Update $sanitizer->removeMB4() method to use a default replacement character (can be set in $options). Plus implement a shorter replacement function (version 2). Also some updates to the $sanitizer->validateFile() method.

wire/core/PagesEditor.php

@@ -1548,7 +1548,7 @@ class PagesEditor extends Wire {
 	 * @param HookEvent $event
 	 * 
 	 */
-	protected function hookFieldtypeSleepValueStripMB4(HookEvent $event) {
+	public function hookFieldtypeSleepValueStripMB4(HookEvent $event) {
 		$event->return = $this->wire('sanitizer')->removeMB4($event->return); 
 	}
 }

wire/core/Sanitizer.php

@@ -3178,18 +3178,29 @@ class Sanitizer extends Wire {
 	 * #pw-group-strings
 	 * 
 	 * @param string|array $value String or array containing strings
+	 * @param array $options Options to modify behavior, 3.0.169+ only:
+	 *  - `replaceWith` (string): Replace MB4+ characters with this character, may not be blank (default='<27>')
+	 *  - `version` (int): Replacement method version (default=2)
 	 * @return string|array|mixed 
 	 * 
 	 */
-	public function removeMB4($value) {
+	public function removeMB4($value, array $options = array()) {
-		if(empty($value)) return $value;
+		$defaults = array(
+			'replaceWith' => "\xEF\xBF\xBD", // Default unicode replacement character: U+FFFD aka <20>
+			'version' => 2, 
+		);
+		$options = array_merge($defaults, $options);
+		if($options['replaceWith'] === '') $options['replaceWidth'] = $defaults['replaceWith'];
 		if(is_array($value)) {
+			if(!count($value)) return array();
 			// process array recursively, looking for strings to convert
 			foreach($value as $key => $val) {
-				if(empty($val)) continue;
+				if(is_string($val) || is_array($val)) $value[$key] = $this->removeMB4($val, $options);
-				if(is_string($val) || is_array($val)) $value[$key] = $this->removeMB4($val);
 			}
 		} else if(is_string($value)) {
+			if($options['version'] >= 2) {
+				$value = preg_replace('/[\x{10000}-\x{10FFFF}]/u', $options['replaceWith'], $value);
+			} else {
 				if(strlen($value) > 3 && max(array_map('ord', str_split($value))) >= 240) {
 					// string contains 4-byte characters
 					$regex =
@@ -3198,7 +3209,8 @@ class Sanitizer extends Wire {
 						'|[\xF1-\xF3][\x80-\xBF]{3}' .
 						'|\xF4[\x80-\x8F][\x80-\xBF]{2}' .
 						')!s';
-				$value = preg_replace($regex, '', $value);
+					$value = preg_replace($regex, $options['replaceWith'], $value);
+				}
 			}
 		} else {
 			// not a string or an array, leave as-is
@@ -4807,9 +4819,10 @@ class Sanitizer extends Wire {
 		$extension = strtolower(pathinfo($filename, PATHINFO_EXTENSION));
 		$validatorNames = array();
 		$validatorResults = array();
-		$isValid = null;
 		$getArray = $options['getArray'];
 		$dryrun = $options['dryrun'] || !empty($options['dryRun']);
+		$numFailed = 0;
+		$numPassed = 0;
 
 		if(!strlen($extension) || (!$dryrun && !is_file($filename))) {
 			return $getArray ? array() : null;
@@ -4857,11 +4870,11 @@ class Sanitizer extends Wire {
 				// true (bool): file is valid as-is
 				// 1 (int): file is valid as a result of sanitization
 				// in either case, continue on to the next applicable FileValidator module
-				continue;
+				$numPassed++;
-			}
 				
+			} else {
 				// at this point we’ve determined file is not valid
-			$isValid = false;
+				$numFailed++;
 
 				// move errors to Sanitizer class so they can be retrieved
 				foreach($validator->errors('clear array') as $error) {
@@ -4872,8 +4885,18 @@ class Sanitizer extends Wire {
 				// unless we are returning an array of results, we can stop now for invalid files
 				if(!$getArray) break;
 			}
+		}
 	
-		return $getArray ? $validatorResults : $isValid;
+		// return array result of all validations
+		if($getArray) return $validatorResults;
+		
+		// return null if no validators could be used
+		if(!$numPassed && !$numFailed) return null;
+
+		// if passsed 1+ validators and failed 0, return true
+		if($numPassed > 0 && $numFailed === 0) return true;
+		
+		return false;
 	}
 
 	/**********************************************************************************************************************