Algorithms: String Algorithms

"use strict";
// Z Algorithm in JavaScript
//
// PURPOSE
// The Z array records how many tokens match the prefix at every position.
//
// METHOD
// Maintain the rightmost matching interval. Reuse a previously computed Z value when inside
// it, then extend only beyond its boundary.
//
// COMPLEXITY
// O(n) time and O(1) auxiliary space beyond caller-owned output.
//
// CONTRACT
// Integer token input; output has n entries. Defines Z[0]=0; returns 0.
// 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
// Node.js 18+: Int32Array inputs and workspaces provide compact integer storage. Automatic
// lengths and checked arithmetic reject values outside signed 32-bit. Invalid inputs or
// overflow throw RangeError.
//
// EXAMPLE
// Expected return: 0.
// Expected output after call: [0, 2, 1, 0, 1].
// Run: node example.js
// Tests for this exact source artifact run before publication. The accompanying source
// archive includes unit cases, runtime results, and source hashes.

function alg_require(valid) { if (!valid) throw new RangeError("invalid algorithm input"); }
function alg_checked(v) { if (!Number.isInteger(v) || v < -2147483648 || v > 2147483647) throw new RangeError("32-bit result overflow"); return v; }
function alg_add(a,b) { return alg_checked(a+b); }
function alg_sub(a,b) { return alg_checked(a-b); }
function alg_mul(a,b) { return alg_checked(a*b); }
function alg_size(n) { alg_require(Number.isInteger(n) && n>=0 && n<=4000000); return n; }

function z_array(a, output) {
    alg_require(a instanceof Int32Array);
    alg_require(output instanceof Int32Array);
    let n = a.length;
    alg_require(((n >= 0) && (n <= 1000000)));
    alg_require((a.length == n));
    alg_require((output.length == n));
    let left = 0;
    let right = 0;
    let i = 0;
    while ((i < n)) {
        output[i] = 0;
        i = alg_add(i, 1);
    }
    i = 1;
    while ((i < n)) {
        if ((i <= right)) {
            output[i] = Math.min(alg_add(alg_sub(right, i), 1), output[alg_sub(i, left)]);
        }
        while (((alg_add(i, output[i]) < n) && (a[output[i]] == a[alg_add(i, output[i])]))) {
            output[i] = alg_add(output[i], 1);
        }
        if ((alg_sub(alg_add(i, output[i]), 1) > right)) {
            left = i;
            right = alg_sub(alg_add(i, output[i]), 1);
        }
        i = alg_add(i, 1);
    }
    return 0;
}

let a = new Int32Array([1, 1, 1, 2, 1]);
let output = new Int32Array([0, 0, 0, 0, 0]);
let result = z_array(a, output);
console.log(result);
// Expected output: [0, 2, 1, 0, 1]
console.log(Array.from(output));

from bisect import bisect_left

# Z Algorithm in Python
#
# PURPOSE
# The Z array records how many tokens match the prefix at every position.
#
# METHOD
# Maintain the rightmost matching interval. Reuse a previously computed Z value when inside
# it, then extend only beyond its boundary.
#
# COMPLEXITY
# O(n) time and O(1) auxiliary space beyond caller-owned output.
#
# CONTRACT
# Integer token input; output has n entries. Defines Z[0]=0; returns 0.
# 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
# Python 3.9+: automatic lengths, native lists, arbitrary-precision arithmetic, and
# bisect_left for LIS. Invalid scalar/shape inputs raise ValueError. Integer element types
# and sorted/tree invariants remain caller preconditions.
#
# EXAMPLE
# Expected return: 0.
# Expected output after call: [0, 2, 1, 0, 1].
# Run: python example.py
# Tests for this exact source artifact run before publication. The accompanying source
# archive includes unit cases, runtime results, and source hashes.

def alg_require(condition):
    if not condition:
        raise ValueError("invalid algorithm input; see contract")
def alg_size(n):
    alg_require(0 <= n <= 4000000)
    return n

def z_array(a: list[int], output: list[int]) -> int:
    alg_require(isinstance(a, (list, tuple)) and len(a) <= 1000000)
    alg_require(isinstance(output, list) and len(output) <= 1000000)
    n: int = len(a)
    alg_require(n >= 0 and n <= 1000000)
    alg_require(len(a) == n)
    alg_require(len(output) == n)
    left: int = 0
    right: int = 0
    i: int = 0
    while i < n:
        output[i] = 0
        i += 1
    i = 1
    while i < n:
        if i <= right:
            output[i] = min(right - i + 1, output[i - left])
        while i + output[i] < n and a[output[i]] == a[i + output[i]]:
            output[i] += 1
        if i + output[i] - 1 > right:
            left = i
            right = i + output[i] - 1
        i += 1
    return 0

a = [1, 1, 1, 2, 1]
output = [0, 0, 0, 0, 0]
result = z_array(a, output)
print(result)
# Expected output: [0, 2, 1, 0, 1]
print(output)

