uniqid

(PHP 4, PHP 5, PHP 7, PHP 8)

uniqid — Erzeugt eine eindeutige ID

Beschreibung

uniqid(string $prefix = "", bool $more_entropy = false): string

Gibt eine eindeutige ID mit Präfix zurück, die auf der aktuellen Zeit in Mikrosekunden basiert.

Achtung

Diese Funktion erzeugt keine kryptografisch sicheren Werte und darf nicht für kryptografische Zwecke verwendet werden oder für Zwecke, bei denen die zurückgegebenen Werte nicht abschätzbar sein dürfen.

Falls kryptographisch sichere Zufallszahlen benötigt werden, kann der Random\Randomizer mit der Random\Engine\Secure-Engine verwendet werden. Für einfache Anwendungsfälle bieten die Funktionen random_int() und random_bytes() eine bequeme und sichere API, die den CSPRNG des Betriebssystems verwendet.

Warnung

Diese Funktion garantiert nicht die Eindeutigkeit der Rückgabewerte. Da die meisten System die Systemzeit durch NTP oder ähnlich justieren, ändert sich die Systemzeit kontinuierlich. Daher ist es möglich, dass diese Funktion keine eindeutige ID für den Prozess/Thread zurückgibt. more_entropy kann verwendet werden, um die Wahrscheinlichkeit der Eindeutigkeit zu erhöhen.

Parameter-Liste

prefix

Kann z. B. dann hilfreich sein, wenn Sie auf mehreren Hosts gleichzeitig IDs erzeugen, was im ungünstigsten Fall zur selben Mikrosekunde geschehen kann.

Bei leerem prefix ist die zurückgegebene Zeichenkette 13 Zeichen lang. Falls more_entropy true ist, sind es 23 Zeichen.

more_entropy

Falls dieser Parameter auf true gesetzt ist, wird uniqid() am Ende des Rückgabewertes zusätzliche mit dem Kongruenzgenerator für Pseudozufallszahlen (Combined Linear Congruential Generator) erzeugte Entropie anfügen, wodurch die Wahrscheinlichkeit erhöht wird, dass das Ergebnis eindeutig ist.

Rückgabewerte

Gibt eine zeitstempelbasierte eindeutige ID als Zeichenkette zurück.

Warnung

Diese Funktion versucht eine eindeutige Kennung zu erzeugen, aber sie garantiert nicht die hunderprozentige Eindeutigkeit des Rückgabewerts.

Beispiele

Beispiel #1 uniqid()-Beispiel

<?php
/* Eine eindeutige ID, beispielsweise: 4b3403665fea6 */
$uniqid = uniqid();

/* Zusätzlich kann ein Präfix genutzt werden. Dabei sind die
 * folgenden beiden Anweisungen identisch:
 */
$uniqid = uniqid($prefix);
$uniqid = $prefix . uniqid();

/* Weiterhin kann der more_entropy Parameter aktiviert werden (auf einigen Systemen
 * wie Cygwin ist dies sogar zwangsweise erforderlich). Dadurch generiert uniqid()
 * einen Wert ähnlich dem folgenden: 4b340550242239.64159797
 */
$uniqid = uniqid('', true);

Anmerkungen

Hinweis:
Unter Cygwin muss der Parameter more_entropy auf true gesetzt sein, damit die Funktion arbeitet.

