING("quote-left",		10,	"`",		1,	"&#8216;",	7)
+STRING("quote-right",		11,	"\'",		1,	"&#8217;",	7)
+STRING("right-bracket",		13,	"]",		1,	"]",		1)
+STRING("right-parenthesis",	17,	")",		1,	")",		1)
+STRING("right-singlequote",	17,	"\'",		1,	"&#8217;",	7)
+STRING("rp",			2,	")",		1,	")",		1)
+
+BOTH("<=",			2,	"<=",		2,	"&#8804;",	7)
+BOTH(">=",			2,	">=",		2,	"&#8804;",	7)
+BOTH("aa",			2,	"\'",		1,	"&#180;",	6)
+BOTH("ga",			2,	"`",		1,	"`",		1)
+BOTH("lq",			2, 	"``",		2,	"&#8220;",	7)
+BOTH("rq",			2, 	"\'\'",		2,	"&#8221;",	7)
+BOTH("ua",			2,	"^",		1,	"",	0)
+
+CHAR(" ",			1,	" ",		1,	" ",		1)
+CHAR("!=",			2, 	"!=",		2,	"",	0)
+CHAR("%",			1,	"",		0,	"",	0)
+CHAR("&",			1,	"",		0,	"",	0)
+CHAR("(=",			2,	"(=",		2,	"",	0)
+CHAR("**",			2, 	"*",		1,	"",	0)
+CHAR("*A",			2,	"A",		1,	"",	0)
+CHAR("*B",			2,	"B",		1,	"",	0)
+CHAR("*C",			2,	"H",		1,	"",	0)
+CHAR("*D",			2,	"/\\",		2,	"",	0)
+CHAR("*E",			2,	"E",		1,	"",	0)
+CHAR("*F",			2,	"O_",		1,	"",	0)
+CHAR("*G",			2,	"|",		1,	"",	0)
+CHAR("*H",			2,	"O",		1,	"",	0)
+CHAR("*I",			2,	"I",		1,	"",	0)
+CHAR("*K",			2,	"K",		1,	"",	0)
+CHAR("*L",			2,	"/\\",		2,	"",	0)
+CHAR("*M",			2,	"M",		1,	"",	0)
+CHAR("*N",			2,	"N",		1,	"",	0)
+CHAR("*O",			2,	"O",		1,	"",	0)
+CHAR("*P",			2,	"TT",		2,	"",	0)
+CHAR("*Q",			2,	"Y",		1,	"",	0)
+CHAR("*R",			2,	"P",		1,	"",	0)
+CHAR("*S",			2,	">",		1,	"",	0)
+CHAR("*T",			2,	"T",		1,	"",	0)
+CHAR("*U",			2,	"Y",		1,	"",	0)
+CHAR("*W",			2,	"O",		1,	"",	0)
+CHAR("*X",			2,	"X",		1,	"",	0)
+CHAR("*Y",			2,	"H",		1,	"",	0)
+CHAR("*Z",			2,	"Z",		1,	"",	0)
+CHAR("*a",			2,	"a",		1,	"",	0)
+CHAR("*b",			2,	"B",		1,	"",	0)
+CHAR("*c",			2,	"E",		1,	"",	0)
+CHAR("*d",			2,	"d",		1,	"",	0)
+CHAR("*e",			2,	"e",		1,	"",	0)
+CHAR("*f",			2,	"o",		1,	"",	0)
+CHAR("*g",			2,	"y",		1,	"",	0)
+CHAR("*h",			2,	"0",		1,	"",	0)
+CHAR("*i",			2,	"i",		1,	"",	0)
+CHAR("*k",			2,	"k",		1,	"",	0)
+CHAR("*l",			2,	"\\",		1,	"",	0)
+CHAR("*m",			2,	"u",		1,	"",	0)
+CHAR("*n",			2,	"v",		1,	"",	0)
+CHAR("*o",			2,	"o",		1,	"",	0)
+CHAR("*p",			2,	"n",		1,	"",	0)
+CHAR("*q",			2,	"u",		1,	"",	0)
+CHAR("*r",			2,	"p",		1,	"",	0)
+CHAR("*s",			2,	"o",		1,	"",	0)
+CHAR("*t",			2,	"t",		1,	"",	0)
+CHAR("*u",			2,	"u",		1,	"",	0)
+CHAR("*w",			2,	"w",		1,	"",	0)
+CHAR("*x",			2,	"x",		1,	"",	0)
+CHAR("*y",			2,	"n",		1,	"",	0)
+CHAR("*z",			2,	"C",		1,	"",	0)
+CHAR("+-",			2, 	"+-",		2,	"",	0)
+CHAR("+f",			2,	"o",		1,	"",	0)
+CHAR("+h",			2,	"0",		1,	"",	0)
+CHAR("+p",			2,	"w",		1,	"",	0)
+CHAR(",C",			2,	"C",		1,	"",	0)
