Rabin Karp Pattern Matching in PHP

<?php
declare(strict_types=1);
// Rabin Karp Pattern Matching in PHP
//
// PURPOSE
// Rabin-Karp uses a rolling hash to filter candidate matches, then compares tokens to
// guarantee correctness despite hash collisions.
//
// METHOD
// Hash the pattern and first window modulo a small prime. At each position verify matching
// hashes, then remove the old token and append the new one.
//
// COMPLEXITY
// Expected O(n+m) time on suitable data; O(nm) worst case due to verification. O(1)
// auxiliary space.
//
// CONTRACT
// Tokens are integers in [0,255]. Empty pattern matches at zero. Returns first index or -1;
// not a cryptographic hash.
// Public APIs infer array lengths except in C, which accepts pointer/length pairs. Vertex
// counts and matrix dimensions remain explicit. Arrays must contain signed 32-bit integers.
// Sorted order, correct tree topology, and caller-owned data-structure invariants are
// preconditions; checking them on every query would change query complexity.
// Scalar and shape checks are included. Value-range checks are included where needed to
// prevent invalid indexing or nontermination. Allocation sizes are capped at 4,000,000
// workspace entries. Arithmetic outside Python raises an error on 32-bit overflow; Python
// uses arbitrary precision. Native allocation failures propagate. In-place outputs may be
// partly updated after arithmetic failure. These are single-threaded, in-memory
// implementations.
//
// LANGUAGE DETAILS
// PHP 8.1+ on a 64-bit runtime: packed arrays, strict scalar types, read-only inputs by
// value using copy-on-write, references only for outputs, and checked 32-bit arithmetic.
// Invalid input and overflow throw exceptions. PHP arrays have higher overhead than native
// packed buffers in C/Java.
//
// EXAMPLE
// Expected return: 2.
// Run: php example.php
// Tests for this exact source artifact run before publication. The accompanying source
// archive includes unit cases, runtime results, and source hashes.

function alg_require(bool $valid): void { if (!$valid) throw new InvalidArgumentException("invalid algorithm input"); }
function alg_checked(int|float $v): int { if ($v < -2147483648 || $v > 2147483647 || floor($v) != $v) throw new OverflowException("32-bit result overflow"); return (int)$v; }
function alg_add(int $a,int $b): int { return alg_checked($a+$b); }
function alg_sub(int $a,int $b): int { return alg_checked($a-$b); }
function alg_mul(int $a,int $b): int { return alg_checked($a*$b); }
function alg_size(int $n): int { alg_require($n>=0 && $n<=4000000); return $n; }

function rabin_karp(array $text, array $pattern): int {
    alg_require(array_is_list($text));
    alg_require(array_is_list($pattern));
    $n = count($text);
    $m = count($pattern);
    alg_require((($n >= 0) && ($n <= 1000000)));
    alg_require((($m >= 0) && ($m <= 1000000)));
    alg_require((count($pattern) == $m));
    alg_require((count($text) == $n));
    for ($alg_i = 0; $alg_i < count($text); $alg_i++) {
        $alg_value = $text[$alg_i];
        alg_require((($alg_value >= 0) && ($alg_value <= 255)));
    }
    for ($alg_i = 0; $alg_i < count($pattern); $alg_i++) {
        $alg_value = $pattern[$alg_i];
        alg_require((($alg_value >= 0) && ($alg_value <= 255)));
    }
    if (($m == 0)) {
        return 0;
    }
    if (($m > $n)) {
        return alg_sub(0, 1);
    }
    $power = 1;
    $target_hash = 0;
    $window_hash = 0;
    $i = 0;
    $j = 0;
    while (($i < $m)) {
        $target_hash = (alg_add(alg_mul($target_hash, 256), $pattern[$i]) % 10007);
        $window_hash = (alg_add(alg_mul($window_hash, 256), $text[$i]) % 10007);
        if (($i < alg_sub($m, 1))) {
            $power = (alg_mul($power, 256) % 10007);
        }
        $i = alg_add($i, 1);
    }
    $i = 0;
    while (($i <= alg_sub($n, $m))) {
        if (($target_hash == $window_hash)) {
            $j = 0;
            while ((($j < $m) && ($text[alg_add($i, $j)] == $pattern[$j]))) {
                $j = alg_add($j, 1);
            }
            if (($j == $m)) {
                return $i;
            }
        }
        if (($i < alg_sub($n, $m))) {
            $window_hash = (alg_add(alg_sub($window_hash, (alg_mul($text[$i], $power) % 10007)), 10007) % 10007);
            $window_hash = (alg_add(alg_mul($window_hash, 256), $text[alg_add($i, $m)]) % 10007);
        }
        $i = alg_add($i, 1);
    }
    return alg_sub(0, 1);
}

$text = [1, 2, 1, 2, 3];
$pattern = [1, 2, 3];
$result = rabin_karp($text, $pattern);
echo $result, PHP_EOL;