Siehe auch

random_bytes() - Get cryptographically secure random bytes

add a note

User Contributed Notes 16 notes

down

173

keith at keithtyler dot com ¶

14 years ago


For the record, the underlying function to uniqid() appears to be roughly as follows:

$m=microtime(true);
sprintf("%8x%05x\n",floor($m),($m-floor($m))*1000000);

In other words, first 8 hex chars = Unixtime, last 5 hex chars = microseconds. This is why it has microsecond precision. Also, it provides a means by which to reverse-engineer the time when a uniqid was generated: 

date("r",hexdec(substr(uniqid(),0,8)));

Increasingly as you go further down the string, the number becomes "more unique" over time, with the exception of digit 9, where numeral prevalence is 0..3>4>5..f, because of the difference between 10^6 and 16^5 (this is presumably true for the remaining digits as well but much less noticeable).

down

135

hackan at gmail dot com ¶

8 years ago


Seriously, avoid using this function. Here's an example of why:

<?php
for($i=0;$i<20;$i++) {
    echo uniqid(), PHP_EOL;
}
?>

Output:

5819f3ad1c0ce
5819f3ad1c0ce
5819f3ad1c0ce
5819f3ad1c0ce
5819f3ad1c0ce
5819f3ad1c0ce
5819f3ad1c0ce
5819f3ad1c0ce
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6
5819f3ad1c4b6

As you can see, using it w/ a DB can cause the creation of documents with repeated ID's. You should instead opt for:

<?php
function uniqidReal($lenght = 13) {
    // uniqid gives 13 chars, but you could adjust it to your needs.
    if (function_exists("random_bytes")) {
        $bytes = random_bytes(ceil($lenght / 2));
    } elseif (function_exists("openssl_random_pseudo_bytes")) {
        $bytes = openssl_random_pseudo_bytes(ceil($lenght / 2));
    } else {
        throw new Exception("no cryptographically secure random function available");
    }
    return substr(bin2hex($bytes), 0, $lenght);
}
?>

Test:
<?php
for($i=0;$i<20;$i++) {
    echo uniqid(), "\t", uniqidReal(), PHP_EOL;
}
?>

Output:

5819fa0b63be3   9f39aa0ecd89d
5819fa0b70ed3   2c0735cabfcce
5819fa0b712bb   15e45d1ca1e90
5819fa0b712bb   89593dc230eb3
5819fa0b712bb   449795704aeef
5819fa0b712bb   b046877b80ac9
5819fa0b712bb   0a6fa0ae3ec7b
5819fa0b712bb   ba2f3f4d6afe0
5819fa0b712bb   af03cfac83fd6
5819fa0b712bb   eb9c3c6d475c0
5819fa0b712bb   edfbbf59d5e1b
5819fa0b712bb   500dca18888d4
5819fa0b716a3   4f5a40ef715f1
5819fa0b716a3   154e42b616825
5819fa0b716a3   a879a22663c9b
5819fa0b716a3   ea7c044ddda8a
5819fa0b716a3   2c81a44dc674e
5819fa0b716a3   bb32f37304fd9
5819fa0b716a3   30cdf6c0317d7
5819fa0b716a3   d25f529d126ae

down

142

Andrew Moore ¶

14 years ago


The following class generates VALID RFC 4211 COMPLIANT Universally Unique IDentifiers (UUID) version 3, 4 and 5.

Version 3 and 5 UUIDs are named based. They require a namespace (another valid UUID) and a value (the name). Given the same namespace and name, the output is always the same.

Version 4 UUIDs are pseudo-random.

UUIDs generated below validates using OSSP UUID Tool, and output for named-based UUIDs are exactly the same. This is a pure PHP implementation.

<?php
class UUID {
  public static function v3($namespace, $name) {
    if(!self::is_valid($namespace)) return false;

    // Get hexadecimal components of namespace
    $nhex = str_replace(array('-','{','}'), '', $namespace);

    // Binary Value
    $nstr = '';

    // Convert Namespace UUID to bits
    for($i = 0; $i < strlen($nhex); $i+=2) {
      $nstr .= chr(hexdec($nhex[$i].$nhex[$i+1]));
    }

    // Calculate hash value
    $hash = md5($nstr . $name);

    return sprintf('%08s-%04s-%04x-%04x-%12s',

      // 32 bits for "time_low"
      substr($hash, 0, 8),

      // 16 bits for "time_mid"
      substr($hash, 8, 4),

      // 16 bits for "time_hi_and_version",
      // four most significant bits holds version number 3
      (hexdec(substr($hash, 12, 4)) & 0x0fff) | 0x3000,

