more end-of-sentence (EOS) handling:

* recognize the end of quoted sentences, and of those in parantheses
* detect EOS in append_delims, so it works after all macros
by kristaps@

4 files changed, 73 insertions, 27 deletions

diff --git a/usr.bin/mandoc/man.7 b/usr.bin/mandoc/man.7
index df94dbf704b..5ac1fb1a56e 100644
--- a/usr.bin/mandoc/man.7
+++ b/usr.bin/mandoc/man.7
@@ -1,4 +1,4 @@
-.\"	$Id: man.7,v 1.23 2010/05/14 19:52:43 schwarze Exp $
+.\"	$Id: man.7,v 1.24 2010/05/15 15:37:53 schwarze Exp $
 .\"
 .\" Copyright (c) 2009 Kristaps Dzonsons <kristaps@bsd.lv>
 .\"
@@ -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: May 14 2010 $
+.Dd $Mdocdate: May 15 2010 $
 .Dt MAN 7
 .Os
 .Sh NAME
@@ -216,8 +216,13 @@ literal text.
 When composing a manual, make sure that your sentences end at the end of
 a line.
 By doing so, front-ends will be able to apply the proper amount of
-spacing after the end of sentence (unescaped) period, exclamation, or question
-mark.
+spacing after the end of sentence (unescaped) period, exclamation mark,
+or question mark followed by zero or more non-sentence closing
+delimiters (
+.Ns Sq \&) ,
+.Sq \&] ,
+.Sq \&' ,
+.Sq \&" ) .
 .Sh MANUAL STRUCTURE
 Each
 .Nm
diff --git a/usr.bin/mandoc/mandoc.c b/usr.bin/mandoc/mandoc.c
index 6b04f5c983d..92c65e9d2e1 100644
--- a/usr.bin/mandoc/mandoc.c
+++ b/usr.bin/mandoc/mandoc.c
@@ -1,4 +1,4 @@
-/*	$Id: mandoc.c,v 1.10 2010/05/15 09:20:01 schwarze Exp $ */
+/*	$Id: mandoc.c,v 1.11 2010/05/15 15:37:53 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -304,19 +304,35 @@ mandoc_eos(const char *p, size_t sz)
 	if (0 == sz)
 		return(0);
 
-	switch (p[(int)sz - 1]) {
-	case ('.'):
-		/* Escaped periods. */
-		if (sz > 1 && '\\' == p[(int)sz - 2])
+	/*
+	 * End-of-sentence recognition must include situations where
+	 * some symbols, such as `)', allow prior EOS punctuation to
+	 * propogate outward.
+	 */
+
+	for ( ; sz; sz--) {
+		switch (p[(int)sz - 1]) {
+		case ('\"'):
+			/* FALLTHROUGH */
+		case ('\''):
+			/* FALLTHROUGH */
+		case (']'):
+			/* FALLTHROUGH */
+		case (')'):
+			break;
+		case ('.'):
+			/* Escaped periods. */
+			if (sz > 1 && '\\' == p[(int)sz - 2])
+				return(0);
+			/* FALLTHROUGH */
+		case ('!'):
+			/* FALLTHROUGH */
+		case ('?'):
+			return(1);
+		default:
 			return(0);
-		/* FALLTHROUGH */
-	case ('!'):
-		/* FALLTHROUGH */
-	case ('?'):
-		break;
-	default:
-		return(0);
+		}
 	}
 
-	return(1);
+	return(0);
 }
diff --git a/usr.bin/mandoc/mdoc.7 b/usr.bin/mandoc/mdoc.7
index aecae284574..d9e3578c063 100644
--- a/usr.bin/mandoc/mdoc.7
+++ b/usr.bin/mandoc/mdoc.7
@@ -1,4 +1,4 @@
-.\"	$Id: mdoc.7,v 1.28 2010/05/14 19:52:43 schwarze Exp $
+.\"	$Id: mdoc.7,v 1.29 2010/05/15 15:37:53 schwarze Exp $
 .\"
 .\" Copyright (c) 2009 Kristaps Dzonsons <kristaps@bsd.lv>
 .\"
@@ -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: May 14 2010 $
+.Dd $Mdocdate: May 15 2010 $
 .Dt MDOC 7
 .Os
 .Sh NAME
@@ -301,7 +301,18 @@ When composing a manual, make sure that your sentences end at the end of
 a line.
 By doing so, front-ends will be able to apply the proper amount of
 spacing after the end of sentence (unescaped) period, exclamation mark,
-or question mark.
+or question mark followed by zero or more non-sentence closing
+delimiters (
+.Ns Sq \&) ,
+.Sq \&] ,
+.Sq \&' ,
+.Sq \&" ) .
+.Pp
+The proper spacing is also intelligently preserved if a sentence ends at
+the boundary of a macro line, e.g.,
+.Pp
+.D1 \&Xr mandoc 1 \.
+.D1 \&Fl T \&Ns \&Cm ascii \.
 .Sh MANUAL STRUCTURE
 A well-formed
 .Nm
diff --git a/usr.bin/mandoc/mdoc_macro.c b/usr.bin/mandoc/mdoc_macro.c
index 4f24b9adb77..df5ee6cec90 100644
--- a/usr.bin/mandoc/mdoc_macro.c
+++ b/usr.bin/mandoc/mdoc_macro.c
@@ -1,4 +1,4 @@
-/*	$Id: mdoc_macro.c,v 1.41 2010/05/15 13:12:55 schwarze Exp $ */
+/*	$Id: mdoc_macro.c,v 1.42 2010/05/15 15:37:53 schwarze Exp $ */
 /*
  * Copyright (c) 2008, 2009 Kristaps Dzonsons <kristaps@kth.se>
  *
@@ -615,26 +615,40 @@ rew_sub(enum mdoc_type t, struct mdoc *m,
 
 
 static int
-append_delims(struct mdoc *mdoc, int line, int *pos, char *buf)
+append_delims(struct mdoc *m, int line, int *pos, char *buf)
 {
-	int		 lastarg;
+	int		 la;
 	enum margserr	 ac;
 	char		*p;
 
-	if (0 == buf[*pos])
+	if ('\0' == buf[*pos])
 		return(1);
 
 	for (;;) {
-		lastarg = *pos;
-		ac = mdoc_zargs(mdoc, line, pos, buf, ARGS_NOWARN, &p);
+		la = *pos;
+		ac = mdoc_zargs(m, line, pos, buf, ARGS_NOWARN, &p);
 
 		if (ARGS_ERROR == ac)
 			return(0);
 		else if (ARGS_EOLN == ac)
 			break;
+
 		assert(DELIM_NONE != mdoc_isdelim(p));
-		if ( ! mdoc_word_alloc(mdoc, line, lastarg, p))
+		if ( ! mdoc_word_alloc(m, line, la, p))
 			return(0);
+		/*
+		 * If we encounter end-of-sentence symbols, then trigger
+		 * the double-space.
+		 *
+		 * XXX: it's easy to allow this to propogate outward to
+		 * the last symbol, such that `. )' will cause the
+		 * correct double-spacing.  However, (1) groff isn't
+		 * smart enough to do this and (2) it would require
+		 * knowing which symbols break this behaviour, for
+		 * example, `.  ;' shouldn't propogate the double-space.
+		 */
+		if (mandoc_eos(p, strlen(p)))
+			m->last->flags |= MDOC_EOS;
 	}
 
 	return(1);