mb_internal_encoding - PHP code example

The Function mb_Internal_Encoding in PHP

The mbstring extension gives PHP multibyte-specific string functions and handles character encoding conversion between the possible encoding pairs. mb_internal_encoding sets the internal character encoding used by mbstring. It is not used by any other function and does not affect the character encoding utilized in HTTP input or output.

mb_internal_encoding has been EXPERIMENTAL since PHP 4.4.0. This means that the behavior of this function (and in fact, ANYTHING documented here) may change in a future release without notification.

If you use mb_internal_encoding to set your own internal character encoding and then later upgrade to a version of PHP that does not support that encoding, it is likely that the PHP interpreter will raise an error when it encounters a string that it cannot encode properly. This is especially a problem for multibyte string functions such as mb_strtolower($str).

To prevent this from happening, make sure that your scripts are using a valid encoding. You can test if your script is using the right encoding by checking for the mb_detect_encoding() function. This will return the mb_internal_encoding for your script. If this is not UTF-8, you should rebuild your server with mbstring and check the mb_internal_encoding setting again. Also, you should move any header sending code before the mb_detect_encoding() call. This will prevent PHP from trying to send headers after it has already written them to the output stream. This would result in an E_WARNING. The most common causes of this E_WARNING are whitespace or new lines after the ?php closing tag and Byte Order Marks.

mb_internal_encoding example

function smarty_mb_from_unicode($unicode, $encoding=null) {
    $t = '';
    if (!$encoding) {
        $encoding = mb_internal_encoding();
    }
    foreach((array) $unicode as $utf32be) {
        $character = pack("N*", $utf32be);
        $t .= mb_convert_encoding($character, $encoding, "UTF-32BE");
    }
    return $t;
}

?>

/** * @param string $charset * * @return Enlight_Template_Manager */
    public function setCharset($charset = null)
    {
        if ($charset !== null) {
            self::$_CHARSET = $charset;
        }
        mb_internal_encoding(self::$_CHARSET);

        return $this;
    }

    /** * @param array|Enlight_Config $options * * @return Enlight_Template_Manager */
    public function setOptions($options = null)
    {

    /** * Takes an unencoded string and produces a Base64 encoded string from it. * * If the charset is iso-2022-jp, it uses mb_encode_mimeheader instead of * default encodeString, otherwise pass to the parent method. */
    public function encodeString(string $string, ?string $charset = 'utf-8', int $firstLineOffset = 0, int $maxLineLength = 0): string
    {
        if ('iso-2022-jp' === strtolower($charset)) {
            $old = mb_internal_encoding();
            mb_internal_encoding('utf-8');
            $newstring = mb_encode_mimeheader($string, 'iso-2022-jp', $this->getName(), "\r\n");
            mb_internal_encoding($old);

            return $newstring;
        }

        return parent::encodeString($string, $charset, $firstLineOffset, $maxLineLength);
    }
}

$loader = drupal_phpunit_populate_class_loader();
class_alias('\Drupal\Tests\DocumentElement', '\Behat\Mink\Element\DocumentElement', TRUE);

ClassWriter::mutateTestBase($loader);

// Set sane locale settings, to ensure consistent string, dates, times and // numbers handling. // @see \Drupal\Core\DrupalKernel::bootEnvironment() setlocale(LC_ALL, 'C');

// Set appropriate configuration for multi-byte strings. mb_internal_encoding('utf-8');
mb_language('uni');

// Set the default timezone. While this doesn't cause any tests to fail, PHP // complains if 'date.timezone' is not set in php.ini. The Australia/Sydney // timezone is chosen so all tests are run using an edge case scenario (UTC+10 // and DST). This choice is made to prevent timezone related regressions and // reduce the fragility of the testing system in general. date_default_timezone_set('Australia/Sydney');

// Runtime assertions. PHPUnit follows the php.ini assert.active setting for // runtime assertions. By default this setting is on. Ensure exceptions are

