sync to 1.8.5: partial rewrite of the handling of escape sequences:

distinguish special characters and predefined strings,
and add and fix several escape sequences

4 files changed, 391 insertions, 261 deletions

diff --git a/usr.bin/mandoc/ascii.c b/usr.bin/mandoc/ascii.c
index 7cc89f6c66f..89816232a9d 100644
--- a/usr.bin/mandoc/ascii.c
+++ b/usr.bin/mandoc/ascii.c
@@ -1,4 +1,4 @@
-/*	$Id: ascii.c,v 1.2 2009/06/14 23:00:57 schwarze Exp $ */
+/*	$Id: ascii.c,v 1.3 2009/08/22 17:04:48 schwarze Exp $ */
 /*
  * Copyright (c) 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -27,9 +27,12 @@
 struct	line {
 	const char	 *code;
 	const char	 *out;
-	/* 32- and 64-bit alignment safe. */
 	size_t		  codesz;
 	size_t		  outsz;
+	int		  type;
+#define	ASCII_CHAR	 (1 << 0)
+#define	ASCII_STRING	 (1 << 1)
+#define ASCII_BOTH	 (0x03)
 };
 
 struct	linep {
@@ -37,8 +40,12 @@ struct	linep {
 	struct linep	 *next;
 };
 
-#define LINE(w, x, y, z) \
-	{ (w), (y), (x), (z) },
+#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"
 };
@@ -50,7 +57,9 @@ struct	asciitab {
 
 
 static	inline int	  match(const struct line *,
-				const char *, size_t);
+				const char *, size_t, int);
+static	const char *	  lookup(struct asciitab *, const char *, 
+				size_t, size_t *, int);
 
 
 void
@@ -125,14 +134,29 @@ term_ascii2htab(void)
 const char *
 term_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz)
 {
-	struct asciitab	 *tab;
+
+	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;
 
-	tab = (struct asciitab *)arg;
-	htab = tab->htab;
-
 	assert(p);
 	assert(sz > 0);
 
@@ -147,19 +171,20 @@ term_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz)
 	 */
 
 	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)) 
+		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)) {
+		if ( ! match(pp->line, p, sz, type)) {
 			prev = pp;
 			continue;
 		}
@@ -181,9 +206,11 @@ term_a2ascii(void *arg, const char *p, size_t sz, size_t *rsz)
 
 
 static inline int
-match(const struct line *line, const char *p, size_t sz)
+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
index 40f190da2e0..56f72d83bf3 100644
--- a/usr.bin/mandoc/ascii.in
+++ b/usr.bin/mandoc/ascii.in
@@ -1,4 +1,4 @@
-/*	$Id: ascii.in,v 1.4 2009/07/13 00:18:12 schwarze Exp $ */
+/*	$Id: ascii.in,v 1.5 2009/08/22 17:04:48 schwarze Exp $ */
 /*
  * Copyright (c) 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -16,205 +16,284 @@
  */
 
 /*
- * The ASCII translation table.  The left-hand side corresponds to the
- * escape sequence (\x, \(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.
+ * 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.
  *
- * Be sure to escape strings.
+ * 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!
  */
 
