├── .gitignore
├── README.md
├── composer.json
└── src
    ├── Bucket
        ├── Bucket.php
        └── ModBucket.php
    ├── Digest
        ├── BKDRDigest.php
        ├── DEKDigest.php
        ├── DJBDigest.php
        ├── Digest.php
        ├── ELFDigest.php
        ├── FNVDigest.php
        ├── JSDigest.php
        ├── PJWDigest.php
        └── SDBMDigest.php
    ├── Filter.php
    └── FilterInterface.php


/.gitignore:
--------------------------------------------------------------------------------
1 | .idea
2 | .php_cs.cache
3 | composer.lock
4 | vendor
5 | test.php


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | ## Bloom Filter
 2 | 
 3 | ## install
 4 | ```shell script
 5 | composer require xiaker/bloom-filter
 6 | ```
 7 | 
 8 | ## usage
 9 | 
10 | ```php
11 | ...
12 | 
13 | use Xiaker\Bloom\Bucket\ModBucket;
14 | use Xiaker\Bloom\Digest\BKDRDigest;
15 | use Xiaker\Bloom\Digest\DEKDigest;
16 | use Xiaker\Bloom\Digest\DJBDigest;
17 | use Xiaker\Bloom\Digest\ELFDigest;
18 | use Xiaker\Bloom\Digest\FNVDigest;
19 | use Xiaker\Bloom\Digest\JSDigest;
20 | use Xiaker\Bloom\Digest\PJWDigest;
21 | use Xiaker\Bloom\Digest\SDBMDigest;
22 | use Xiaker\Bloom\Filter;
23 | 
24 | $redis = new Redis();
25 | $redis->connect('127.0.0.1');
26 | 
27 | $digests = [ // you can select several or all of them
28 |     new BKDRDigest(),
29 |     new DEKDigest(),
30 |     new DJBDigest(),
31 |     new ELFDigest(),
32 |     new FNVDigest(),
33 |     new JSDigest(),
34 |     new PJWDigest(),
35 |     new SDBMDigest(),
36 | ];
37 | 
38 | $filter = new Filter(new ModBucket(24), $redis, ...$digests);
39 | 
40 | $filter->add('fatrbaby');
41 | ...
42 | 
43 | if ($filter->exists('fatrbaby')) {
44 |     ...
45 | }
46 | ```


--------------------------------------------------------------------------------
/composer.json:
--------------------------------------------------------------------------------
 1 | {
 2 |     "name": "xiaker/bloom-filter",
 3 |     "description": "A bloom filter wrote by PHP & storaged by Redis",
 4 |     "type": "library",
 5 |     "license": "MIT",
 6 |     "authors": [
 7 |         {
 8 |             "name": "fatrbaby",
 9 |             "email": "fatrbaby@qq.com"
10 |         }
11 |     ],
12 |     "require": {
13 |         "php": ">=7.2",
14 |         "ext-redis": "*"
15 |     },
16 |     "require-dev": {
17 |         "symfony/finder": "^5.1"
18 |     },
19 |     "autoload": {
20 |         "psr-4": {
21 |             "Xiaker\\Bloom\\": "src"
22 |         }
23 |     },
24 |     "autoload-dev": {
25 |         "psr-4": {
26 |             "Tests\\": "tests"
27 |         }
28 |     }
29 | }
30 | 


--------------------------------------------------------------------------------
/src/Bucket/Bucket.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Bucket;
 6 | 
 7 | abstract class Bucket
 8 | {
 9 |     /**
10 |      * @var int
11 |      */
12 |     protected $total;
13 | 
14 |     public function __construct(int $total)
15 |     {
16 |         $this->total = $total;
17 |     }
18 | 
19 |     protected function warp($value)
20 |     {
21 |         // PSR-16
22 |         return '___xiaker.'.$value;
23 |     }
24 | 
25 |     abstract public function dispatch(string $key);
26 | }
27 | 


--------------------------------------------------------------------------------
/src/Bucket/ModBucket.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Bucket;
 6 | 
 7 | class ModBucket extends Bucket
 8 | {
 9 |     public function dispatch(string $key)
10 |     {
11 |         return $this->warp(crc32($key) % $this->total);
12 |     }
13 | }
14 | 


