PHP: mb_range() - Unicode compatible range('A', 'Z')

mb_range.php

<?php

mb_internal_encoding('UTF-8');

/**
 * multibyte string compatible range('A', 'Z')
 *
 * @param string $start Character to start from (included)
 * @param string $end Character to end with (included)
 * @return array list of characters in unicode alphabet from $start to $end
 * @author Rodney Rehm
 */
function mb_range($start, $end) {
    // if start and end are the same, well, there's nothing to do
    if ($start == $end) {
        return array($start);
    }
    
    $_result = array();
    // get unicodes of start and end
    list(, $_start, $_end) = unpack("N*", mb_convert_encoding($start . $end, "UTF-32BE", "UTF-8"));
    // determine movement direction
    $_offset = $_start < $_end ? 1 : -1;
    $_current = $_start;
    while ($_current != $_end) {
        $_result[] = mb_convert_encoding(pack("N*", $_current), "UTF-8", "UTF-32BE");
        $_current += $_offset;
    }
    $_result[] = $end;
    return $_result;
}

var_dump(mb_range('क', 'म'));
var_dump(mb_range('A', 'C'));
var_dump(mb_range('A', 'Z'));
// beware, ä…ö are not a sequence you'd expect!
var_dump(mb_range('ä', 'ö'));

How to use?

@tegardazzly

How to use?

You can encode decimals to Nth base number more than N=64.
For example, the code below you can shorten MD5 hash to 10 Chinese characters.

<?php

include('mb_range.php');

mb_internal_encoding('UTF-8');

// Sample usage
$s = 'This is a sample string to be hashed.';
echo $s . PHP_EOL;                              // 'This is a sample string to be hashed.'
echo md5($s) . PHP_EOL;                         // 'baad33e1e97f316b9750c27c86bf64d6'
echo mb_base_encode(hexdec(md5($s))) . PHP_EOL; // '䙔倁屩劷䋾彨䏔䂌䤘剒'
echo mb_md5($s) . PHP_EOL;                      // '䙔倁屩劷䋾彨䏔䂌䤘剒'

/**
 * Returns base Nth encoded string from decimal number input.
 *
 * @param  integer $number Decimal number to encode
 * @return string          Encoded string
 */
function mb_base_encode($number)
{
    $char = array_merge(
        mb_range('䀀', '䶵'), // U+4000 - U+4DB5
        mb_range('一', '俿'), // U+4E00 - U+4FFF
        mb_range('倀', '忿') //  U+5000 - U+5FFF
        // REF: http://www.utf8-chartable.com/unicode-utf8-table.pl?start=16384&number=1024
    );

    $base   = count($char);
    $result = "";
    while ($number > 0) {
        $result = $char[ fmod($number, $base) ] . $result;
        $number = floor($number / $base);
    }

    return ( $result == "" ) ? 0 : $result;
}

function mb_md5($string)
{
    return mb_base_encode(hexdec(md5($string)));
}