From f7e3d0dbf6a763557d3431c6d12a6dde1be5648b Mon Sep 17 00:00:00 2001
From: zotlabs <mike@macgirvin.com>
Date: Thu, 9 Nov 2017 21:26:18 -0800
Subject: hubzilla issue #901 - unicode characters in urls tripping up url
 regexes - these regexes have been modified to accept unicode "letters" which
 may preclude emojis and control sequences and symbols in url links; but
 should suffice for most legal URLs containing language context "text" glyphs.

---
 include/bbcode.php | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'include/bbcode.php')

diff --git a/include/bbcode.php b/include/bbcode.php
index 9a2a6eb9b..050ab2d29 100644
--- a/include/bbcode.php
+++ b/include/bbcode.php
@@ -838,13 +838,13 @@ function bbcode($Text, $preserve_nl = false, $tryoembed = true, $cache = false)
 
 	// Perform URL Search
 
-	$urlchars = '[a-zA-Z0-9\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,\@\(\)]';
+	$urlchars = '[a-zA-Z0-9\pL\:\/\-\?\&\;\.\=\_\~\#\%\$\!\+\,\@\(\)]';
 
 	if (strpos($Text,'http') !== false) {
 		if($tryoembed) {
-			$Text = preg_replace_callback("/([^\]\='".'"'."\/]|^|\#\^)(https?\:\/\/$urlchars+)/ism", 'tryoembed', $Text);
+			$Text = preg_replace_callback("/([^\]\='".'"'."\/]|^|\#\^)(https?\:\/\/$urlchars+)/ismu", 'tryoembed', $Text);
 		}
-		$Text = preg_replace("/([^\]\='".'"'."\/]|^|\#\^)(https?\:\/\/$urlchars+)/ism", '$1<a href="$2" target="_blank" rel="nofollow noopener">$2</a>', $Text);
+		$Text = preg_replace("/([^\]\='".'"'."\/]|^|\#\^)(https?\:\/\/$urlchars+)/ismu", '$1<a href="$2" target="_blank" rel="nofollow noopener">$2</a>', $Text);
 	}
 
 	if (strpos($Text,'[/share]') !== false) {
-- 
cgit v1.2.3