Correct handling of \*(xx, \*[xxx], \*x versus \x, \(xx, \([xxx]: predefined strings and escape characters, respectively.

5 files changed, 305 insertions, 250 deletions

diff --git a/ascii.c b/ascii.c
index 00ca4d39..60698339 100644
--- a/ascii.c
+++ b/ascii.c
@@ -1,4 +1,4 @@
-/*	$Id: ascii.c,v 1.8 2009/06/10 20:18:43 kristaps Exp $ */
+/*	$Id: ascii.c,v 1.9 2009/07/27 12:02:49 kristaps 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/ascii.in b/ascii.in
index d925587d..b1dc0168 100644
--- a/ascii.in
+++ b/ascii.in
@@ -1,4 +1,4 @@
-/*	$Id: ascii.in,v 1.12 2009/07/26 19:53:40 kristaps Exp $ */
+/*	$Id: ascii.in,v 1.13 2009/07/27 12:02:49 kristaps Exp $ */
 /*
  * Copyright (c) 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -16,205 +16,227 @@
  */
 
 /*
- * 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,	  "", 0)
-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("left-parenthesis",	16,	"(",		1)
+STRING("right-parenthesis",	17,	")",		1)
+STRING("lp",			2,	"(",		1)
+STRING("rp",			2,	")",		1)
+STRING("left-bracket",		12,	"[",		1)
+STRING("right-bracket",		13,	"]",		1)
+STRING("left-singlequote",	16,	"`",		1)
+STRING("right-singlequote",	17,	"\'",		1)
+STRING("quote-left",		10,	"`",		1)
+STRING("quote-right",		11,	"\'",		1)
+STRING("q",			1,	"\"",		1)
+STRING("Ne",			2,	"!=",		2)
+STRING("Le",			2,	"<=",		2)
+STRING("Ge",			2,	">=",		2)
+STRING("Lt",			2,	"<",		1)
+STRING("Gt",			2,	">",		1)
+STRING("Pm",			2,	"+-",		2)
+STRING("Na",			2,	"NaN",		3)
+STRING("Ba",			2,	"|",		1)
+STRING("Am",			2,	"&",		1)
+STRING("Rq",			2,	"\'\'",		2)
+STRING("Lq",			2,	"``",		2)
+STRING("Pi",			2,	"pi",		2)
+STRING("If",			2,	"infinity",	8)
+
+BOTH("aa",			2,	"\'",		1)
+BOTH("ga",			2,	"`",		1)
+BOTH("ua",			2,	"^",		1)
+BOTH("<=",			2,	"<=",		2)
+BOTH(">=",			2,	">=",		2)
+
+CHAR("\\",			1,	"\\",		1)
+CHAR("\'",			1,	"\'",		1)
+CHAR("`",			1,	"`",		1)
+CHAR("%",			1,	"",		0)
+CHAR("-",			1,	"-",		1)
+CHAR(" ",			1,	" ",		1)
+CHAR("~",			1,	" ",		1)
+CHAR("^",			1,	"",		0)
+CHAR("0",			1,	" ",		1)
+CHAR(".",			1,	".",		1)
+CHAR("&",			1,	"",		0)
+CHAR("e",			1,	"\\",		1)
+CHAR("|",			1,	"",		0)
+CHAR("rC",			2, 	"}",		1)
+CHAR("lC",			2, 	"{",		1)
+CHAR("rB",			2, 	"]",		1)
+CHAR("lB",			2, 	"[",		1)
+CHAR("ra",			2, 	">",		1)
+CHAR("la",			2, 	"<",		1)
+CHAR("lq",			2, 	"``",		2)
+CHAR("rq",			2, 	"\'\'",		2)
+CHAR("oq",			2, 	"`",		1)
+CHAR("aq",			2, 	"\'",		1)
+CHAR("Bq",			2, 	",,",		2)
+CHAR("bq",			2, 	",,",		2)
+CHAR("<-",			2, 	"<-",		2)
+CHAR("->",			2, 	"->",		2)
+CHAR("<>",			2, 	"<>",		2)
+CHAR("ua",			2, 	"^",		1)
+CHAR("da",			2, 	"v",		1)
+CHAR("bu",			2, 	"o",		1)
+CHAR("ci",			2, 	"O",		1)
+CHAR("ba",			2, 	"|",		1)
+CHAR("bb",			2, 	"|",		1)
+CHAR("co",			2, 	"(C)",		3)
+CHAR("rg",			2, 	"(R)",		3)
+CHAR("tm",			2, 	"tm",		2)
+CHAR("<=",			2, 	"<=",		2)
+CHAR(">=",			2, 	">=",		2)
+CHAR("==",			2, 	"==",		2)
+CHAR("!=",			2, 	"!=",		2)
+CHAR("+-",			2, 	"+-",		2)
+CHAR("if",			2, 	"oo",		2)
+CHAR("na",			2, 	"NaN",		3)
+CHAR("**",			2, 	"*",		1)
+CHAR("aa",			2, 	"\'",		1)
+CHAR("a~",			2, 	"~",		1)
+CHAR("ga",			2, 	"`",		1)
+CHAR("en",			2, 	"-",		1)