+CHAR(",c",			2,	"c",		1,	"",	0)
+CHAR("-",			1,	"-",		1,	"-",		1)
+CHAR("->",			2, 	"->",		2,	"",	0)
+CHAR("-D",			2,	"D",		1,	"",	0)
+CHAR(".",			1,	".",		1,	".",		1)
+CHAR("/L",			2,	"L",		1,	"",	0)
+CHAR("/O",			2,	"O",		1,	"",	0)
+CHAR("/l",			2,	"l",		1,	"",	0)
+CHAR("/o",			2,	"o",		1,	"",	0)
+CHAR("0",			1,	" ",		1,	"",	0)
+CHAR(":A",			2,	"A",		1,	"",	0)
+CHAR(":E",			2,	"E",		1,	"",	0)
+CHAR(":I",			2,	"I",		1,	"",	0)
+CHAR(":O",			2,	"O",		1,	"",	0)
+CHAR(":U",			2,	"U",		1,	"",	0)
+CHAR(":a",			2,	"a",		1,	"",	0)
+CHAR(":e",			2,	"e",		1,	"",	0)
+CHAR(":i",			2,	"i",		1,	"",	0)
+CHAR(":o",			2,	"o",		1,	"",	0)
+CHAR(":u",			2,	"u",		1,	"",	0)
+CHAR(":y",			2,	"y",		1,	"",	0)
+CHAR("<-",			2, 	"<-",		2,	"",	0)
+CHAR("<=",			2, 	"<=",		2,	"",	0)
+CHAR("<>",			2, 	"<>",		2,	"",	0)
+CHAR("=)",			2,	"=)",		2,	"",	0)
+CHAR("==",			2, 	"==",		2,	"",	0)
+CHAR("=~",			2,	"=~",		2,	"",	0)
+CHAR(">=",			2, 	">=",		2,	"",	0)
+CHAR("AE",			2,	"AE",		2,	"",	0)
+CHAR("AN",			2,	"^",		1,	"",	0)
+CHAR("Ah",			2,	"N",		1,	"",	0)
+CHAR("Bq",			2, 	",,",		2,	"",	0)
+CHAR("Cs",			2,	"x",		1,	"",	0)
+CHAR("Do",			2,	"$",		1,	"",	0)
+CHAR("Eu",			2,	"EUR",		3,	"",	0)
+CHAR("Fc",			2,	">>",		2,	"",	0)
+CHAR("Fi",			2,	"ffi",		3,	"",	0)
+CHAR("Fl",			2,	"ffl",		3,	"",	0)
+CHAR("Fn",			2,	"f",		1,	"",	0)
+CHAR("Fo",			2,	"<<",		2,	"",	0)
+CHAR("Im",			2,	"I",		1,	"",	0)
+CHAR("OE",			2,	"OE",		2,	"",	0)
+CHAR("OR",			2,	"v",		1,	"",	0)
+CHAR("Po",			2,	"L",		1,	"",	0)
+CHAR("Re",			2,	"R",		1,	"",	0)
+CHAR("Sd",			2,	"o",		1,	"",	0)
+CHAR("TP",			2,	"b",		1,	"",	0)
+CHAR("Tp",			2,	"b",		1,	"",	0)
+CHAR("Ye",			2,	"Y",		1,	"",	0)
+CHAR("\'",			1,	"\'",		1,	"",	0)
+CHAR("\'A",			2,	"A",		1,	"",	0)
+CHAR("\'E",			2,	"E",		1,	"",	0)
+CHAR("\'I",			2,	"I",		1,	"",	0)
+CHAR("\'O",			2,	"O",		1,	"",	0)
+CHAR("\'U",			2,	"U",		1,	"",	0)
+CHAR("\'a",			2,	"a",		1,	"",	0)
+CHAR("\'e",			2,	"e",		1,	"",	0)
+CHAR("\'i",			2,	"i",		1,	"",	0)
+CHAR("\'o",			2,	"o",		1,	"",	0)
+CHAR("\'u",			2,	"u",		1,	"",	0)
+CHAR("\\",			1,	"\\",		1,	"",	0)
+CHAR("^",			1,	"",		0,	"",	0)
+CHAR("^A",			2,	"A",		1,	"",	0)
+CHAR("^E",			2,	"E",		1,	"",	0)
+CHAR("^I",			2,	"I",		1,	"",	0)
+CHAR("^O",			2,	"O",		1,	"",	0)
+CHAR("^U",			2,	"U",		1,	"",	0)
+CHAR("^a",			2,	"a",		1,	"",	0)
+CHAR("^e",			2,	"e",		1,	"",	0)
+CHAR("^i",			2,	"i",		1,	"",	0)
+CHAR("^o",			2,	"o",		1,	"",	0)
+CHAR("^u",			2,	"u",		1,	"",	0)
+CHAR("`",			1,	"`",		1,	"",	0)
+CHAR("`A",			2,	"A",		1,	"",	0)
+CHAR("`E",			2,	"E",		1,	"",	0)
+CHAR("`I",			2,	"I",		1,	"",	0)
