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.

This commit is contained in:
zotlabs
2017-11-09 21:26:18 -08:00
parent 766a7fb1b4
commit 9b19e40a74
4 changed files with 13 additions and 13 deletions

View File

@@ -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) {