Parse and ignore the \k, \o, \w, and \z roff escapes, and recursively

ignore embedded escapes and mathematical roff subexpressions.
In roff copy mode, resolve "\\" to '\'.
Allow ".xx\}" where xx is a macro to close roff conditional scope.
Mandoc now handles the special character definitions in the pod2man(1)
preamble, so remove the explicit redefinitions in chars.c/chars.in.
From kristaps@.

I have checked that this causes no relevant change to the Perl manuals.
The only change introduced is that some non-ASCII characters rendered
incorrectly before are now rendered incorrectly in a different way.
For example, e accent aigu was "e", now is "e'"
and c cedille was "c", now is "c,".

5 files changed, 117 insertions, 26 deletions

diff --git a/usr.bin/mandoc/chars.c b/usr.bin/mandoc/chars.c
index 4b5f93c07f8..120a66140b0 100644
--- a/usr.bin/mandoc/chars.c
+++ b/usr.bin/mandoc/chars.c
@@ -1,4 +1,4 @@
-/*	$Id: chars.c,v 1.12 2010/08/20 00:53:35 schwarze Exp $ */
+/*	$Id: chars.c,v 1.13 2010/09/13 22:04:01 schwarze Exp $ */
 /*
  * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -36,7 +36,7 @@ struct	ln {
 #define CHARS_BOTH	 (CHARS_CHAR | CHARS_STRING)
 };
 
-#define	LINES_MAX	  369
+#define	LINES_MAX	  362
 
 #define CHAR(in, ch, code) \
 	{ NULL, (in), (ch), (code), CHARS_CHAR },
diff --git a/usr.bin/mandoc/chars.in b/usr.bin/mandoc/chars.in
index 7df6cbd6dde..4c48e3d3209 100644
--- a/usr.bin/mandoc/chars.in
+++ b/usr.bin/mandoc/chars.in
@@ -1,4 +1,4 @@
-/*	$Id: chars.in,v 1.11 2010/08/18 02:46:37 schwarze Exp $ */
+/*	$Id: chars.in,v 1.12 2010/09/13 22:04:01 schwarze Exp $ */
 /*
  * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -322,8 +322,8 @@ CHAR("fi",			"fi",		64257)
 CHAR("fl",			"fl",		64258)
 CHAR("Fi",			"ffi",		64259)
 CHAR("Fl",			"ffl",		64260)
-BOTH("AE",			"AE",		198)
-BOTH("ae",			"ae",		230)
+CHAR("AE",			"AE",		198)
+CHAR("ae",			"ae",		230)
 CHAR("OE",			"OE",		338)
 CHAR("oe",			"oe",		339)
 CHAR("ss",			"ss",		223)
@@ -360,13 +360,6 @@ STRING("^",			"^",		94)
 STRING(",",			",",		44)
 STRING("~",			"~",		126)
 STRING("/",			"/",		47)
-STRING(":",			"\"",		776)
-STRING("8",			"B",		946)
-STRING("o",			"o",		176)
-STRING("D-",			"D",		208)
-STRING("d-",			"o",		240)
-STRING("Th",			"b",		222)
-STRING("th",			"b",		254)
 
 /* Old style. */
 STRING("Am",			"&",		38)
diff --git a/usr.bin/mandoc/mandoc.c b/usr.bin/mandoc/mandoc.c
index 9abaced9a54..8d0421f11bd 100644
--- a/usr.bin/mandoc/mandoc.c
+++ b/usr.bin/mandoc/mandoc.c
@@ -1,4 +1,4 @@
-/*	$Id: mandoc.c,v 1.18 2010/08/20 00:53:35 schwarze Exp $ */
+/*	$Id: mandoc.c,v 1.19 2010/09/13 22:04:01 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -51,14 +51,10 @@ mandoc_special(char *p)
 		/* FALLTHROUGH */
 	case ('x'):
 		/* FALLTHROUGH */
-	case ('w'):
-		/* FALLTHROUGH */
 	case ('S'):
 		/* FALLTHROUGH */
 	case ('R'):
 		/* FALLTHROUGH */
-	case ('o'):
-		/* FALLTHROUGH */
 	case ('N'):
 		/* FALLTHROUGH */
 	case ('l'):
@@ -126,6 +122,27 @@ mandoc_special(char *p)
 			p++;
 		} 
 		
+		/* Handle embedded numerical subexp or escape. */
+
+		if ('(' == *p) {
+			while (*p && ')' != *p)
+				if ('\\' == *p++) {
+					i = mandoc_special(--p);
+					if (0 == i)
+						return(0);
+					p += i;
+				}
+
+			if (')' == *p++)
+				break;
+
+			return(0);
+		} else if ('\\' == *p) {
+			if (0 == (i = mandoc_special(p)))
+				return(0);
+			p += i;
+		}
+
 		break;
 #if 0
 	case ('Y'):
@@ -136,9 +153,9 @@ mandoc_special(char *p)
 		/* FALLTHROUGH */
 	case ('n'):
 		/* FALLTHROUGH */
+#endif
 	case ('k'):
 		/* FALLTHROUGH */
-#endif
 	case ('M'):
 		/* FALLTHROUGH */
 	case ('m'):
@@ -167,6 +184,23 @@ mandoc_special(char *p)
 	case ('['):
 		term = ']';
 		break;
+	case ('z'):
+		len = 1;
+		if ('\\' == *p) {
+			if (0 == (i = mandoc_special(p)))
+				return(0);
+			p += i;
+			return(*p ? (int)(p - sv) : 0);
+		}
+		break;
+	case ('o'):
+		/* FALLTHROUGH */
+	case ('w'):
+		if ('\'' == *p++) {
+			term = '\'';
+			break;
+		}
+		/* FALLTHROUGH */
 	default:
 		len = 1;
 		p--;
diff --git a/usr.bin/mandoc/out.c b/usr.bin/mandoc/out.c
index 4da5bb43634..6e1f812fa29 100644
--- a/usr.bin/mandoc/out.c
+++ b/usr.bin/mandoc/out.c
@@ -1,4 +1,4 @@
-/*	$Id: out.c,v 1.7 2010/08/18 02:38:40 schwarze Exp $ */
+/*	$Id: out.c,v 1.8 2010/09/13 22:04:01 schwarze Exp $ */
 /*
  * Copyright (c) 2009, 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  *
@@ -168,6 +168,7 @@ a2roffdeco(enum roffdeco *d, const char **word, size_t *sz)
 	int		 i, j, lim;
 	char		 term, c;
 	const char	*wp;
+	enum roffdeco	 dd;
 
