From: Miecio Za <miecio@miecio.net>
Date: Thu, 7 Mar 2019 11:40:46 +0000 (+0100)
Subject: [Fix] Fix memor leaks and whitespace processing
X-Git-Tag: 1.9.0~19^2
X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=refs%2Fpull%2F2771%2Fhead;p=thirdparty%2Frspamd.git

[Fix] Fix memor leaks and whitespace processing

Fixed for memory leaks, add simple whitespace character
processing. Everything like '\n' '\r' '\t' should be
replaced by simple space.
---

diff --git a/src/libserver/html.c b/src/libserver/html.c
index cdcaf4ebbf..fef769ec10 100644
--- a/src/libserver/html.c
+++ b/src/libserver/html.c
@@ -2373,11 +2373,28 @@ rspamd_html_check_displayed_url (rspamd_mempool_t *pool,
 		return;
 	}
 
-	gint visible_part_len = dest->len - href_offset;
-	url->visible_part = rspamd_mempool_alloc0(pool, visible_part_len +1);
-	url->visible_partlen = visible_part_len;
-	gchar *visible_part = g_strndup(dest->data + href_offset, visible_part_len);
-	g_stpcpy(url->visible_part, visible_part);
+	url->visible_part = rspamd_mempool_alloc0(pool, dest->len - href_offset+1);
+	gchar *current_processed_char = dest->data + href_offset;
+	gchar *current_char_in_struct = url->visible_part;
+	gboolean previous_char_was_space = false;
+
+	while (current_processed_char < (gchar*) dest->data + dest->len) {
+		if (g_ascii_isspace(*current_processed_char)) {
+			if (previous_char_was_space) {
+				current_processed_char++;
+				continue;
+			}
+			previous_char_was_space = true;
+			*current_char_in_struct = ' ';
+		} else {
+			*current_char_in_struct = *current_processed_char;
+			previous_char_was_space = false;
+		}
+		current_char_in_struct++;
+		current_processed_char++;
+	}
+	*current_char_in_struct = '\0';
+	url->visible_partlen = current_char_in_struct - url->visible_part;
 
 	rspamd_html_url_is_phished (pool, url,
 			dest->data + href_offset,