+CHAR("em",			2, 	"--",		2)
+CHAR("hy",			2, 	"-",		1)
+CHAR("Fo",			2,	"<<",		2)
+CHAR("Fc",			2,	">>",		2)
+CHAR("fo",			2,	"<",		1)
+CHAR("fc",			2,	">",		1)
+CHAR("lh",			2,	"<=",		2)
+CHAR("rh",			2,	"=>",		2)
+CHAR("ae",			2,	"ae",		2)
+CHAR("AE",			2,	"AE",		2)
+CHAR("oe",			2,	"oe",		2)
+CHAR("OE",			2,	"OE",		2)
+CHAR("ss",			2,	"ss",		2)
+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,	"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,	"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("lA",			2, 	"<=",		2)
+CHAR("rA",			2, 	"=>",		2)
+CHAR("uA",			2, 	"^",		1)
+CHAR("dA",			2, 	"v",		1)
+CHAR("hA",			2, 	"<=>",		3)
+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("^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("-D",			2,	"D",		1)
+CHAR("Sd",			2,	"o",		1)
+CHAR("TP",			2,	"b",		1)
+CHAR("Tp",			2,	"b",		1)
+CHAR(",C",			2,	"C",		1)
+CHAR(",c",			2,	"c",		1)
+CHAR("/L",			2,	"L",		1)
+CHAR("/l",			2,	"l",		1)
+CHAR("/O",			2,	"O",		1)
+CHAR("/o",			2,	"o",		1)
+CHAR("oA",			2,	"A",		1)
+CHAR("oa",			2,	"a",		1)
+CHAR("a^",			2,	"^",		1)
+CHAR("ac",			2,	",",		1)
+CHAR("ad",			2,	"\"",		1)
+CHAR("ah",			2,	"v",		1)
+CHAR("ao",			2,	"o",		1)
+CHAR("ho",			2,	",",		1)
+CHAR("ab",			2,	"`",		1)
+CHAR("a\"",			2,	"\"",		1)
+CHAR("a-",			2,	"-",		1)
+CHAR("Cs",			2,	"x",		1)
+CHAR("Do",			2,	"$",		1)
+CHAR("Po",			2,	"L",		1)
+CHAR("Ye",			2,	"Y",		1)
+CHAR("Fn",			2,	"f",		1)
+CHAR("ct",			2,	"c",		1)
+CHAR("ff",			2,	"ff",		2)
+CHAR("fi",			2,	"fi",		2)
+CHAR("fl",			2,	"fl",		2)
+CHAR("Fi",			2,	"ffi",		3)
+CHAR("Fl",			2,	"ffl",		3)
+CHAR("r!",			2,	"i",		1)
+CHAR("r?",			2,	"c",		1)
+CHAR("dd",			2,	"=",		1)
+CHAR("dg",			2,	"-",		1)
+CHAR("ps",			2,	"9|",		2)
+CHAR("sc",			2,	"S",		1)
+CHAR("de",			2,	"o",		1)
+CHAR("tf",			2,	".:.",		3)
+CHAR("~~",			2,	"~~",		2)
+CHAR("~=",			2,	"~=",		2)
+CHAR("=~",			2,	"=~",		2)
+CHAR("AN",			2,	"^",		1)
+CHAR("OR",			2,	"v",		1)
+CHAR("no",			2,	"~",		1)
+CHAR("fa",			2,	"V",		1)
+CHAR("te",			2,	"3",		1)
+CHAR("Ah",			2,	"N",		1)
+CHAR("Im",			2,	"I",		1)
+CHAR("Re",			2,	"R",		1)
+CHAR("mo",			2,	"E",		1)
+CHAR("nm",			2,	"E",		1)
+CHAR("eq",			2,	"=",		1)
+CHAR("pl",			2,	"+",		1)
+CHAR("di",			2,	"-:-",		3)
+CHAR("mu",			2,	"x",		1)
+CHAR("(=",			2,	"(=",		2)
+CHAR("=)",			2,	"=)",		2)
+CHAR("ap",			2,	"~",		1)
+CHAR("pd",			2,	"a",		1)
+CHAR("gr",			2,	"V",		1)
+CHAR("ca",			2,	"(^)",		3)
+CHAR("cu",			2,	"U",		1)
+CHAR("es",			2,	"{}",		2)
+CHAR("st",			2,	"-)",		2)
diff --git a/mandoc_char.7 b/mandoc_char.7
index 1d90fc69..f4ce7041 100644
--- a/mandoc_char.7
+++ b/mandoc_char.7
@@ -1,4 +1,4 @@
-.\"	$Id: mandoc_char.7,v 1.8 2009/07/26 19:53:40 kristaps Exp $
+.\"	$Id: mandoc_char.7,v 1.9 2009/07/27 12:02:49 kristaps Exp $
 .\"
 .\" Copyright (c) 2009 Kristaps Dzonsons <kristaps@kth.se>
 .\"
@@ -14,7 +14,7 @@
 .\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
 .\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 .\" 
-.Dd $Mdocdate: July 26 2009 $
+.Dd $Mdocdate: July 27 2009 $
 .Dt MANDOC_CHAR 7
 .Os
 .\" SECTION
@@ -39,8 +39,18 @@ Both
 .Xr mdoc 7
 and
 .Xr man 7
-encode special characters with slightly different semantics; consult the