      // 16 bits, 8 bits for "clk_seq_hi_res",
      // 8 bits for "clk_seq_low",
      // two most significant bits holds zero and one for variant DCE1.1
      (hexdec(substr($hash, 16, 4)) & 0x3fff) | 0x8000,

      // 48 bits for "node"
      substr($hash, 20, 12)
    );
  }

  public static function v4() {
    return sprintf('%04x%04x-%04x-%04x-%04x-%04x%04x%04x',

      // 32 bits for "time_low"
      mt_rand(0, 0xffff), mt_rand(0, 0xffff),

      // 16 bits for "time_mid"
      mt_rand(0, 0xffff),

      // 16 bits for "time_hi_and_version",
      // four most significant bits holds version number 4
      mt_rand(0, 0x0fff) | 0x4000,

      // 16 bits, 8 bits for "clk_seq_hi_res",
      // 8 bits for "clk_seq_low",
      // two most significant bits holds zero and one for variant DCE1.1
      mt_rand(0, 0x3fff) | 0x8000,

      // 48 bits for "node"
      mt_rand(0, 0xffff), mt_rand(0, 0xffff), mt_rand(0, 0xffff)
    );
  }

  public static function v5($namespace, $name) {
    if(!self::is_valid($namespace)) return false;

    // Get hexadecimal components of namespace
    $nhex = str_replace(array('-','{','}'), '', $namespace);

    // Binary Value
    $nstr = '';

    // Convert Namespace UUID to bits
    for($i = 0; $i < strlen($nhex); $i+=2) {
      $nstr .= chr(hexdec($nhex[$i].$nhex[$i+1]));
    }

    // Calculate hash value
    $hash = sha1($nstr . $name);

    return sprintf('%08s-%04s-%04x-%04x-%12s',

      // 32 bits for "time_low"
      substr($hash, 0, 8),

      // 16 bits for "time_mid"
      substr($hash, 8, 4),

      // 16 bits for "time_hi_and_version",
      // four most significant bits holds version number 5
      (hexdec(substr($hash, 12, 4)) & 0x0fff) | 0x5000,

      // 16 bits, 8 bits for "clk_seq_hi_res",
      // 8 bits for "clk_seq_low",
      // two most significant bits holds zero and one for variant DCE1.1
      (hexdec(substr($hash, 16, 4)) & 0x3fff) | 0x8000,

      // 48 bits for "node"
      substr($hash, 20, 12)
    );
  }

  public static function is_valid($uuid) {
    return preg_match('/^\{?[0-9a-f]{8}\-?[0-9a-f]{4}\-?[0-9a-f]{4}\-?'.
                      '[0-9a-f]{4}\-?[0-9a-f]{12}\}?$/i', $uuid) === 1;
  }
}

// Usage
// Named-based UUID.

$v3uuid = UUID::v3('1546058f-5a25-4334-85ae-e68f2a44bbaf', 'SomeRandomString');
$v5uuid = UUID::v5('1546058f-5a25-4334-85ae-e68f2a44bbaf', 'SomeRandomString');

// Pseudo-random UUID

$v4uuid = UUID::v4();
?>

down

rommel at rommelsantor dot com ¶

13 years ago


The php5-uuid functions could definitely use some documentation to clarify how they should be used, but here's what I've gleaned by examining the OSSP source code (found here: http://ossp-uuid.sourcearchive.com/documentation/1.5.1-1ubuntu1/php_2uuid_8c-source.html).

The uuid_make() function takes two arguments when generating v1 or v4, but four arguments are required when generating v3 or v5. The first two arguments have been demonstrated below and are straightforward, so I'll skip to the as-yet non-described arguments.

