From 9b38cbd9069ed0dc215f7029e03b3af63ee0f699 Mon Sep 17 00:00:00 2001
From: Ingo Schwarze <schwarze@openbsd.org>
Date: Sun, 23 Mar 2014 12:26:58 +0000
Subject: If a man(7) NAME section contains macros, avoid truncated or empty
 entries for .Nd in mandocdb(8), instead use the macro content recursively. 
 This improves indexing of more than 200 manuals in Xenocara, i.e. more than
 15%, in particular GL and some Xkb.

---
 mandocdb.c | 55 ++++++++-----------------------------------------------
 1 file changed, 8 insertions(+), 47 deletions(-)

(limited to 'mandocdb.c')

diff --git a/mandocdb.c b/mandocdb.c
index 50614d3d..535f0e36 100644
--- a/mandocdb.c
+++ b/mandocdb.c
@@ -1,4 +1,4 @@
-/*	$Id: mandocdb.c,v 1.120 2014/03/23 12:11:18 schwarze Exp $ */
+/*	$Id: mandocdb.c,v 1.121 2014/03/23 12:26:58 schwarze Exp $ */
 /*
  * Copyright (c) 2011, 2012 Kristaps Dzonsons <kristaps@bsd.lv>
  * Copyright (c) 2011, 2012, 2013, 2014 Ingo Schwarze <schwarze@openbsd.org>
@@ -1263,9 +1263,9 @@ static void
 parse_man(struct mpage *mpage, const struct man_node *n)
 {
 	const struct man_node *head, *body;
-	char		*start, *sv, *title;
+	char		*start, *title;
 	char		 byte;
-	size_t		 sz, titlesz;
+	size_t		 sz;
 
 	if (NULL == n)
 		return;
@@ -1285,11 +1285,7 @@ parse_man(struct mpage *mpage, const struct man_node *n)
 				NULL != (head = (head->child)) &&
 				MAN_TEXT == head->type &&
 				0 == strcmp(head->string, "NAME") &&
-				NULL != (body = body->child) &&
-				MAN_TEXT == body->type) {
-
-			title = NULL;
-			titlesz = 0;
+				NULL != body->child) {
 
 			/*
 			 * Suck the entire NAME section into memory.
@@ -1298,47 +1294,11 @@ parse_man(struct mpage *mpage, const struct man_node *n)
 			 * NAME sections over many lines.
 			 */
 
-			for ( ; NULL != body; body = body->next) {
-				if (MAN_TEXT != body->type)
-					break;
-				if (0 == (sz = strlen(body->string)))
-					continue;
-				title = mandoc_realloc
-					(title, titlesz + sz + 1);
-				memcpy(title + titlesz, body->string, sz);
-				titlesz += sz + 1;
-				title[titlesz - 1] = ' ';
-			}
+			title = NULL;
+			man_deroff(&title, body);
 			if (NULL == title)
 				return;
 
-			title = mandoc_realloc(title, titlesz + 1);
-			title[titlesz] = '\0';
-
-			/* Skip leading space.  */
-
-			sv = title;
-			while (isspace((unsigned char)*sv))
-				sv++;
-
-			if (0 == (sz = strlen(sv))) {
-				free(title);
-				return;
-			}
-
-			/* Erase trailing space. */
-
-			start = &sv[sz - 1];
-			while (start > sv && isspace((unsigned char)*start))
-				*start-- = '\0';
-
-			if (start == sv) {
-				free(title);
-				return;
-			}
-
-			start = sv;
-
 			/* 
 			 * Go through a special heuristic dance here.
 			 * Conventionally, one or more manual names are
@@ -1347,6 +1307,7 @@ parse_man(struct mpage *mpage, const struct man_node *n)
 			 * the name parts here.
 			 */
 
+			start = title;
 			for ( ;; ) {
 				sz = strcspn(start, " ,");
 				if ('\0' == start[sz])
@@ -1377,7 +1338,7 @@ parse_man(struct mpage *mpage, const struct man_node *n)
 					start++;
 			}
 
-			if (sv == start) {
+			if (start == title) {
 				putkey(mpage, start, TYPE_Nm);
 				free(title);
 				return;
-- 
cgit v1.2.3