From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from localhost (fantadrom.bsd.lv [local])
	by fantadrom.bsd.lv (OpenSMTPD) with ESMTPA id 5d755537
	for <source@mandoc.bsd.lv>;
	Fri, 12 Apr 2019 01:47:15 -0500 (EST)
Date: Fri, 12 Apr 2019 01:47:15 -0500 (EST)
X-Mailinglist: mandoc-source
Reply-To: source@mandoc.bsd.lv
MIME-Version: 1.0
From: schwarze@mandoc.bsd.lv
To: source@mandoc.bsd.lv
Subject: docbook2mdoc: If a text node follows a non-text node without intervening 
X-Mailer: activitymail 1.26, http://search.cpan.org/dist/activitymail/
Content-Type: text/plain; charset=utf-8
Message-ID: <e3fdb0ca08af148d@fantadrom.bsd.lv>

Log Message:
-----------
If a text node follows a non-text node without intervening whitespace, 
keep the text in it as short as possible.

Modified Files:
--------------
    docbook2mdoc:
        parse.c

Revision Data
-------------
Index: parse.c
===================================================================
RCS file: /home/cvs/mdocml/docbook2mdoc/parse.c,v
retrieving revision 1.34
retrieving revision 1.35
diff -Lparse.c -Lparse.c -u -p -r1.34 -r1.35
--- parse.c
+++ parse.c
@@ -322,44 +322,88 @@ warn_msg(struct parse *p, const char *fm
  * Otherwise, create a new one as a child of the current node.
  */
 static void
-xml_char(struct parse *p, const char *word, int sz)
+xml_text(struct parse *p, const char *word, int sz)
 {
-	struct pnode	*n;
+	struct pnode	*n, *np;
 	size_t		 oldsz, newsz;
+	int		 i;
 
 	assert(sz > 0);
 	if (p->del > 0)
 		return;
 
 	if ((n = p->cur) == NULL) {
-		error_msg(p, "discarding text before document: %.*s", sz, word);
+		error_msg(p, "discarding text before document: %.*s",
+		    sz, word);
 		return;
 	}
 
-	if (n->node != NODE_TEXT) {
-		if ((n = pnode_alloc(p->cur)) == NULL)
+	/* Append to the current text node, if one is open. */
+
+	if (n->node == NODE_TEXT) {
+		oldsz = strlen(n->b);
+		newsz = oldsz + sz;
+		if (oldsz && (p->flags & PFLAG_SPC))
+			newsz++;
+		if ((n->b = realloc(n->b, newsz + 1)) == NULL)
 			fatal(p);
-		n->node = NODE_TEXT;
-		n->spc = (p->flags & PFLAG_SPC) != 0;
-		p->cur = n;
+		if (oldsz && (p->flags & PFLAG_SPC))
+			n->b[oldsz++] = ' ';
+		memcpy(n->b + oldsz, word, sz);
+		n->b[newsz] = '\0';
+		p->flags &= ~PFLAG_SPC;
+		return;
 	}
 
-	if (p->tree->flags & TREE_CLOSED && n->parent == p->tree->root)
+	if (p->tree->flags & TREE_CLOSED && n == p->tree->root)
 		warn_msg(p, "text after end of document: %.*s", sz, word);
 
-	/* Append to the current text node. */
+	/* Create a new text node. */
 
-	oldsz = n->b == NULL ? 0 : strlen(n->b);
-	newsz = oldsz + sz;
-	if (oldsz && (p->flags & PFLAG_SPC))
-		newsz++;
-	if ((n->b = realloc(n->b, newsz + 1)) == NULL)
+	if ((n = pnode_alloc(p->cur)) == NULL)
 		fatal(p);
-	if (oldsz && (p->flags & PFLAG_SPC))
-		n->b[oldsz++] = ' ';
-	memcpy(n->b + oldsz, word, sz);
-	n->b[newsz] = '\0';
+	n->node = NODE_TEXT;
+	n->spc = (p->flags & PFLAG_SPC) != 0;
 	p->flags &= ~PFLAG_SPC;
+
+	/*
+	 * If this node follows a non-text node without intervening
+	 * whitespace, keep the text in it as short as possible,
+	 * and do not keep it open.
+	 */
+
+	if (n->spc == 0 &&
+	    (np = TAILQ_PREV(n, pnodeq, child)) != NULL &&
+	    np->node != NODE_TEXT && np->node != NODE_ESCAPE) {
+		i = 0;
+		while (i < sz && !isspace((unsigned char)word[i]))
+			i++;
+		if ((n->b = strndup(word, i)) == NULL)
+			fatal(p);
+		if (i == sz)
+			return;
+		while (i < sz && isspace((unsigned char)word[i]))
+			i++;
+		if (i == sz) {
+			p->flags |= PFLAG_SPC;
+			return;
+		}
+
+		/* Put any remaining text into a second node. */
+
+		if ((n = pnode_alloc(p->cur)) == NULL)
+			fatal(p);
+		n->node = NODE_TEXT;
+		n->spc = 1;
+		word += i;
+		sz -= i;
+	}
+	if ((n->b = strndup(word, sz)) == NULL)
+		fatal(p);
+
+	/* The new node remains open for later pnode_closetext(). */
+
+	p->cur = n;
 }
 
 /*
@@ -959,7 +1003,7 @@ parse_string(struct parse *p, char *b, s
 			advance(p, b, rlen, &pend,
 			    p->ncur == NODE_DOCTYPE ? "<&]\n" : "<&\n",
 			    refill);
-			xml_char(p, b + poff, pend - poff);
+			xml_text(p, b + poff, pend - poff);
 			if (b[pend] == '\n')
 				pnode_closetext(p);
 		}
--
 To unsubscribe send an email to source+unsubscribe@mandoc.bsd.lv