The third argument to uuid_make() is: $namespace
  - this is a secondary resource created with uuid_create(); it is apparently used to generate an internal UUID, which is used as the namespace of the output UUID

The fourth argument to uuid_make() is: $url
  - this is the value that is to be hashed (MD5 for v3, SHA-1 for v5); it may be any string or even null

Here's a simple class illustrating the proper usage (note that if php5-uuid is not installed on your system, each function call will just return false):

<?php
class UUID {
  /**
   * Generates version 1: MAC address
   */
  public static function v1() {
    if (!function_exists('uuid_create'))
      return false;

    uuid_create(&$context);
    uuid_make($context, UUID_MAKE_V1);
    uuid_export($context, UUID_FMT_STR, &$uuid);
    return trim($uuid);
  }

  /**
   * Generates version 3 UUID: MD5 hash of URL
   */
  public static function v3($i_url) {
    if (!function_exists('uuid_create'))
      return false;

    if (!strlen($i_url))
      $i_url = self::v1();

    uuid_create(&$context);
    uuid_create(&$namespace);

    uuid_make($context, UUID_MAKE_V3, $namespace, $i_url);
    uuid_export($context, UUID_FMT_STR, &$uuid);
    return trim($uuid);
  }

  /**
   * Generates version 4 UUID: random
   */
  public static function v4() {
    if (!function_exists('uuid_create'))
      return false;

    uuid_create(&$context);

    uuid_make($context, UUID_MAKE_V4);
    uuid_export($context, UUID_FMT_STR, &$uuid);
    return trim($uuid);
  }

  /**
   * Generates version 5 UUID: SHA-1 hash of URL
   */
  public static function v5($i_url) {
    if (!function_exists('uuid_create'))
      return false;

    if (!strlen($i_url))
      $i_url = self::v1();

    uuid_create(&$context);
    uuid_create(&$namespace);

    uuid_make($context, UUID_MAKE_V5, $namespace, $i_url);
    uuid_export($context, UUID_FMT_STR, &$uuid);
    return trim($uuid);
  }
}
?>

And here's a demonstration:

<?php
for ($i = 1; $i <= 3; ++$i) {
  echo 'microtime = ' . microtime(true) . '<br/>';
  echo "V1 UUID: " . UUID::v1() . '<br/>';
  echo "V3 UUID of URL='abc': " . UUID::v3('abc') . '<br/>';
  echo "V4 UUID: " . UUID::v4() . '<br/>';
  echo "V5 UUID of URL=null: " . UUID::v5(null) . '<br/>';
  echo '<hr/>';
}
?>

And the output:

microtime = 1306620716.0457
V1 UUID: 7fddae8e-8977-11e0-bc11-003048c3b1f2
V3 UUID of URL='abc': 522ec739-ca63-3ec5-b082-08ce08ad65e2
V4 UUID: b3851ec7-4871-4527-92b5-ef5616bae1e6
V5 UUID of URL=null: e129f27c-5103-5c5c-844b-cdf0a15e160d
-------------------
microtime = 1306620716.0465
V1 UUID: 7fddb83e-8977-11e0-9e6e-003048c3b1f2
V3 UUID of URL='abc': 522ec739-ca63-3ec5-b082-08ce08ad65e2
V4 UUID: 7e78fe0d-59b8-4637-af7f-e88d221a7d1e
V5 UUID of URL=null: e129f27c-5103-5c5c-844b-cdf0a15e160d
-------------------
microtime = 1306620716.0467
V1 UUID: 7fddbfb4-8977-11e0-a2bc-003048c3b1f2
V3 UUID of URL='abc': 522ec739-ca63-3ec5-b082-08ce08ad65e2
V4 UUID: 12a940c7-0f3f-46a1-bb5f-bdd602e10654
V5 UUID of URL=null: e129f27c-5103-5c5c-844b-cdf0a15e160d

As you can see, the calls to v3() always return the same UUID because the same URL parameter, "abc", is always supplied. The same goes for the v5() function which is always supplied a null URL.

