How to keep json_encode() from dropping strings with invalid characters

Question

Is there a way to keep json_encode() from returning null for a string that contains an invalid (non-UTF-8) character?

It can be a pain in t

Answer 1

to get a informational error notification on json failures we use this helper:

• installs temporarily a custom error handler to catch json errors for encoding/decoding
• throws RuntimeException on error

'foo']);
 * $array = HelperJson::decode('{"bla":"foo"}');
 * 
 * throws exception on failure
 * 
 */
class HelperJson {

    /**
     * @var array
     */
    static private $jsonErrors = [
            JSON_ERROR_NONE => '',
            JSON_ERROR_UTF8 => 'Malformed UTF-8 characters, possibly incorrectly encoded',
            JSON_ERROR_DEPTH => 'Maximum stack depth exceeded',
            JSON_ERROR_STATE_MISMATCH => 'Underflow or the modes mismatch',
            JSON_ERROR_CTRL_CHAR => 'Unexpected control character found',
            JSON_ERROR_SYNTAX => 'Syntax error, malformed JSON',
    ];

    /**
     * ! assoc ! (reverse logic to php function)
     * @param string $jsonString
     * @param bool $assoc
     * @throws RuntimeException
     * @return array|null
     */
    static public function decode($jsonString, $assoc=true){

        HelperJson_ErrorHandler::reset(); // siehe unten
        set_error_handler('HelperJson_ErrorHandler::handleError');

        $result = json_decode($jsonString, $assoc);

        $errStr = HelperJson_ErrorHandler::getErrstr();
        restore_error_handler();

        $jsonError = json_last_error();
        if( $jsonError!=JSON_ERROR_NONE ) {
            $errorMsg = isset(self::$jsonErrors[$jsonError]) ? self::$jsonErrors[$jsonError] : 'unknown error code: '.$jsonError;
            throw new \RuntimeException('json decoding error: '.$errorMsg.' JSON: '.substr($jsonString,0, 250));
        }
        if( $errStr!='' ){
            throw new \RuntimeException('json decoding problem: '.$errStr.' JSON: '.substr($jsonString,0, 250));
        }
        return $result;
    }

    /**
     * encode with error "throwing"
     * @param mixed $data
     * @param int $options   $options=JSON_PRESERVE_ZERO_FRACTION+JSON_UNESCAPED_SLASHES : 1024 + 64 = 1088
     * @return string
     * @throws \RuntimeException
     */
    static public function encode($data, $options=1088){

        HelperJson_ErrorHandler::reset();// scheint notwendg da sonst bei utf-8 problemen nur eine warnung geflogen ist, die hier aber nicht durchschlug, verdacht der error handler macht selbst was mit json und reset damit json_last_error
        set_error_handler('HelperJson_ErrorHandler::handleError');

        $result = json_encode($data, $options);

        $errStr = HelperJson_ErrorHandler::getErrstr();
        restore_error_handler();

        $jsonError = json_last_error();
        if( $jsonError!=JSON_ERROR_NONE ){
            $errorMsg = isset(self::$jsonErrors[$jsonError]) ? self::$jsonErrors[$jsonError] : 'unknown error code: '.$jsonError;
            throw new \RuntimeException('json encoding error: '.$errorMsg);
        }
        if( $errStr!='' ){
            throw new \RuntimeException('json encoding problem: '.$errStr);
        }
        return $result;
    }

}

/**

HelperJson_ErrorHandler::install();
preg_match('~a','');
$errStr = HelperJson_ErrorHandler::getErrstr();
HelperJson_ErrorHandler::remove();

 *
 */
class HelperJson_ErrorHandler {

    static protected  $errno = 0;
    static protected  $errstr = '';
    static protected  $errfile = '';
    static protected  $errline = '';
    static protected  $errcontext = array();

    /**
     * @param int $errno
     * @param string $errstr
     * @param string $errfile
     * @param int $errline
     * @param array $errcontext
     * @return bool
     */
    static public function handleError($errno, $errstr, $errfile, $errline, $errcontext){
        self::$errno = $errno;
        self::$errstr = $errstr;
        self::$errfile = $errfile;
        self::$errline = $errline;
        self::$errcontext = $errcontext;
        return true;
    }

    /**
     * @return int
     */
    static public function getErrno(){
        return self::$errno;
    }
    /**
     * @return int
     */
    static public function getErrstr(){
        return self::$errstr;
    }
    /**
     * @return int
     */
    static public function getErrfile(){
        return self::$errfile;
    }
    /**
     * @return int
     */
    static public function getErrline(){
        return self::$errline;
    }
    /**
     * @return array
     */
    static public function getErrcontext(){
        return self::$errcontext;
    }
    /**
     * reset last error
     */
    static public function reset(){
        self::$errno = 0;
        self::$errstr = '';
        self::$errfile = '';
        self::$errline = 0;
        self::$errcontext = array();
    }

    /**
     * set black-hole error handler
     */
    static public function install(){
        self::reset();
        set_error_handler('HelperJson_ErrorHandler::handleError');
    }

    /**
     * restore previous error handler
     */
    static function remove(){
        restore_error_handler();
    }
}