--------------------------------------------------------------------------------
/src/Digest/BKDRDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * From "The C Programming Language" author: Brian Kernighan & Dennis Ritchie
 9 |  */
10 | class BKDRDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         $seed = 131;  // 31 131 1313 13131 131313 etc..
15 |         $hash = 0;
16 | 
17 |         if (null === $length) {
18 |             $length = strlen($string);
19 |         }
20 | 
21 |         for ($i = 0; $i < $length; ++$i) {
22 |             $hash = (int) (($hash * $seed) + ord($string[$i]));
23 |         }
24 | 
25 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
26 |     }
27 | }
28 | 


--------------------------------------------------------------------------------
/src/Digest/DEKDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * Invented by Donald E. Knuth
 9 |  */
10 | class DEKDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         if (null === $length) {
15 |             $length = strlen($string);
16 |         }
17 | 
18 |         $hash = $length;
19 | 
20 |         for ($i = 0; $i < $length; ++$i) {
21 |             $hash = (($hash << 5) ^ ($hash >> 27)) ^ ord($string[$i]);
22 |         }
23 | 
24 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
25 |     }
26 | }
27 | 


--------------------------------------------------------------------------------
/src/Digest/DJBDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * Invented by Daniel J. Bernstein
 9 |  */
10 | class DJBDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         $hash = 5381;
15 | 
16 |         if (null === $length) {
17 |             $length = strlen($string);
18 |         }
19 | 
20 |         for ($i = 0; $i < $length; ++$i) {
21 |             $hash = (int) (($hash << 5) + $hash) + ord($string[$i]);
22 |         }
23 | 
24 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
25 |     }
26 | }
27 | 


--------------------------------------------------------------------------------
/src/Digest/Digest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | interface Digest
 8 | {
 9 |     public function hash($string, $length = null);
10 | }
11 | 


--------------------------------------------------------------------------------
/src/Digest/ELFDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * Like PjwDigest
 9 |  */
10 | class ELFDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         $hash = 0;
15 | 
16 |         if (null === $length) {
17 |             $length = strlen($string);
18 |         }
19 | 
20 |         for ($i = 0; $i < $length; ++$i) {
21 |             $hash = ($hash << 4) + ord($string[$i]);
22 |             $x = $hash & 0xF0000000;
23 | 
24 |             if (0 != $x) {
25 |                 $hash ^= ($x >> 24);
26 |             }
27 | 
28 |             $hash &= ~$x;
29 |         }
30 | 
31 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
32 |     }
33 | }
34 | 


--------------------------------------------------------------------------------
/src/Digest/FNVDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * @see http://www.isthe.com/chongo/tech/comp/fnv/
 9 |  */
10 | class FNVDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         $prime = 16777619; // prime 2^24 + 2^8 + 0x93 = 16777619
15 |         $hash = 2166136261; // offset
16 | 
17 |         if (null === $length) {
18 |             $length = strlen($string);
19 |         }
20 | 
21 |         for ($i = 0; $i < $length; ++$i) {
22 |             $hash = (int) ($hash * $prime) % 0xFFFFFFFF;
23 |             $hash ^= ord($string[$i]);
24 |         }
25 | 
26 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
27 |     }
28 | }
29 | 


--------------------------------------------------------------------------------
/src/Digest/JSDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * Invented by Justin Sobel.
 9 |  */
10 | class JSDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         $hash = 1315423911;
15 | 
16 |         if (null === $length) {
17 |             $length = strlen($string);
18 |         }
19 | 
20 |         for ($i = 0; $i < $length; ++$i) {
21 |             $hash ^= (($hash << 5) + ord($string[$i]) + ($hash >> 2));
22 |         }
23 | 
24 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
25 |     }
26 | }
27 | 