The v4() UUIDs are always entirely different because they are (pseudo)random. And the v1() calls are very similar but just slightly different because it's based on the computer's MAC address and the current time.

down

nodkz at mail dot ru ¶

16 years ago


I use such UUID (it not RFC!!!)
(server_id)-(clientIP)-(unixtime)-(milliseconds)-(random)

 I can easyly determine which server at which time and who initiate creating of object.

<?php

$u=uuid();   // 0001-7f000001-478c8000-4801-47242987
echo $u;
echo "<br>";
print_r(uuidDecode($u)); // Array ( [serverID] => 0001 [ip] => 127.0.0.1 [unixtime] => 1200390144 [micro] => 0.28126525878906 ) 

function uuid($serverID=1)
{
    $t=explode(" ",microtime());
    return sprintf( '%04x-%08s-%08s-%04s-%04x%04x',
        $serverID,
        clientIPToHex(),
        substr("00000000".dechex($t[1]),-8),   // get 8HEX of unixtime
        substr("0000".dechex(round($t[0]*65536)),-4), // get 4HEX of microtime
        mt_rand(0,0xffff), mt_rand(0,0xffff));
}

function uuidDecode($uuid) {
    $rez=Array();
    $u=explode("-",$uuid);
    if(is_array($u)&&count($u)==5) {
        $rez=Array(
            'serverID'=>$u[0],
            'ip'=>clientIPFromHex($u[1]),
            'unixtime'=>hexdec($u[2]),
            'micro'=>(hexdec($u[3])/65536)
        );
    }
    return $rez;
}

function clientIPToHex($ip="") {
    $hex="";
    if($ip=="") $ip=getEnv("REMOTE_ADDR");
    $part=explode('.', $ip);
    for ($i=0; $i<=count($part)-1; $i++) {
        $hex.=substr("0".dechex($part[$i]),-2);
    }
    return $hex;
}

function clientIPFromHex($hex) {
    $ip="";
    if(strlen($hex)==8) {
        $ip.=hexdec(substr($hex,0,2)).".";
        $ip.=hexdec(substr($hex,2,2)).".";
        $ip.=hexdec(substr($hex,4,2)).".";
        $ip.=hexdec(substr($hex,6,2));
    }
    return $ip;
}

?>

down

Enrico Pallazzo ¶

14 years ago


Running:

rand_uniqid(9007199254740989);

will return 'PpQXn7COf' and:

rand_uniqid('PpQXn7COf', true);

will return '9007199254740989'

---

If you want the rand_uniqid to be at least 6 letter long, use the $pad_up = 6 argument

---

You can support even more characters (making the resulting rand_uniqid even smaller) by adding characters to the $index var at the top of the function body.

---

<?php
function rand_uniqid($in, $to_num = false, $pad_up = false, $passKey = null)
{
    $index = "abcdefghijklmnopqrstuvwxyz0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";
    if ($passKey !== null) {
        // Although this function's purpose is to just make the
        // ID short - and not so much secure,
        // you can optionally supply a password to make it harder
        // to calculate the corresponding numeric ID

        for ($n = 0; $n<strlen($index); $n++) {
            $i[] = substr( $index,$n ,1);
        }

        $passhash = hash('sha256',$passKey);
        $passhash = (strlen($passhash) < strlen($index))
            ? hash('sha512',$passKey)
            : $passhash;

        for ($n=0; $n < strlen($index); $n++) {
            $p[] =  substr($passhash, $n ,1);
        }

        array_multisort($p,  SORT_DESC, $i);
        $index = implode($i);
    }

    $base  = strlen($index);

    if ($to_num) {
        // Digital number  <<--  alphabet letter code
        $in  = strrev($in);
        $out = 0;
        $len = strlen($in) - 1;
        for ($t = 0; $t <= $len; $t++) {
            $bcpow = bcpow($base, $len - $t);
            $out   = $out + strpos($index, substr($in, $t, 1)) * $bcpow;
        }

        if (is_numeric($pad_up)) {
            $pad_up--;
            if ($pad_up > 0) {
                $out -= pow($base, $pad_up);
            }
        }
        $out = sprintf('%F', $out);
        $out = substr($out, 0, strpos($out, '.'));
    } else {
        // Digital number  -->>  alphabet letter code
        if (is_numeric($pad_up)) {
            $pad_up--;
            if ($pad_up > 0) {
                $in += pow($base, $pad_up);
            }
        }