      // Send an empty string to disable the cache limiter.       ini_set('session.cache_limiter', '');
      // Use httponly session cookies.       ini_set('session.cookie_httponly', '1');
    }

    // Set sane locale settings, to ensure consistent string, dates, times and     // numbers handling.     setlocale(LC_ALL, 'C');

    // Set appropriate configuration for multi-byte strings.     mb_internal_encoding('utf-8');
    mb_language('uni');

    // Indicate that code is operating in a test child site.     if (!defined('DRUPAL_TEST_IN_CHILD_SITE')) {
      if ($test_prefix = drupal_valid_test_ua()) {
        $test_db = new TestDatabase($test_prefix);
        // Only code that interfaces directly with tests should rely on this         // constant; e.g., the error/exception handler conditionally adds further         // error information into HTTP response headers that are consumed by         // the internal browser.         define('DRUPAL_TEST_IN_CHILD_SITE', TRUE);

if (1 > $split_length = (int) $split_length) {
            if (80000 > \PHP_VERSION_ID) {
                trigger_error('The length of each segment must be greater than zero', \E_USER_WARNING);

                return false;
            }

            throw new \ValueError('Argument #2 ($length) must be greater than 0');
        }

        if (null === $encoding) {
            $encoding = mb_internal_encoding();
        }

        if ('UTF-8' === $encoding = self::getEncoding($encoding)) {
            $rx = '/(';
            while (65535 < $split_length) {
                $rx .= '.{65535}';
                $split_length -= 65535;
            }
            $rx .= '.{'.$split_length.'})/us';

            return preg_split($rx, $string, -1, \PREG_SPLIT_DELIM_CAPTURE | \PREG_SPLIT_NO_EMPTY);
        }

    public static function decodeChunkedBody($body)
    {
        $decBody = '';

        // If mbstring overloads substr and strlen functions, we have to         // override it's internal encoding         if (function_exists('mb_internal_encoding') &&
           ((int) ini_get('mbstring.func_overload')) & 2) {

            $mbIntEnc = mb_internal_encoding();
            mb_internal_encoding('ASCII');
        }

        while (trim($body)) {
            if (! preg_match("/^([\da-fA-F]+)[^\r\n]*\r\n/sm", $body, $m)) {
                throw new Zend_Http_Exception("Error parsing body - doesn't seem to be a chunked message");
            }

            $length = hexdec(trim($m[1]));
            $cut = strlen($m[0]);
            $decBody .= substr($body, $cut, $length);

$overloaded = true;
        } else {
            $overloaded = false;
        }
    }

    if ( false === $overloaded ) {
        return;
    }

    if ( ! $reset ) {
        $encoding = mb_internal_encoding();
        array_push( $encodings, $encoding );
        mb_internal_encoding( 'ISO-8859-1' );
    }

    if ( $reset && $encodings ) {
        $encoding = array_pop( $encodings );
        mb_internal_encoding( $encoding );
    }
}

/** * Resets the mbstring internal encoding to a users previously set encoding. * * @see mbstring_binary_safe_encoding() * * @since 3.7.0 */

$options = $options->toArray();
        } else if (!is_array($options)) {
            $options = func_get_args();
            $temp    = array();
            if (!empty($options)) {
                $temp['encoding'] = array_shift($options);
            }
            $options = $temp;
        }

        if (!array_key_exists('encoding', $options) && function_exists('mb_internal_encoding')) {
            $options['encoding'] = mb_internal_encoding();
        }

        if (array_key_exists('encoding', $options)) {
            $this->setEncoding($options['encoding']);
        }
    }

    /** * Returns the set encoding * * @return string */

        if ($K && null === self::$KD) {
            self::$KD = self::getData('compatibilityDecomposition');
        }

        if (null === self::$D) {
            self::$D = self::getData('canonicalDecomposition');
            self::$cC = self::getData('combiningClass');
        }

        if (null !== $mbEncoding = (2 /* MB_OVERLOAD_STRING */ & (int) \ini_get('mbstring.func_overload')) ? mb_internal_encoding() : null) {
            mb_internal_encoding('8bit');
        }

