Bellman-Ford permits negative edge weights and detects negative cycles reachable from the source.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 | <?php declare(strict_types=1); // Bellman Ford Shortest Path in PHP // // PURPOSE // Bellman-Ford permits negative edge weights and detects negative cycles reachable from the // source. Repeated edge relaxation finds progressively longer optimal paths. // // METHOD // Relax every edge up to n-1 times, stopping early if nothing changes. Perform one extra // pass to detect a reachable negative cycle. Guard unreachable endpoints before adding // weights. // // COMPLEXITY // O(VE) time and O(V) auxiliary space. // // CONTRACT // Edge arrays from_vertex, to_vertex, weight each have m entries; n >= 1. Finite distances // and all relaxed sums stay strictly between -1,000,000,000 and 1,000,000,000. Returns // distance, INF if unreachable, or -INF for any reachable negative cycle (even if it cannot // reach target). // 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 bellman_ford(array $from_vertex, array $to_vertex, array $weight, int $n, int $source, int $target): int { alg_require(array_is_list($from_vertex)); alg_require(array_is_list($to_vertex)); alg_require(array_is_list($weight)); $m = count($from_vertex); alg_require((($n >= 0) && ($n <= 1000000))); alg_require((($m >= 0) && ($m <= 1000000))); alg_require((count($from_vertex) == $m)); alg_require((count($to_vertex) == $m)); alg_require((count($weight) == $m)); alg_require((($n >= 1) && ($source >= 0) && ($source < $n))); alg_require((($target >= 0) && ($target < $n))); for ($alg_i = 0; $alg_i < count($from_vertex); $alg_i++) { $alg_value = $from_vertex[$alg_i]; alg_require((($alg_value >= 0) && ($alg_value < $n))); } for ($alg_i = 0; $alg_i < count($to_vertex); $alg_i++) { $alg_value = $to_vertex[$alg_i]; alg_require((($alg_value >= 0) && ($alg_value < $n))); } $distance = array_fill(0, alg_size($n), 0); $i = 0; $edge = 0; $iteration = 0; $changed = 0; $u = 0; $v = 0; while (($i < $n)) { $distance[$i] = 1000000000; $i = alg_add($i, 1); } $distance[$source] = 0; while (($iteration < alg_sub($n, 1))) { $changed = 0; $edge = 0; while (($edge < $m)) { $u = $from_vertex[$edge]; $v = $to_vertex[$edge]; if ((($distance[$u] != 1000000000) && (alg_add($distance[$u], $weight[$edge]) < $distance[$v]))) { $distance[$v] = alg_add($distance[$u], $weight[$edge]); $changed = 1; } $edge = alg_add($edge, 1); } if (($changed == 0)) { break; } $iteration = alg_add($iteration, 1); } $edge = 0; while (($edge < $m)) { $u = $from_vertex[$edge]; $v = $to_vertex[$edge]; if ((($distance[$u] != 1000000000) && (alg_add($distance[$u], $weight[$edge]) < $distance[$v]))) { return alg_sub(0, 1000000000); } $edge = alg_add($edge, 1); } return $distance[$target]; } $from_vertex = [0, 1, 0]; $to_vertex = [1, 2, 2]; $weight = [4, -2, 9]; $n = 3; $source = 0; $target = 2; $result = bellman_ford($from_vertex, $to_vertex, $weight, $n, $source, $target); echo $result, PHP_EOL; |