-LINE("\\", 1,	  "\\", 1)
-LINE("\'", 1,	  "\'", 1)
-LINE("`", 1,	  "`", 1)
-LINE("%", 1,	  "%", 1)
-LINE("-", 1,	  "-", 1)
-LINE(" ", 1,	  " ", 1)
-LINE("~", 1,	  " ", 1)
-LINE("^", 1,	  "", 0)
-LINE("0", 1,	  " ", 1)
-LINE(".", 1,	  ".", 1)
-LINE("&", 1,	  "", 0)
-LINE("e", 1,	  "\\", 1)
-LINE("q", 1,	  "\"", 1)
-LINE("|", 1,	  "", 0)
-LINE("rC", 2, 	  "}", 1)
-LINE("lC", 2, 	  "{", 1)
-LINE("rB", 2, 	  "]", 1)
-LINE("lB", 2, 	  "[", 1)
-LINE("ra", 2, 	  ">", 1)
-LINE("la", 2, 	  "<", 1)
-LINE("Lq", 2, 	  "``", 2)
-LINE("lq", 2, 	  "``", 2)
-LINE("Rq", 2, 	  "\'\'", 2)
-LINE("rq", 2, 	  "\'\'", 2)
-LINE("oq", 2, 	  "`", 1)
-LINE("aq", 2, 	  "\'", 1)
-LINE("Bq", 2, 	  ",,", 2)
-LINE("bq", 2, 	  ",,", 2)
-LINE("<-", 2, 	  "<-", 2)
-LINE("->", 2, 	  "->", 2)
-LINE("<>", 2, 	  "<>", 2)
-LINE("ua", 2, 	  "^", 1)
-LINE("da", 2, 	  "v", 1)
-LINE("bu", 2, 	  "o", 1)
-LINE("ci", 2, 	  "O", 1)
-LINE("Ba", 2, 	  "|", 1)
-LINE("ba", 2, 	  "|", 1)
-LINE("bb", 2, 	  "|", 1)
-LINE("co", 2, 	  "(C)", 3)
-LINE("rg", 2, 	  "(R)", 3)
-LINE("tm", 2, 	  "tm", 2)
-LINE("Am", 2, 	  "&", 1)
-LINE("Le", 2, 	  "<=", 2)
-LINE("<=", 2, 	  "<=", 2)
-LINE("Ge", 2, 	  ">=", 2)
-LINE(">=", 2, 	  ">=", 2)
-LINE("==", 2, 	  "==", 2)
-LINE("Ne", 2, 	  "!=", 2)
-LINE("!=", 2, 	  "!=", 2)
-LINE("Pm", 2, 	  "+-", 2)
-LINE("+-", 2, 	  "+-", 2)
-LINE("If", 2, 	  "infinity", 8)
-LINE("if", 2, 	  "oo", 2)
-LINE("Na", 2, 	  "NaN", 3)
-LINE("na", 2, 	  "NaN", 3)
-LINE("**", 2, 	  "*", 1)
-LINE("Gt", 2, 	  ">", 1)
-LINE("Lt", 2, 	  "<", 1)
-LINE("aa", 2, 	  "\'", 1)
-LINE("a~", 2, 	  "~", 1)
-LINE("ga", 2, 	  "`", 1)
-LINE("en", 2, 	  "-", 1)
-LINE("em", 2, 	  "--", 2)
-LINE("hy", 2, 	  "-", 1)
-LINE("Pi", 2, 	  "pi", 2)
-LINE("Fo", 2,	  "<<", 2)
-LINE("Fc", 2,	  ">>", 2)
-LINE("fo", 2,	  "<", 1)
-LINE("fc", 2,	  ">", 1)
-LINE("lh", 2,	  "<=", 2)
-LINE("rh", 2,	  "=>", 2)
-LINE("ae", 2,	  "ae", 2)
-LINE("AE", 2,	  "AE", 2)
-LINE("oe", 2,	  "oe", 2)
-LINE("OE", 2,	  "OE", 2)
-LINE("ss", 2,	  "ss", 2)
-LINE("\'A", 2,	  "A", 1)
-LINE("\'E", 2,	  "E", 1)
-LINE("\'I", 2,	  "I", 1)
-LINE("\'O", 2,	  "O", 1)
-LINE("\'U", 2,	  "U", 1)
-LINE("\'a", 2,	  "a", 1)
-LINE("\'e", 2,	  "e", 1)
-LINE("\'i", 2,	  "i", 1)
-LINE("\'o", 2,	  "o", 1)
-LINE("\'u", 2,	  "u", 1)
-LINE("`A", 2,	  "A", 1)
-LINE("`E", 2,	  "E", 1)
-LINE("`I", 2,	  "I", 1)
-LINE("`O", 2,	  "O", 1)
-LINE("`U", 2,	  "U", 1)
-LINE("`a", 2,	  "a", 1)
-LINE("`e", 2,	  "e", 1)
-LINE("`i", 2,	  "i", 1)
-LINE("`o", 2,	  "o", 1)
-LINE("`u", 2,	  "u", 1)
-LINE("~A", 2,	  "A", 1)
-LINE("~N", 2,	  "N", 1)
-LINE("~O", 2,	  "O", 1)
-LINE("~a", 2,	  "a", 1)
-LINE("~n", 2,	  "n", 1)
-LINE("~o", 2,	  "o", 1)
-LINE("lA", 2, 	  "<=", 2)
-LINE("rA", 2, 	  "=>", 2)
