Feedparser: Change handling of libxml error 9 (cycle all errors)

classes/feedparser.php

@@ -23,18 +23,12 @@ class FeedParser {
 		$error = libxml_get_last_error();
 
 		// libxml compiled without iconv?
-		if ($error && ($error->code == 32 || $error->code == 9)) {
+		if ($error && $error->code == 32) {
-			if (preg_match('/^(<\?xml[\t\n\r ].*?encoding=["\'])(.+?)(["\'].*?\?>)/s', $data, $matches) === 1) {
+			if (preg_match('/^(<\?xml[\t\n\r ].*?encoding[\t\n\r ]*=[\t\n\r ]*["\'])(.+?)(["\'].*?\?>)/s', $data, $matches) === 1) {
-				$enc = $matches[2];
+				$data = mb_convert_encoding($data, 'UTF-8', $matches[2]);
-
-				$data = mb_convert_encoding($data, 'UTF-8', $enc);
 
 				$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();
 
@@ -47,9 +41,15 @@ class FeedParser {
 		}
 
 		// some terrible invalid unicode entity?
-		if ($error && $error->code == 9) {
+		if ($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();
 
@@ -58,6 +58,9 @@ class FeedParser {
 
 						$error = libxml_get_last_error();
 					}
+					break;
+				}
+			}
 		}
 
 		$this->error = $this->format_error($error);