 	*d = DECO_NONE;
 	lim = i = 0;
@@ -215,6 +216,8 @@ a2roffdeco(enum roffdeco *d, const char **word, size_t *sz)
 			break;
 		}
 		break;
+	case ('k'):
+		/* FALLTHROUGH */
 	case ('M'):
 		/* FALLTHROUGH */
 	case ('m'):
@@ -271,7 +274,27 @@ a2roffdeco(enum roffdeco *d, const char **word, size_t *sz)
 				return(i);
 			i++;
 		} 
-		
+
+		/* Handle embedded numerical subexp or escape. */
+
+		if ('(' == wp[i]) {
+			while (wp[i] && ')' != wp[i])
+				if ('\\' == wp[i++]) {
+					/* Handle embedded escape. */
+					*word = &wp[i];
+					i += a2roffdeco(&dd, word, sz);
+				}
+
+			if (')' == wp[i++])
+				break;
+
+			*d = DECO_NONE;
+			return(i - 1);
+		} else if ('\\' == wp[i]) {
+			*word = &wp[++i];
+			i += a2roffdeco(&dd, word, sz);
+		}
+
 		break;
 	case ('['):
 		*d = DECO_SPECIAL;
@@ -280,6 +303,22 @@ a2roffdeco(enum roffdeco *d, const char **word, size_t *sz)
 	case ('c'):
 		*d = DECO_NOSPACE;
 		return(i);
+	case ('z'):
+		*d = DECO_NONE;
+		if ('\\' == wp[i]) {
+			*word = &wp[++i];
+			return(i + a2roffdeco(&dd, word, sz));
+		} else
+			lim = 1;
+		break;
+	case ('o'):
+		/* FALLTHROUGH */
+	case ('w'):
+		if ('\'' == wp[i++]) {
+			term = '\'';
+			break;
+		} 
+		/* FALLTHROUGH */
 	default:
 		*d = DECO_SSPECIAL;
 		i--;
diff --git a/usr.bin/mandoc/roff.c b/usr.bin/mandoc/roff.c
index f2a84ddf13f..af5eb60ce5f 100644
--- a/usr.bin/mandoc/roff.c
+++ b/usr.bin/mandoc/roff.c
@@ -1,4 +1,4 @@
-/*	$Id: roff.c,v 1.11 2010/08/20 00:53:35 schwarze Exp $ */
+/*	$Id: roff.c,v 1.12 2010/09/13 22:04:01 schwarze Exp $ */
 /*
  * Copyright (c) 2010 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2010 Ingo Schwarze <schwarze@openbsd.org>
@@ -128,6 +128,7 @@ static	int		 roff_res(struct roff *,
 				char **, size_t *, int);
 static	void		 roff_setstr(struct roff *,
 				const char *, const char *);
+static	char		*roff_strdup(const char *);
 
 /* See roff_hash_find() */
 
@@ -744,11 +745,13 @@ roff_cond_sub(ROFF_ARGS)
 	l = r->last;
 	roffnode_cleanscope(r);
 
-	if (l != r->last)
-		return(ROFFRULE_DENY == rr ? ROFF_IGN : ROFF_CONT);
-
-	if (ROFF_MAX == (t = roff_parse(*bufp, &pos)))
+	if (ROFF_MAX == (t = roff_parse(*bufp, &pos))) {
+		if ('\\' == (*bufp)[pos] && '}' == (*bufp)[pos + 1])
+			return(roff_ccond
+				(r, ROFF_ccond, bufp, szp,
+				 ln, pos, pos + 2, offs));
 		return(ROFFRULE_DENY == rr ? ROFF_IGN : ROFF_CONT);
+	}
 
 	/*
 	 * A denied conditional must evaluate its children if and only
@@ -1005,6 +1008,27 @@ roff_nr(ROFF_ARGS)
 }
 
 
+static char *
+roff_strdup(const char *name)
+{
+	char		*namecopy, *sv;
+
+	/* 
+	 * This isn't a nice simple mandoc_strdup() because we must
+	 * handle roff's stupid double-escape rule. 
+	 */
+	sv = namecopy = mandoc_malloc(strlen(name) + 1);
+	while (*name) {
+		if ('\\' == *name && '\\' == *(name + 1))
+			name++;
+		*namecopy++ = *name++;
+	}
+
+	*namecopy = '\0';
+	return(sv);
+}
+
+
 static void
 roff_setstr(struct roff *r, const char *name, const char *string)
 {
@@ -1024,7 +1048,8 @@ roff_setstr(struct roff *r, const char *name, const char *string)
 	} else
 		free(n->string);
 
-	n->string = string ? strdup(string) : NULL;
+	/* Don't use mandoc_strdup: clean out double-escapes. */
+	n->string = string ? roff_strdup(string) : NULL;
 }