-respective manuals for these escapes.
+encode these special characters with 
+.Sq \eX
+.Pq for a one-character escape ,
+.Sq \e(XX
+.Pq two-character ,
+and
+.Sq \e[N]
+.Pq N-character .
+One may generalise
+.Sq \e(XX
+as
+.Sq \e[XX] .
 .\" PARAGRAPH
 .Pp
 Typographic:
@@ -51,10 +61,12 @@ Typographic:
 .Pq space
 .It \e^
 .Pq zero-width space
-.It \e^
+.It \e%
 .Pq zero-width space
 .It \e&
 .Pq zero-width space
+.It \e|
+.Pq zero-width space
 .El
 .\" PARAGRAPH
 .Pp
@@ -111,12 +123,8 @@ Enclosures:
 .Pq double-quote
 .It \e(lq
 .Pq left double-quote
-.It \e(Lq
-.Pq left double-quote, deprecated
 .It \e(rq
 .Pq right double-quote
-.It \e(Rq
-.Pq right double-quote, deprecated
 .It \e(oq
 .Pq left single-quote
 .It \e(aq
@@ -211,36 +219,20 @@ Mathematical:
 .Pq approximately equals
 .It \e(=~
 .Pq congruent
-.It \e(Gt
-.Pq greater-than, deprecated
-.It \e(Lt
-.Pq less-than, deprecated
 .It \e(<=
 .Pq less-than-equal
-.It \e(Le
-.Pq less-than-equal, deprecated
 .It \e(>=
 .Pq greater-than-equal
-.It \e(Ge
-.Pq greater-than-equal
 .It \e(==
 .Pq equal
 .It \e(!=
 .Pq not equal
-.It \e(Ne
-.Pq not equal, deprecated
 .It \e(if
 .Pq infinity
-.It \e(If
-.Pq infinity, deprecated
 .It \e(na
 .Pq NaN , an extension
-.It \e(Na
-.Pq NaN, deprecated
 .It \e(+-
 .Pq plus-minus
-.It \e(Pm
-.Pq plus-minus, deprecated
 .It \e(**
 .Pq asterisk
 .El
@@ -457,20 +449,14 @@ Special symbols:
 .Pq bar
 .It \e(bb
 .Pq broken bar
-.It \e(Ba
-.Pq bar, deprecated
 .It \e(co
 .Pq copyright
 .It \e(rg
 .Pq registered
 .It \e(tm
 .Pq trademarked
-.It \e&
-.Pq non-breaking space
 .It \ee
 .Pq escape
-.It \e(Am
-.Pq ampersand, deprecated
 .El 
 .\" SECTION
 .Sh COMPATIBILITY
diff --git a/term.c b/term.c
index 89a4a34f..fa9d8d68 100644
--- a/term.c
+++ b/term.c
@@ -1,4 +1,4 @@
-/*	$Id: term.c,v 1.93 2009/07/24 11:54:25 kristaps Exp $ */
+/*	$Id: term.c,v 1.94 2009/07/27 12:02:49 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -34,6 +34,8 @@ static	void		  term_free(struct termp *);
 static	void		  term_pescape(struct termp *, const char **);
 static	void		  term_nescape(struct termp *,
 				const char *, size_t);
+static	void		  term_sescape(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 *);
@@ -399,9 +401,26 @@ 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)
+		return;
+	for (i = 0; i < (int)sz; i++) 
+		term_encodea(p, rhs[i]);
+}
+
+
+static void
+term_sescape(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)
+		return;
+	for (i = 0; i < (int)sz; i++) 
+		term_encodea(p, rhs[i]);
 }
 
 
@@ -448,13 +467,13 @@ term_pescape(struct termp *p, const char **word)
 				return;
 			}
 
-			term_nescape(p, wp, 2);
+			term_sescape(p, wp, 2);
 			*word = ++wp;
 			return;
 		case ('['):
 			break;
 		default:
-			term_nescape(p, wp, 1);
+			term_sescape(p, wp, 1);
 			*word = wp;
 			return;
 		}
diff --git a/term.h b/term.h
index 29f1afd2..9b3a33b8 100644
--- a/term.h
+++ b/term.h
@@ -1,4 +1,4 @@
-/*	$Id: term.h,v 1.42 2009/07/24 12:47:35 kristaps Exp $ */
+/*	$Id: term.h,v 1.43 2009/07/27 12:02:49 kristaps 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 *);