        $out = "";
        for ($t = floor(log($in, $base)); $t >= 0; $t--) {
            $bcp = bcpow($base, $t);
            $a   = floor($in / $bcp) % $base;
            $out = $out . substr($index, $a, 1);
            $in  = $in - ($a * $bcp);
        }
        $out = strrev($out); // reverse
    }

    return $out;
}

echo rand_uniqid(1);
?>

down

DimeCadmium ¶

5 years ago


Most of these notes are grossly wrong.

First: you probably shouldn't be using uniqid at all. I can't think of a good reason to use it. Maybe to use as an identifier for a particular run of your script in logging, where you don't particularly care if there's a collision? I don't know.

Second: don't rely on the output of uniqid to be, in any way, unique. This will break if you use it twice in quick succession, or if you have two different users at the same time, or if the phase of the moon is wrong.

Third: to nodkz who wrote the clientIPToHex function - you reinvented so many wheels. bin2hex(inet_pton($ip)) will do the same thing.

Fourth: do not use random bytes for a unique identifier (like hackan suggests). That only serves to virtually guarantee that you will eventually have a collision. (You can use random bytes, and get another set of random bytes if there's a collision... but it's a lot easier to just use a sequential identifier like you should, possibly along with a random key)

down

Anonymous ¶

13 years ago


Prefix can be useful, for instance, if you generate identifiers simultaneously on several hosts that might happen to generate the identifier at the same microsecond.
So we can include the hostname / servername in the id.
<?php
echo uniqid(php_uname('n'), true);
// Output: darkstar4dfa8c27aea106.40781203
?>

down

redtraider at gmail dot com ¶

12 years ago


I use this function to generate microsoft-compatible GUID's.

<?php
 public function create_guid($namespace = '') {     
    static $guid = '';
    $uid = uniqid("", true);
    $data = $namespace;
    $data .= $_SERVER['REQUEST_TIME'];
    $data .= $_SERVER['HTTP_USER_AGENT'];
    $data .= $_SERVER['LOCAL_ADDR'];
    $data .= $_SERVER['LOCAL_PORT'];
    $data .= $_SERVER['REMOTE_ADDR'];
    $data .= $_SERVER['REMOTE_PORT'];
    $hash = strtoupper(hash('ripemd128', $uid . $guid . md5($data)));
    $guid = '{' .   
            substr($hash,  0,  8) . 
            '-' .
            substr($hash,  8,  4) .
            '-' .
            substr($hash, 12,  4) .
            '-' .
            substr($hash, 16,  4) .
            '-' .
            substr($hash, 20, 12) .
            '}';
    return $guid;
  }
?>

down

php at dot dot dot metehanarslan dot com ¶

8 years ago


Here is my approach to generate short string unique id. I am just increasing base of UUID. If you need even shorter keys you may add more chars to $index . When $index gets longer $out will be shorter.