--------------------------------------------------------------------------------
/src/Digest/PJWDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * Invented by Peter J. Weinberger of AT＆T
 9 |  */
10 | class PJWDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         $bitsInUnsignedInt = 4 * 8; //（unsigned int）（sizeof（unsigned int）* 8）;
15 |         $threeQuarters = ($bitsInUnsignedInt * 3) / 4;
16 |         $oneEighth = $bitsInUnsignedInt / 8;
17 |         $highBits = 0xFFFFFFFF << (int) ($bitsInUnsignedInt - $oneEighth);
18 |         $hash = 0;
19 | 
20 |         if (null === $length) {
21 |             $length = strlen($string);
22 |         }
23 | 
24 |         for ($i = 0; $i < $length; ++$i) {
25 |             $hash = ($hash << (int) ($oneEighth)) + ord($string[$i]);
26 |         }
27 | 
28 |         $test = $hash & $highBits;
29 | 
30 |         if (0 != $test) {
31 |             $hash = (($hash ^ ($test >> (int) ($threeQuarters))) & (~$highBits));
32 |         }
33 | 
34 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
35 |     }
36 | }
37 | 


--------------------------------------------------------------------------------
/src/Digest/SDBMDigest.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom\Digest;
 6 | 
 7 | /**
 8 |  * From SDBM project
 9 |  */
10 | class SDBMDigest implements Digest
11 | {
12 |     public function hash($string, $length = null)
13 |     {
14 |         $hash = 0;
15 | 
16 |         if (null === $length) {
17 |             $length = strlen($string);
18 |         }
19 | 
20 |         for ($i = 0; $i < $length; ++$i) {
21 |             $hash = (int) (ord($string[$i]) + ($hash << 6) + ($hash << 16) - $hash);
22 |         }
23 | 
24 |         return ($hash % 0xFFFFFFFF) & 0xFFFFFFFF;
25 |     }
26 | }
27 | 


--------------------------------------------------------------------------------
/src/Filter.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom;
 6 | 
 7 | use Redis;
 8 | use Xiaker\Bloom\Bucket\Bucket;
 9 | use Xiaker\Bloom\Digest\Digest;
10 | 
11 | class Filter implements FilterInterface
12 | {
13 |     /**
14 |      * @var Bucket
15 |      */
16 |     protected $bucket;
17 | 
18 |     /**
19 |      * @var Digest[]
20 |      */
21 |     protected $digests;
22 | 
23 |     /**
24 |      * @var Redis
25 |      */
26 |     protected $redis;
27 | 
28 |     public function __construct(Bucket $bucket, Redis $redis, Digest ...$digests)
29 |     {
30 |         $this->bucket = $bucket;
31 |         $this->redis = $redis;
32 |         $this->digests = $digests;
33 |     }
34 | 
35 |     public function add(string $string)
36 |     {
37 |         $bucket = $this->bucket->dispatch($string);
38 | 
39 |         $pipe = $this->redis->multi();
40 |         foreach ($this->digests as $digest) {
41 |             $pipe->setBit($bucket, $digest->hash($string), true);
42 |         }
43 | 
44 |         return $pipe->exec();
45 |     }
46 | 
47 |     public function exists(string $string)
48 |     {
49 |         $bucket = $this->bucket->dispatch($string);
50 | 
51 |         $pipe = $this->redis->multi();
52 |         $length = strlen($string);
53 | 
54 |         foreach ($this->digests as $digest) {
55 |             $pipe->getBit($bucket, $digest->hash($string, $length));
56 |         }
57 | 
58 |         $payloads = $pipe->exec();
59 | 
60 |         foreach ($payloads as $bit) {
61 |             if (0 == $bit) {
62 |                 return false;
63 |             }
64 |         }
65 | 
66 |         return true;
67 |     }
68 | }
69 | 


--------------------------------------------------------------------------------
/src/FilterInterface.php:
--------------------------------------------------------------------------------
 1 | <?php
 2 | 
 3 | declare(strict_types=1);
 4 | 
 5 | namespace Xiaker\Bloom;
 6 | 
 7 | interface FilterInterface
 8 | {
 9 |     public function add(string $string);
10 | 
11 |     public function exists(string $string);
12 | }
13 | 


--------------------------------------------------------------------------------