+CHAR("`O",			2,	"O",		1,	"",	0)
+CHAR("`U",			2,	"U",		1,	"",	0)
+CHAR("`a",			2,	"a",		1,	"",	0)
+CHAR("`e",			2,	"e",		1,	"",	0)
+CHAR("`i",			2,	"i",		1,	"",	0)
+CHAR("`o",			2,	"o",		1,	"",	0)
+CHAR("`u",			2,	"u",		1,	"",	0)
+CHAR("a-",			2,	"-",		1,	"",	0)
+CHAR("a\"",			2,	"\"",		1,	"",	0)
+CHAR("a^",			2,	"^",		1,	"",	0)
+CHAR("aa",			2, 	"\'",		1,	"",	0)
+CHAR("ab",			2,	"`",		1,	"",	0)
+CHAR("ac",			2,	",",		1,	"",	0)
+CHAR("ad",			2,	"\"",		1,	"",	0)
+CHAR("ae",			2,	"ae",		2,	"",	0)
+CHAR("ah",			2,	"v",		1,	"",	0)
+CHAR("ao",			2,	"o",		1,	"",	0)
+CHAR("ap",			2,	"~",		1,	"",	0)
+CHAR("aq",			2, 	"\'",		1,	"&#8217;",	7)
+CHAR("a~",			2, 	"~",		1,	"",	0)
+CHAR("ba",			2, 	"|",		1,	"",	0)
+CHAR("bb",			2, 	"|",		1,	"",	0)
+CHAR("bq",			2, 	",",		1,	"",	0)
+CHAR("bu",			2, 	"o",		1,	"&#8226;",	7)
+CHAR("c",			1,	"",		0,	"",	0)
+CHAR("ca",			2,	"(^)",		3,	"",	0)
+CHAR("ci",			2, 	"O",		1,	"",	0)
+CHAR("co",			2, 	"(C)",		3,	"",	0)
+CHAR("ct",			2,	"c",		1,	"",	0)
+CHAR("cu",			2,	"U",		1,	"",	0)
+CHAR("dA",			2, 	"v",		1,	"",	0)
+CHAR("da",			2, 	"v",		1,	"",	0)
+CHAR("dd",			2,	"=",		1,	"",	0)
+CHAR("de",			2,	"o",		1,	"",	0)
+CHAR("dg",			2,	"-",		1,	"",	0)
+CHAR("di",			2,	"-:-",		3,	"",	0)
+CHAR("e",			1,	"\\",		1,	"\\",		1)
+CHAR("em",			2, 	"--",		2,	"&#8212;",	7)
+CHAR("en",			2, 	"-",		1,	"&#8211;",	7)
+CHAR("eq",			2,	"=",		1,	"",	0)
+CHAR("es",			2,	"{}",		2,	"",	0)
+CHAR("eu",			2,	"EUR",		3,	"",	0)
+CHAR("fa",			2,	"V",		1,	"",	0)
+CHAR("fc",			2,	">",		1,	"",	0)
+CHAR("ff",			2,	"ff",		2,	"",	0)
+CHAR("fi",			2,	"fi",		2,	"",	0)
+CHAR("fl",			2,	"fl",		2,	"",	0)
+CHAR("fo",			2,	"<",		1,	"",	0)
+CHAR("ga",			2, 	"`",		1,	"",	0)
+CHAR("gr",			2,	"V",		1,	"",	0)
+CHAR("hA",			2, 	"<=>",		3,	"",	0)
+CHAR("ho",			2,	",",		1,	"",	0)
+CHAR("hy",			2, 	"-",		1,	"",	0)
+CHAR("if",			2, 	"oo",		2,	"",	0)
+CHAR("lA",			2, 	"<=",		2,	"",	0)
+CHAR("lB",			2, 	"[",		1,	"[",		1)
+CHAR("lC",			2, 	"{",		1,	"",	0)
+CHAR("la",			2, 	"<",		1,	"&lt;",		4)
+CHAR("lh",			2,	"<=",		2,	"",	0)
+CHAR("mo",			2,	"E",		1,	"",	0)
+CHAR("mu",			2,	"x",		1,	"",	0)
+CHAR("na",			2, 	"NaN",		3,	"",	0)
+CHAR("nm",			2,	"E",		1,	"",	0)
+CHAR("no",			2,	"~",		1,	"",	0)
+CHAR("oA",			2,	"A",		1,	"",	0)
+CHAR("oa",			2,	"a",		1,	"",	0)
+CHAR("oe",			2,	"oe",		2,	"",	0)
+CHAR("oq",			2, 	"`",		1,	"&#8216;",	7)
+CHAR("pd",			2,	"a",		1,	"",	0)
+CHAR("pl",			2,	"+",		1,	"",	0)
+CHAR("ps",			2,	"9|",		2,	"",	0)
+CHAR("r!",			2,	"i",		1,	"",	0)
+CHAR("r?",			2,	"c",		1,	"",	0)
+CHAR("rA",			2, 	"=>",		2,	"",	0)
