===================================================================
RCS file: /cvs/mandoc/mdoc.c,v
retrieving revision 1.124
retrieving revision 1.135
diff -u -p -r1.124 -r1.135
--- mandoc/mdoc.c	2010/05/07 15:49:36	1.124
+++ mandoc/mdoc.c	2010/05/16 10:59:36	1.135
@@ -1,4 +1,4 @@
-/*	$Id: mdoc.c,v 1.124 2010/05/07 15:49:36 kristaps Exp $ */
+/*	$Id: mdoc.c,v 1.135 2010/05/16 10:59:36 kristaps Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -69,7 +69,6 @@ const	char *const __mdoc_merrnames[MERRMAX] = {		 
 	"prologue macro out of conventional order", /* EPROLOOO */
 	"prologue macro repeated", /* EPROLREP */
 	"invalid manual section", /* EBADMSEC */
-	"invalid section", /* EBADSEC */
 	"invalid font mode", /* EFONT */
 	"invalid date syntax", /* EBADDATE */
 	"invalid number format", /* ENUMFMT */
@@ -151,9 +150,10 @@ static	struct mdoc_node *node_alloc(struct mdoc *, int
 				enum mdoct, enum mdoc_type);
 static	int		  node_append(struct mdoc *, 
 				struct mdoc_node *);
-static	int		  mdoc_ptext(struct mdoc *, int, char *);
-static	int		  mdoc_pmacro(struct mdoc *, int, char *);
-static	int		  macrowarn(struct mdoc *, int, const char *);
+static	int		  mdoc_ptext(struct mdoc *, int, char *, int);
+static	int		  mdoc_pmacro(struct mdoc *, int, char *, int);
+static	int		  macrowarn(struct mdoc *, int, 
+				const char *, int);
 
 
 const struct mdoc_node *
@@ -191,6 +191,8 @@ mdoc_free1(struct mdoc *mdoc)
 		free(mdoc->meta.arch);
 	if (mdoc->meta.vol)
 		free(mdoc->meta.vol);
+	if (mdoc->meta.msec)
+		free(mdoc->meta.msec);
 }
 
 
@@ -283,14 +285,16 @@ mdoc_endparse(struct mdoc *m)
  * the macro (mdoc_pmacro()) or text parser (mdoc_ptext()).
  */
 int
-mdoc_parseln(struct mdoc *m, int ln, char *buf)
+mdoc_parseln(struct mdoc *m, int ln, char *buf, int offs)
 {
 
 	if (MDOC_HALT & m->flags)
 		return(0);
 
-	return('.' == *buf ? mdoc_pmacro(m, ln, buf) :
-			mdoc_ptext(m, ln, buf));
+	m->flags |= MDOC_NEWLINE;
+	return(('.' == buf[offs] || '\'' == buf[offs]) ? 
+			mdoc_pmacro(m, ln, buf, offs) :
+			mdoc_ptext(m, ln, buf, offs));
 }
 
 
@@ -453,7 +457,9 @@ node_alloc(struct mdoc *m, int line, int pos, 
 	p->pos = pos;
 	p->tok = tok;
 	p->type = type;
-
+	if (MDOC_NEWLINE & m->flags)
+		p->flags |= MDOC_LINE;
+	m->flags &= ~MDOC_NEWLINE;
 	return(p);
 }
 
@@ -542,7 +548,7 @@ mdoc_word_alloc(struct mdoc *m, int line, int pos, con
 
 	len = strlen(p);
 
-	n = node_alloc(m, line, pos, -1, MDOC_TEXT);
+	n = node_alloc(m, line, pos, MDOC_MAX, MDOC_TEXT);
 	n->string = mandoc_malloc(len + 1);
 	sv = strlcpy(n->string, p, len + 1);
 
@@ -625,32 +631,34 @@ mdoc_node_delete(struct mdoc *m, struct mdoc_node *p)
  * control character.
  */
 static int
-mdoc_ptext(struct mdoc *m, int line, char *buf)
+mdoc_ptext(struct mdoc *m, int line, char *buf, int offs)
 {
 	int		 i;
 
 	/* Ignore bogus comments. */
 
-	if ('\\' == buf[0] && '.' == buf[1] && '\"' == buf[2])
-		return(mdoc_pwarn(m, line, 0, EBADCOMMENT));
+	if ('\\' == buf[offs] && 
+			'.' == buf[offs + 1] && 
+			'"' == buf[offs + 2])
+		return(mdoc_pwarn(m, line, offs, EBADCOMMENT));
 
 	/* No text before an initial macro. */
 
 	if (SEC_NONE == m->lastnamed)
-		return(mdoc_perr(m, line, 0, ETEXTPROL));
+		return(mdoc_perr(m, line, offs, ETEXTPROL));
 
 	/* Literal just gets pulled in as-is. */
 	
 	if (MDOC_LITERAL & m->flags)
-		return(mdoc_word_alloc(m, line, 0, buf));
+		return(mdoc_word_alloc(m, line, offs, buf + offs));
 
 	/* Check for a blank line, which may also consist of spaces. */
 
-	for (i = 0; ' ' == buf[i]; i++)
+	for (i = offs; ' ' == buf[i]; i++)
 		/* Skip to first non-space. */ ;
 
 	if ('\0' == buf[i]) {
-		if ( ! mdoc_pwarn(m, line, 0, ENOBLANK))
+		if ( ! mdoc_pwarn(m, line, offs, ENOBLANK))
 			return(0);
 