-LINE("uA", 2, 	  "^", 1)
-LINE("dA", 2, 	  "v", 1)
-LINE("hA", 2, 	  "<=>", 3)
-LINE(":A", 2,	  "A", 1)
-LINE(":E", 2,	  "E", 1)
-LINE(":I", 2,	  "I", 1)
-LINE(":O", 2,	  "O", 1)
-LINE(":U", 2,	  "U", 1)
-LINE(":a", 2,	  "a", 1)
-LINE(":e", 2,	  "e", 1)
-LINE(":i", 2,	  "i", 1)
-LINE(":o", 2,	  "o", 1)
-LINE(":u", 2,	  "u", 1)
-LINE(":y", 2,	  "y", 1)
-LINE("^A", 2,	  "A", 1)
-LINE("^E", 2,	  "E", 1)
-LINE("^I", 2,	  "I", 1)
-LINE("^O", 2,	  "O", 1)
-LINE("^U", 2,	  "U", 1)
-LINE("^a", 2,	  "a", 1)
-LINE("^e", 2,	  "e", 1)
-LINE("^i", 2,	  "i", 1)
-LINE("^o", 2,	  "o", 1)
-LINE("^u", 2,	  "u", 1)
-LINE("-D", 2,	  "D", 1)
-LINE("Sd", 2,	  "o", 1)
-LINE("TP", 2,	  "b", 1)
-LINE("Tp", 2,	  "b", 1)
-LINE(",C", 2,	  "C", 1)
-LINE(",c", 2,	  "c", 1)
-LINE("/L", 2,	  "L", 1)
-LINE("/l", 2,	  "l", 1)
-LINE("/O", 2,	  "O", 1)
-LINE("/o", 2,	  "o", 1)
-LINE("oA", 2,	  "A", 1)
-LINE("oa", 2,	  "a", 1)
-LINE("a^", 2,	  "^", 1)
-LINE("ac", 2,	  ",", 1)
-LINE("ad", 2,	  "\"", 1)
-LINE("ah", 2,	  "v", 1)
-LINE("ao", 2,	  "o", 1)
-LINE("ho", 2,	  ",", 1)
-LINE("ab", 2,	  "`", 1)
-LINE("a\"", 2,	  "\"", 1)
-LINE("a-", 2,	  "-", 1)
-LINE("Cs", 2,	  "x", 1)
-LINE("Do", 2,	  "$", 1)
-LINE("Po", 2,	  "L", 1)
-LINE("Ye", 2,	  "Y", 1)
-LINE("Fn", 2,	  "f", 1)
-LINE("ct", 2,	  "c", 1)
-LINE("ff", 2,	  "ff", 2)
-LINE("fi", 2,	  "fi", 2)
-LINE("fl", 2,	  "fl", 2)
-LINE("Fi", 2,	  "ffi", 3)
-LINE("Fl", 2,	  "ffl", 3)
-LINE("r!", 2,	  "i", 1)
-LINE("r?", 2,	  "c", 1)
-LINE("dd", 2,	  "=", 1)
-LINE("dg", 2,	  "-", 1)
-LINE("ps", 2,	  "9|", 2)
-LINE("sc", 2,	  "S", 1)
-LINE("de", 2,	  "o", 1)
-LINE("tf", 2,	  ".:.", 3)
-LINE("~~", 2,	  "~~", 2)
-LINE("~=", 2,	  "~=", 2)
-LINE("=~", 2,	  "=~", 2)
-LINE("AN", 2,	  "^", 1)
-LINE("OR", 2,	  "v", 1)
-LINE("no", 2,	  "~", 1)
-LINE("fa", 2,	  "V", 1)
-LINE("te", 2,	  "3", 1)
-LINE("Ah", 2,	  "N", 1)
-LINE("Im", 2,	  "I", 1)
-LINE("Re", 2,	  "R", 1)
-LINE("mo", 2,	  "E", 1)
-LINE("nm", 2,	  "E", 1)
-LINE("eq", 2,	  "=", 1)
-LINE("pl", 2,	  "+", 1)
-LINE("di", 2,	  "-:-", 3)
-LINE("mu", 2,	  "x", 1)
-LINE("(=", 2,	  "(=", 2)
-LINE("=)", 2,	  "=)", 2)
-LINE("ap", 2,	  "~", 1)
-LINE("pd", 2,	  "a", 1)
-LINE("gr", 2,	  "V", 1)
-LINE("ca", 2,	  "(^)", 3)
-LINE("cu", 2,	  "U", 1)
-LINE("es", 2,	  "{}", 2)
-LINE("st", 2,	  "-)", 2)
+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/term.c b/usr.bin/mandoc/term.c
index 908040530b8..324be180478 100644
--- a/usr.bin/mandoc/term.c
+++ b/usr.bin/mandoc/term.c
@@ -1,4 +1,4 @@
-/*	$Id: term.c,v 1.10 2009/08/09 21:28:57 schwarze Exp $ */
+/*	$Id: term.c,v 1.11 2009/08/22 17:04:48 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -31,13 +31,16 @@ extern	int		  mdoc_run(struct termp *,
 