<?php
function struuid($entropy)
{
    $s=uniqid("",$entropy);
    $num= hexdec(str_replace(".","",(string)$s));
    $index = '1234567890ABCDEFGHIJKLMNOPQRSTUVWXYZ';
    $base= strlen($index);
    $out = '';
        for($t = floor(log10($num) / log10($base)); $t >= 0; $t--) {
            $a = floor($num / pow($base,$t));
            $out = $out.substr($index,$a,1);
            $num = $num-($a*pow($base,$t));
        }
    return $out;
}

echo struuid(false); //Return sample: F4518NTQTQ
echo struuid(true);  //Return sample: F451FAHSUCD90N6YNRBQHLZ9E1W

?>

down

ken at smallboxsoftware ¶

17 years ago


Just to note this function is fairly slow, and can bring your script to a crawl if it is in a loop. Strangely if you run it as uniqid('', true) it runs much more quickly

down

ciantic at NOSPAM dot oksidi dot com ¶

10 years ago


If you are storing the value to database it might be more efficient to use more letters than in hexadecimal, for me I decided 0-9a-z is good enough.

Notice also that many databases are case insensitive by default so using capitals maybe unwise anyway:

<?php

function uniqid_base36($more_entropy=false) {
    $s = uniqid('', $more_entropy);
    if (!$more_entropy)
        return base_convert($s, 16, 36);
        
    $hex = substr($s, 0, 13);
    $dec = $s[13] . substr($s, 15); // skip the dot
    return base_convert($hex, 16, 36) . base_convert($dec, 10, 36);
}

echo uniqid_base36() . "\n"; // eb98xzzhq7
echo uniqid_base36(true) . "\n"; // eb98xzzhr8dervre

?>

String length may vary with this method.

down

david at ramaboo dot com ¶

15 years ago


Another UUID function. This time using /dev/random
<?php
/**
     * @brief Generates a Universally Unique IDentifier, version 4.
     * 
     * This function generates a truly random UUID. The built in CakePHP String::uuid() function
     * is not cryptographically secure. You should uses this function instead.
     * 
     * @see http://tools.ietf.org/html/rfc4122#section-4.4
     * @see http://en.wikipedia.org/wiki/UUID
     * @return string A UUID, made up of 32 hex digits and 4 hyphens.
     */
    public function uuidSecure() {
        
        $pr_bits = null;
        $fp = @fopen('/dev/urandom','rb');
        if ($fp !== false) {
            $pr_bits .= @fread($fp, 16);
            @fclose($fp);
        } else {
            $this->cakeError('randomNumber');
        }
        
        $time_low = bin2hex(substr($pr_bits,0, 4));
        $time_mid = bin2hex(substr($pr_bits,4, 2));
        $time_hi_and_version = bin2hex(substr($pr_bits,6, 2));
        $clock_seq_hi_and_reserved = bin2hex(substr($pr_bits,8, 2));
        $node = bin2hex(substr($pr_bits,10, 6));
        
        /**
         * Set the four most significant bits (bits 12 through 15) of the
         * time_hi_and_version field to the 4-bit version number from
         * Section 4.1.3.
         * @see http://tools.ietf.org/html/rfc4122#section-4.1.3
         */
        $time_hi_and_version = hexdec($time_hi_and_version);
        $time_hi_and_version = $time_hi_and_version >> 4;
        $time_hi_and_version = $time_hi_and_version | 0x4000;
        
        /**
         * Set the two most significant bits (bits 6 and 7) of the
         * clock_seq_hi_and_reserved to zero and one, respectively.
         */
        $clock_seq_hi_and_reserved = hexdec($clock_seq_hi_and_reserved);
        $clock_seq_hi_and_reserved = $clock_seq_hi_and_reserved >> 2;
        $clock_seq_hi_and_reserved = $clock_seq_hi_and_reserved | 0x8000;
        
        return sprintf('%08s-%04s-%04x-%04x-%012s',
            $time_low, $time_mid, $time_hi_and_version, $clock_seq_hi_and_reserved, $node);
    }
?>

down

php at ryanmckeel dot com ¶

14 years ago


Calls to uuid_make that use the constants UUID_MAKE_V5 or UUID_MAKE_V3 (using Debian package php5-uuid available June 2010) will not work with only two variables.