        $r = self::decompose($s, $K);

        if ($C) {
            if (null === self::$C) {
                self::$C = self::getData('canonicalComposition');
            }

            $r = self::recompose($r);
        }

    if (!function_exists('iconv_strpos')) {
        function iconv_strpos(?string $haystack, ?string $needle, ?int $offset = 0, ?string $encoding = null): int|false { null === $encoding && $encoding = p\Iconv::$internalEncoding; return mb_strpos((string) $haystack, (string) $needle, (int) $offset, $encoding); }
    }
    if (!function_exists('iconv_strrpos')) {
        function iconv_strrpos(?string $haystack, ?string $needle, ?string $encoding = null): int|false { null === $encoding && $encoding = p\Iconv::$internalEncoding; return mb_strrpos((string) $haystack, (string) $needle, 0, $encoding); }
    }
    if (!function_exists('iconv_substr')) {
        function iconv_substr(?string $string, ?int $offset, ?int $length = null, ?string $encoding = null): string|false { null === $encoding && $encoding = p\Iconv::$internalEncoding; return mb_substr((string) $string, (int) $offset, $length, $encoding); }
    }
    if (!function_exists('iconv_mime_decode')) {
        function iconv_mime_decode($string, $mode = 0, $encoding = null) { $currentMbEncoding = mb_internal_encoding(); null === $encoding && $encoding = p\Iconv::$internalEncoding; mb_internal_encoding($encoding); $decoded = mb_decode_mimeheader($string); mb_internal_encoding($currentMbEncoding); return $decoded; }
    }
} else {
    if (!function_exists('iconv_strlen')) {
        if (extension_loaded('xml')) {
            function iconv_strlen(?string $string, ?string $encoding = null): int|false { return p\Iconv::strlen1((string) $string, $encoding); }
        } else {
            function iconv_strlen(?string $string, ?string $encoding = null): int|false { return p\Iconv::strlen2((string) $string, $encoding); }
        }
    }

    if (!function_exists('iconv_strpos')) {

/** * Returns the current encoding (see: mb_internal_encoding()) * * @see http://php.net/manual/en/function.mb-internal-encoding.php * * @param array $params * @param array $smarty * @param string $template */
function smarty_function_encoding($params, $smarty, $template = null)
{
return mb_internal_encoding();
}

/**#@-*/

    /** * Initialize new Smarty object * */
    public function __construct()
    {
        // selfpointer needed by some other class methods         $this->smarty = $this;
        if (is_callable('mb_internal_encoding')) {
            mb_internal_encoding(Smarty::$_CHARSET);
        }
        $this->start_time = microtime(true);
        // set default dirs         $this->setTemplateDir('.' . DS . 'templates' . DS)
            ->setCompileDir('.' . DS . 'templates_c' . DS)
            ->setPluginsDir(SMARTY_PLUGINS_DIR)
            ->setCacheDir('.' . DS . 'cache' . DS)
            ->setConfigDir('.' . DS . 'configs' . DS);

        $this->debug_tpl = 'file:' . dirname(__FILE__) . '/debug.tpl';
        if (isset($_SERVER['SCRIPT_NAME'])) {

    public function min()
    {
        if ($this->output !== '') { // min already run             return $this->output;
        }

        $mbIntEnc = null;
        if (function_exists('mb_strlen') && ((int)ini_get('mbstring.func_overload') & 2)) {
            $mbIntEnc = mb_internal_encoding();
            mb_internal_encoding('8bit');
        }
        $this->input = str_replace("\r\n", "\n", $this->input);
        $this->inputLength = strlen($this->input);

        $this->action(self::ACTION_DELETE_A_B);

        while ($this->a !== null) {
            // determine next command             $command = self::ACTION_KEEP_A; // default             if ($this->a === ' ') {