 static	struct termp	 *term_alloc(enum termenc);
 static	void		  term_free(struct termp *);
-static	void		  term_pescape(struct termp *, const char **);
-static	void		  term_nescape(struct termp *,
+
+static	void		  do_escaped(struct termp *, const char **);
+static	void		  do_special(struct termp *,
+				const char *, size_t);
+static	void		  do_reserved(struct termp *,
 				const char *, size_t);
-static	void		  term_chara(struct termp *, char);
-static	void		  term_encodea(struct termp *, char);
-static	int		  term_isopendelim(const char *);
-static	int		  term_isclosedelim(const char *);
+static	void		  buffer(struct termp *, char);
+static	void		  encode(struct termp *, char);
+static	int		  isopendelim(const char *);
+static	int		  isclosedelim(const char *);
 
 
 void *
@@ -110,7 +113,7 @@ term_alloc(enum termenc enc)
 
 
 static int
-term_isclosedelim(const char *p)
+isclosedelim(const char *p)
 {
 
 	if ( ! (*p && 0 == *(p + 1)))
@@ -144,7 +147,7 @@ term_isclosedelim(const char *p)
 
 
 static int
-term_isopendelim(const char *p)
+isopendelim(const char *p)
 {
 
 	if ( ! (*p && 0 == *(p + 1)))
@@ -205,12 +208,6 @@ term_isopendelim(const char *p)
  *  Otherwise, the line will break at the right margin.  Extremely long
  *  lines will cause the system to emit a warning (TODO: hyphenate, if
  *  possible).
- *
- *  FIXME: newline breaks occur (in groff) also occur when a single
- *  space follows a NOBREAK (try `Bl -tag')
- *
- *  FIXME: there's a newline error where a `Bl -diag' will have a
- *  trailing newline if the line is exactly 73 chars long.
  */
 void
 term_flushln(struct termp *p)
@@ -385,13 +382,8 @@ term_vspace(struct termp *p)
 }
 
 
-/*
- * Determine the symbol indicated by an escape sequences, that is, one
- * starting with a backslash.  Once done, we pass this value into the
- * output buffer by way of the symbol table.
- */
 static void
-term_nescape(struct termp *p, const char *word, size_t len)
+do_special(struct termp *p, const char *word, size_t len)
 {
 	const char	*rhs;
 	size_t		 sz;
@@ -399,9 +391,40 @@ term_nescape(struct termp *p, const char *word, size_t len)
 
 	rhs = term_a2ascii(p->symtab, word, len, &sz);
 
-	if (rhs)
-		for (i = 0; i < (int)sz; i++) 
-			term_encodea(p, rhs[i]);
+	if (NULL == rhs) {
+#if 0
+		fputs("Unknown special character: ", stderr);
+		for (i = 0; i < (int)len; i++)
+			fputc(word[i], stderr);
+		fputc('\n', stderr);
+#endif
+		return;
+	}
+	for (i = 0; i < (int)sz; i++) 
+		encode(p, rhs[i]);
+}
+
+
+static void
+do_reserved(struct termp *p, const char *word, size_t len)
+{
+	const char	*rhs;
+	size_t		 sz;
+	int		 i;
+
+	rhs = term_a2res(p->symtab, word, len, &sz);
+
+	if (NULL == rhs) {
+#if 0
+		fputs("Unknown reserved word: ", stderr);
+		for (i = 0; i < (int)len; i++)
+			fputc(word[i], stderr);
+		fputc('\n', stderr);
+#endif
+		return;
+	}
+	for (i = 0; i < (int)sz; i++) 
+		encode(p, rhs[i]);
 }
 