I could not find good documentation, so I read some source code and figured out that this would work:

uuid_create(&$v5);
//uuid_make($v5, UUID_MAKE_V5);
uuid_make($v5, UUID_MAKE_V5,$v5,$uniqid());
uuid_export($v5, UUID_FMT_STR, &$v5String);

Please use at your own risk.  This may not be the best way to give this param variables, but it at least makes it work what appears to be properly (generating unique ID's).

I imagine that UUID_MAKE_V3 is similar in what it needs.

  Ryan

down

Albert Rolando ¶

1 year ago


Don't try to overthink UUID V4 generation…

<?php
function generateUuid4() {
    $randomBytes = bin2hex(random_bytes(16));
    return sprintf("%s-%s-%s-%s-%s",
        substr($randomBytes, 0, 8),
        substr($randomBytes, 8, 4),
        substr($randomBytes, 12, 4),
        substr($randomBytes, 16, 4),
        substr($randomBytes, 20)
    );
}
?>

down

-3

lance_rushing at hotmail dot com ¶

17 years ago


wooshoofoo, the reason mimec is calling mt_rand multiple times is because the largest number mt_rand can produce is 2^31 (2147483647, as reported by mt_getrandmax() on my server).  RFC 4122 requires a 128 bit value.

Also they are not "4 digit sequeces", but 4 digit hexadecimal numbers.  16^4 == 2^16.

mimec's limiting each random result to 2^16 avoids problem of PHP's 2^32 integer max (http://php.net/manual/en/language.types.integer.php).

If you want to call mt_rand fewer times:  mimec's version calls mt_rand 8 times ( 16 bits * 8 = 128 bits ).  You *could* call mt_rand 5 times ( 31 bits + 31 bits + 31 bits + 31 bits + 4 bits = 128 bits ).  But then you would have keep all your values as strings.

Something like: 

<?php
/**
 * Another (ugly) "random or pseudo-random" version of RFC 4122
 *
 * This version calls mt_rand() the fewest possible times.
 * if mt_getrandmax() == 2^31 then this will call mt_rand() 5 times YMMV
 *
 * Personally, I would use mimec's version
 * To handle the large values, we'll keep everything as strings.
 *
 * @return string
 */
function uuid() {    

     // Generate 128 bit random sequence 
     $randmax_bits = strlen(base_convert(mt_getrandmax(), 10, 2));  // how many bits is mt_getrandmax()
     $x = '';
     while (strlen($x) < 128) {
         $maxbits = (128 - strlen($x) < $randmax_bits) ? 128 - strlen($x) :  $randmax_bits;
         $x .= str_pad(base_convert(mt_rand(0, pow(2,$maxbits)), 10, 2), $maxbits, "0", STR_PAD_LEFT);
     }

     // break into fields
     $a = array();
     $a['time_low_part'] = substr($x, 0, 32);
     $a['time_mid'] = substr($x, 32, 16);
     $a['time_hi_and_version'] = substr($x, 48, 16);
     $a['clock_seq'] = substr($x, 64, 16);
     $a['node_part'] =  substr($x, 80, 48);
     
     // Apply bit masks for "random or pseudo-random" version per RFC
     $a['time_hi_and_version'] = substr_replace($a['time_hi_and_version'], '0100', 0, 4);
     $a['clock_seq'] = substr_replace($a['clock_seq'], '10', 0, 2);

    // Format output
    return sprintf('%s-%s-%s-%s-%s',
        str_pad(base_convert($a['time_low_part'], 2, 16), 8, "0", STR_PAD_LEFT),
        str_pad(base_convert($a['time_mid'], 2, 16), 4, "0", STR_PAD_LEFT),
        str_pad(base_convert($a['time_hi_and_version'], 2, 16), 4, "0", STR_PAD_LEFT),
        str_pad(base_convert($a['clock_seq'], 2, 16), 4, "0", STR_PAD_LEFT),
        str_pad(base_convert($a['node_part'], 2, 16), 12, "0", STR_PAD_LEFT));
}

?>

However, I think mimec's version is much more elegant.

add a note