 		/*
@@ -658,38 +666,61 @@ mdoc_ptext(struct mdoc *m, int line, char *buf)
 		 * blank lines aren't allowed, but enough manuals assume this
 		 * behaviour that we want to work around it.
 		 */
-		if ( ! mdoc_elem_alloc(m, line, 0, MDOC_Pp, NULL))
+		if ( ! mdoc_elem_alloc(m, line, offs, MDOC_Pp, NULL))
 			return(0);
 
 		m->next = MDOC_NEXT_SIBLING;
 		return(1);
 	}
 
-	/* Warn if the last un-escaped character is whitespace. */
+	/* 
+	 * Warn if the last un-escaped character is whitespace. Then
+	 * strip away the remaining spaces (tabs stay!).   
+	 */
 
 	i = (int)strlen(buf);
 	assert(i);
 
-	if (' ' == buf[i - 1] || '\t' == buf[i - 1])
-		if (1 == i || ('\\' != buf[i - 2]))
+	if (' ' == buf[i - 1] || '\t' == buf[i - 1]) {
+		if (i > 1 && '\\' != buf[i - 2])
 			if ( ! mdoc_pwarn(m, line, i - 1, ETAILWS))
 				return(0);
 
+		for (--i; i && ' ' == buf[i]; i--)
+			/* Spin back to non-space. */ ;
+
+		/* Jump ahead of escaped whitespace. */
+		i += '\\' == buf[i] ? 2 : 1;
+
+		buf[i] = '\0';
+	}
+
 	/* Allocate the whole word. */
 
-	return(mdoc_word_alloc(m, line, 0, buf));
-}
+	if ( ! mdoc_word_alloc(m, line, offs, buf + offs))
+		return(0);
 
+	/*
+	 * End-of-sentence check.  If the last character is an unescaped
+	 * EOS character, then flag the node as being the end of a
+	 * sentence.  The front-end will know how to interpret this.
+	 */
 
+	assert(i);
+	if (mandoc_eos(buf, (size_t)i))
+		m->last->flags |= MDOC_EOS;
 
+	return(1);
+}
+
+
 static int
-macrowarn(struct mdoc *m, int ln, const char *buf)
+macrowarn(struct mdoc *m, int ln, const char *buf, int offs)
 {
 	if ( ! (MDOC_IGN_MACRO & m->pflags))
-		return(mdoc_verr(m, ln, 0, 
-				"unknown macro: %s%s", 
+		return(mdoc_verr(m, ln, offs, "unknown macro: %s%s", 
 				buf, strlen(buf) > 3 ? "..." : ""));
-	return(mdoc_vwarn(m, ln, 0, "unknown macro: %s%s",
+	return(mdoc_vwarn(m, ln, offs, "unknown macro: %s%s",
 				buf, strlen(buf) > 3 ? "..." : ""));
 }
 
@@ -699,17 +730,20 @@ macrowarn(struct mdoc *m, int ln, const char *buf)
  * character.
  */
 int
-mdoc_pmacro(struct mdoc *m, int ln, char *buf)
+mdoc_pmacro(struct mdoc *m, int ln, char *buf, int offs)
 {
-	int		  i, j, c;
-	char		  mac[5];
+	enum mdoct	tok;
+	int		i, j, sv;
+	char		mac[5];
 
 	/* Empty lines are ignored. */
 
-	if ('\0' == buf[1])
+	offs++;
+
+	if ('\0' == buf[offs])
 		return(1);
 
-	i = 1;
+	i = offs;
 
 	/* Accept whitespace after the initial control char. */
 
@@ -721,6 +755,8 @@ mdoc_pmacro(struct mdoc *m, int ln, char *buf)
 			return(1);
 	}
 
+	sv = i;
+
 	/* Copy the first word into a nil-terminated buffer. */
 
 	for (j = 0; j < 4; j++, i++) {
@@ -736,16 +772,16 @@ mdoc_pmacro(struct mdoc *m, int ln, char *buf)
 		return(mdoc_perr(m, ln, i, EPRINT));
 	}
 
-	mac[j] = 0;
+	mac[j] = '\0';
 
 	if (j == 4 || j < 2) {
-		if ( ! macrowarn(m, ln, mac))
+		if ( ! macrowarn(m, ln, mac, sv))
 			goto err;
 		return(1);
 	} 
 	
-	if (MDOC_MAX == (c = mdoc_hash_find(mac))) {
-		if ( ! macrowarn(m, ln, mac))
+	if (MDOC_MAX == (tok = mdoc_hash_find(mac))) {
+		if ( ! macrowarn(m, ln, mac, sv))
 			goto err;
 		return(1);
 	}
@@ -755,7 +791,10 @@ mdoc_pmacro(struct mdoc *m, int ln, char *buf)
 	while (buf[i] && ' ' == buf[i])
 		i++;
 
-	/* Trailing whitespace? */
+	/* 
+	 * Trailing whitespace.  Note that tabs are allowed to be passed
+	 * into the parser as "text", so we only warn about spaces here.
+	 */
 
 	if ('\0' == buf[i] && ' ' == buf[i - 1])
 		if ( ! mdoc_pwarn(m, ln, i - 1, ETAILWS))
@@ -765,7 +804,7 @@ mdoc_pmacro(struct mdoc *m, int ln, char *buf)
 	 * Begin recursive parse sequence.  Since we're at the start of
 	 * the line, we don't need to do callable/parseable checks.
 	 */
-	if ( ! mdoc_macro(m, c, ln, 1, &i, buf)) 
+	if ( ! mdoc_macro(m, tok, ln, sv, &i, buf)) 
 		goto err;
 
 	return(1);