+CHAR("rB",			2, 	"]",		1,	"]",		1)
+CHAR("rC",			2, 	"}",		1,	"",	0)
+CHAR("ra",			2, 	">",		1,	"&gt;",		4)
+CHAR("rg",			2, 	"(R)",		3,	"",	0)
+CHAR("rh",			2,	"=>",		2,	"",	0)
+CHAR("sc",			2,	"S",		1,	"",	0)
+CHAR("ss",			2,	"ss",		2,	"",	0)
+CHAR("st",			2,	"-)",		2,	"",	0)
+CHAR("te",			2,	"3",		1,	"",	0)
+CHAR("tf",			2,	".:.",		3,	"",	0)
+CHAR("tm",			2, 	"tm",		2,	"",	0)
+CHAR("ts",			2,	"s",		1,	"",	0)
+CHAR("uA",			2, 	"^",		1,	"",	0)
+CHAR("ua",			2, 	"^",		1,	"",	0)
+CHAR("|",			1,	"",		0,	"",	0)
+CHAR("~",			1,	" ",		1,	"",	0)
+CHAR("~=",			2,	"~=",		2,	"",	0)
+CHAR("~A",			2,	"A",		1,	"",	0)
+CHAR("~N",			2,	"N",		1,	"",	0)
+CHAR("~O",			2,	"O",		1,	"",	0)
+CHAR("~a",			2,	"a",		1,	"",	0)
+CHAR("~n",			2,	"n",		1,	"",	0)
+CHAR("~o",			2,	"o",		1,	"",	0)
+CHAR("~~",			2,	"~~",		2,	"",	0)
diff --git a/usr.bin/mandoc/term.c b/usr.bin/mandoc/term.c
index a6a1c36502d..c61f3ac193b 100644
--- a/usr.bin/mandoc/term.c
+++ b/usr.bin/mandoc/term.c
@@ -1,4 +1,4 @@
-/*	$Id: term.c,v 1.14 2009/10/18 21:03:31 schwarze Exp $ */
+/*	$Id: term.c,v 1.15 2009/10/19 09:16:58 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -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);
 }
@@ -89,8 +90,8 @@ term_free(struct termp *p)
 
 	if (p->buf)
 		free(p->buf);
-	if (TERMENC_ASCII == p->enc && p->symtab)
-		term_asciifree(p->symtab);
+	if (p->symtab)
+		chars_free(p->symtab);
 
 	free(p);
 }
@@ -102,7 +103,7 @@ term_alloc(enum termenc enc)
 	struct termp *p;
 
 	if (NULL == (p = malloc(sizeof(struct termp))))
-		err(1, "malloc");
+		return(NULL);
 	bzero(p, sizeof(struct termp));
 	p->maxrmargin = 78;
 	p->enc = enc;
@@ -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
@@ -555,7 +556,7 @@ buffer(struct termp *p, char c)
 		s = p->maxcols * 2;
 		p->buf = realloc(p->buf, s);
 		if (NULL == p->buf)
-			err(1, "realloc");
+			err(1, "realloc"); /* FIXME: shouldn't be here! */
 		p->maxcols = s;
 	}
 	p->buf[(int)(p->col)++] = c;
diff --git a/usr.bin/mandoc/term.h b/usr.bin/mandoc/term.h
index 618e6eb4818..61e5a5a3876 100644
--- a/usr.bin/mandoc/term.h
+++ b/usr.bin/mandoc/term.h
@@ -1,4 +1,4 @@
-/*	$Id: term.h,v 1.10 2009/09/21 20:28:43 schwarze Exp $ */
+/*	$Id: term.h,v 1.11 2009/10/19 09:16:58 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -20,9 +20,7 @@
 __BEGIN_DECLS
 
 enum	termenc {
-	TERMENC_ASCII,
-	TERMENC_LATIN1,		/* Not implemented. */
-	TERMENC_UTF8		/* Not implemented. */
+	TERMENC_ASCII
 };
 
 struct	termp {
@@ -50,11 +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 *);