diff --git a/classes/feedparser.php b/classes/feedparser.php index 4a2c6c2d..22052bdb 100644 --- a/classes/feedparser.php +++ b/classes/feedparser.php @@ -23,18 +23,12 @@ class FeedParser { $error = libxml_get_last_error(); // libxml compiled without iconv? - if ($error && ($error->code == 32 || $error->code == 9)) { - if (preg_match('/^(<\?xml[\t\n\r ].*?encoding=["\'])(.+?)(["\'].*?\?>)/s', $data, $matches) === 1) { - $enc = $matches[2]; - - $data = mb_convert_encoding($data, 'UTF-8', $enc); + if ($error && $error->code == 32) { + if (preg_match('/^(<\?xml[\t\n\r ].*?encoding[\t\n\r ]*=[\t\n\r ]*["\'])(.+?)(["\'].*?\?>)/s', $data, $matches) === 1) { + $data = mb_convert_encoding($data, 'UTF-8', $matches[2]); $data = preg_replace('/^<\?xml[\t\n\r ].*?\?>/s', $matches[1] . "UTF-8" . $matches[3] , $data); - - // apparently not all UTF-8 characters are valid for XML - $data = preg_replace('/[^\x{0009}\x{000a}\x{000d}\x{0020}-\x{D7FF}\x{E000}-\x{FFFD}]+/u', ' ', $data); - if ($data) { libxml_clear_errors(); @@ -43,20 +37,29 @@ class FeedParser { $error = libxml_get_last_error(); } - } + } } // some terrible invalid unicode entity? - if ($error && $error->code == 9) { - $data = mb_convert_encoding($data, 'UTF-8', 'UTF-8'); + if ($error) { + foreach(libxml_get_errors() as $err) { + if ($err->code == 9) { + // remove dangling bytes + $data = mb_convert_encoding($data, 'UTF-8', 'UTF-8'); + + // apparently not all UTF-8 characters are valid for XML + $data = preg_replace('/[^\x{0009}\x{000a}\x{000d}\x{0020}-\x{D7FF}\x{E000}-\x{FFFD}]+/u', ' ', $data); - if ($data) { - libxml_clear_errors(); + if ($data) { + libxml_clear_errors(); - $this->doc = new DOMDocument(); - $this->doc->loadXML($data); + $this->doc = new DOMDocument(); + $this->doc->loadXML($data); - $error = libxml_get_last_error(); + $error = libxml_get_last_error(); + } + break; + } } }