 
@@ -411,12 +434,13 @@ term_nescape(struct termp *p, const char *word, size_t len)
  * the escape sequence (we assert upon badly-formed escape sequences).
  */
 static void
-term_pescape(struct termp *p, const char **word)
+do_escaped(struct termp *p, const char **word)
 {
-	int		 j;
+	int		 j, type;
 	const char	*wp;
 
 	wp = *word;
+	type = 1;
 
 	if (0 == *(++wp)) {
 		*word = wp;
@@ -430,7 +454,7 @@ term_pescape(struct termp *p, const char **word)
 			return;
 		}
 
-		term_nescape(p, wp, 2);
+		do_special(p, wp, 2);
 		*word = ++wp;
 		return;
 
@@ -448,13 +472,14 @@ term_pescape(struct termp *p, const char **word)
 				return;
 			}
 
-			term_nescape(p, wp, 2);
+			do_reserved(p, wp, 2);
 			*word = ++wp;
 			return;
 		case ('['):
+			type = 0;
 			break;
 		default:
-			term_nescape(p, wp, 1);
+			do_reserved(p, wp, 1);
 			*word = wp;
 			return;
 		}
@@ -485,7 +510,7 @@ term_pescape(struct termp *p, const char **word)
 		return;
 
 	} else if ('[' != *wp) {
-		term_nescape(p, wp, 1);
+		do_special(p, wp, 1);
 		*word = wp;
 		return;
 	}
@@ -499,7 +524,10 @@ term_pescape(struct termp *p, const char **word)
 		return;
 	}
 
-	term_nescape(p, wp - j, (size_t)j);
+	if (type)
+		do_special(p, wp - j, (size_t)j);
+	else
+		do_reserved(p, wp - j, (size_t)j);
 	*word = wp;
 }
 
@@ -514,28 +542,23 @@ term_word(struct termp *p, const char *word)
 {
 	const char	 *sv;
 
-	if (term_isclosedelim(word))
+	if (isclosedelim(word))
 		if ( ! (TERMP_IGNDELIM & p->flags))
 			p->flags |= TERMP_NOSPACE;
 
 	if ( ! (TERMP_NOSPACE & p->flags))
-		term_chara(p, ' ');
+		buffer(p, ' ');
 
 	if ( ! (p->flags & TERMP_NONOSPACE))
 		p->flags &= ~TERMP_NOSPACE;
 
-	/* 
-	 * If ANSI (word-length styling), then apply our style now,
-	 * before the word.
-	 */
-
 	for (sv = word; *word; word++)
 		if ('\\' != *word)
-			term_encodea(p, *word);
+			encode(p, *word);
 		else
-			term_pescape(p, &word);
+			do_escaped(p, &word);
 
-	if (term_isopendelim(sv))
+	if (isopendelim(sv))
 		p->flags |= TERMP_NOSPACE;
 }
 
@@ -546,7 +569,7 @@ term_word(struct termp *p, const char *word)
  * size.
  */
 static void
-term_chara(struct termp *p, char c)
+buffer(struct termp *p, char c)
 {
 	size_t		 s;
 
@@ -564,18 +587,18 @@ term_chara(struct termp *p, char c)
 
 
 static void
-term_encodea(struct termp *p, char c)
+encode(struct termp *p, char c)
 {
 	
 	if (' ' != c && TERMP_STYLE & p->flags) {
 		if (TERMP_BOLD & p->flags) {
-			term_chara(p, c);
-			term_chara(p, 8);
+			buffer(p, c);
+			buffer(p, 8);
 		}
 		if (TERMP_UNDER & p->flags) {
-			term_chara(p, '_');
-			term_chara(p, 8);
+			buffer(p, '_');
+			buffer(p, 8);
 		}
 	}
-	term_chara(p, c);
+	buffer(p, c);
 }
diff --git a/usr.bin/mandoc/term.h b/usr.bin/mandoc/term.h
index 9812ed1b383..3b8d7c3947e 100644
--- a/usr.bin/mandoc/term.h
+++ b/usr.bin/mandoc/term.h
@@ -1,4 +1,4 @@
-/*	$Id: term.h,v 1.7 2009/08/09 21:59:41 schwarze Exp $ */
+/*	$Id: term.h,v 1.8 2009/08/22 17:04:48 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -58,6 +58,7 @@ struct	termp {
 
 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 *);