From c78c9e8936467a043d7391bd6994eb881eb0cc70 Mon Sep 17 00:00:00 2001
From: Kristaps Dzonsons <kristaps@bsd.lv>
Date: Thu, 1 Sep 2011 22:25:53 +0000
Subject: Finishing touches on demandoc.  It now backs over ending punctuation
 as well as leading punctuation.  Again, this isn't the same as deroff (which
 uses, I think, some punctuation as delimiters), but it's easier to explain
 and simpler to audit.

---
 demandoc.1 | 14 ++++++++++++--
 demandoc.c | 18 +++++++++++++++---
 2 files changed, 27 insertions(+), 5 deletions(-)

diff --git a/demandoc.1 b/demandoc.1
index 169616d2..1c1cd585 100644
--- a/demandoc.1
+++ b/demandoc.1
@@ -41,8 +41,10 @@ This outputs each word of text on its own line.
 A
 .Qq word ,
 in this case, refers to whitespace-delimited terms beginning with at
-least two letters after opening punctuation and not consisting of any
-escape sequences.
+least two letters and not consisting of any escape sequences.
+Terms have their leading and trailing punctuation
+.Pq double-quotes, sentence punctuation, etc.
+stripped.
 .It Ar
 The input files.
 .El
@@ -84,6 +86,14 @@ Such errors cause
 to exit at once, possibly in the middle of parsing or formatting a file.
 The output databases are corrupt and should be removed .
 .El
+.Sh EXAMPLES
+The traditional usage of
+.Nm
+is for spell-checking manuals on
+.Bx .
+This is accomplished as follows (assuming British spelling):
+.Pp
+.Dl $ demandoc -w file.1 | spell -b
 .Sh SEE ALSO
 .Xr mandoc 1 ,
 .Xr man 7
diff --git a/demandoc.c b/demandoc.c
index b61e1fe8..8b6533f2 100644
--- a/demandoc.c
+++ b/demandoc.c
@@ -132,7 +132,7 @@ static void
 pstring(const char *p, int col, int *colp, int list)
 {
 	enum mandoc_esc	 esc;
-	const char	*start;
+	const char	*start, *end;
 	int		 emit;
 
 	/*
@@ -161,8 +161,20 @@ again:
 			} else if (isspace((unsigned char)*p))
 				break;
 
-		if (emit && p - start >= 2) {
-			for ( ; start != p; start++)
+		end = p - 1;
+
+		while (end > start)
+			if ('.' == *end || ',' == *end || 
+					'\'' == *end || '"' == *end ||
+					')' == *end || '!' == *end ||
+					'?' == *end || ':' == *end ||
+					';' == *end)
+				end--;
+			else
+				break;
+
+		if (emit && end - start >= 1) {
+			for ( ; start <= end; start++)
 				if (ASCII_HYPH == *start)
 					putchar('-');
 				else
-- 
cgit