#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>
#include <limits.h>
#include <errno.h>
// Z Algorithm in C
//
// PURPOSE
// The Z array records how many tokens match the prefix at every position.
//
// METHOD
// Maintain the rightmost matching interval. Reuse a previously computed Z value when inside
// it, then extend only beyond its boundary.
//
// COMPLEXITY
// O(n) time and O(1) auxiliary space beyond caller-owned output.
//
// CONTRACT
// Integer token input; output has n entries. Defines Z[0]=0; returns 0.
// 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
// C99: checked calloc/free workspace, const input pointers, explicit size_t lengths. Errors
// return INT_MIN with errno=EINVAL, ERANGE, or ENOMEM. Check errno before using a result.
// Input/output arrays must not alias. No variable-length stack arrays.
//
// EXAMPLE
// Expected return: 0.
// Expected output after call: [0, 2, 1, 0, 1].
// Run: gcc -std=c99 -O2 example.c -o example && ./example
// Tests for this exact source artifact run before publication. The accompanying source
// archive includes unit cases, runtime results, and source hashes.

static int alg_min(int a, int b) { return a < b ? a : b; }
static int alg_max(int a, int b) { return a > b ? a : b; }
static int alg_checked(int64_t v, int *error) {
    if (v < INT_MIN || v > INT_MAX) { *error = 1; return 0; }
    return (int)v;
}
static int alg_add(int a, int b, int *e) { return alg_checked((int64_t)a+b,e); }
static int alg_sub(int a, int b, int *e) { return alg_checked((int64_t)a-b,e); }
static int alg_mul(int a, int b, int *e) { return alg_checked((int64_t)a*b,e); }

int z_array(const int *a, size_t a_length, int *output, size_t output_length) {
    int alg_error = 0;
    int alg_result = INT_MIN;
    errno = 0;
    if (a_length > 1000000 || (a_length && !a)) { errno = EINVAL; alg_result = INT_MIN; goto alg_cleanup; }
    if (output_length > 1000000 || (output_length && !output)) { errno = EINVAL; alg_result = INT_MIN; goto alg_cleanup; }
    int n = (int)a_length;
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    if (!(((n >= 0) && (n <= 1000000)))) { errno = EINVAL; alg_result = INT_MIN; goto alg_cleanup; }
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    if (!(((int)a_length == n))) { errno = EINVAL; alg_result = INT_MIN; goto alg_cleanup; }
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    if (!(((int)output_length == n))) { errno = EINVAL; alg_result = INT_MIN; goto alg_cleanup; }
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    int left = 0;
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    int right = 0;
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    int i = 0;
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    while (1) {
        int alg_condition = (i < n);
        if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        if (!alg_condition) break;
        output[i] = 0;
        if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        i = alg_add(i, 1, &alg_error);
        if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    }
    i = 1;
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    while (1) {
        int alg_condition = (i < n);
        if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        if (!alg_condition) break;
        int alg_condition_1 = (i <= right);
        if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        if (alg_condition_1) {
            output[i] = alg_min(alg_add(alg_sub(right, i, &alg_error), 1, &alg_error), output[alg_sub(i, left, &alg_error)]);
            if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        }
        while (1) {
            int alg_condition = ((alg_add(i, output[i], &alg_error) < n) && (a[output[i]] == a[alg_add(i, output[i], &alg_error)]));
            if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
            if (!alg_condition) break;
            output[i] = alg_add(output[i], 1, &alg_error);
            if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        }
        int alg_condition_2 = (alg_sub(alg_add(i, output[i], &alg_error), 1, &alg_error) > right);
        if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        if (alg_condition_2) {
            left = i;
            if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
            right = alg_sub(alg_add(i, output[i], &alg_error), 1, &alg_error);
            if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
        }
        i = alg_add(i, 1, &alg_error);
        if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    }
    alg_result = 0;
    if (alg_error) { errno = ERANGE; alg_result = INT_MIN; goto alg_cleanup; }
    goto alg_cleanup;
alg_cleanup:
    return alg_result;
}

int main(void) {
    int a[] = {1, 1, 1, 2, 1};
    int output[] = {0, 0, 0, 0, 0};
    int result = z_array(a, 5, output, 5);
    printf("%d\n", result);
    // Expected output: [0, 2, 1, 0, 1]
    for (int alg_print = 0; alg_print < 5; alg_print++) printf("%d ", output[alg_print]);
    printf("\n");
    if (errno) { perror("algorithm"); return 1; }
    return 0;
}

#include <stdio.h>
#include <vector>
#include <algorithm>
#include <stdexcept>
#include <cstdint>
#include <limits>
// Z Algorithm in C++
//
// PURPOSE
// The Z array records how many tokens match the prefix at every position.
//
// METHOD
// Maintain the rightmost matching interval. Reuse a previously computed Z value when inside
// it, then extend only beyond its boundary.
//
// COMPLEXITY
// O(n) time and O(1) auxiliary space beyond caller-owned output.
//
// CONTRACT
// Integer token input; output has n entries. Defines Z[0]=0; returns 0.
// 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
// C++17: const vector references for inputs, mutable references for outputs, RAII workspace,
// standard lower_bound for LIS. Invalid inputs throw invalid_argument; overflow throws
// overflow_error; allocation failure propagates bad_alloc.
//
// EXAMPLE
// Expected return: 0.
// Expected output after call: [0, 2, 1, 0, 1].
// Run: g++ -std=c++17 -O2 example.cpp -o example && ./example
// Tests for this exact source artifact run before publication. The accompanying source
// archive includes unit cases, runtime results, and source hashes.

static int alg_min(int a, int b) { return a < b ? a : b; }
static int alg_max(int a, int b) { return a > b ? a : b; }
static void alg_require(bool valid) { if (!valid) throw std::invalid_argument("invalid algorithm input"); }
static int alg_checked(int64_t v) { if (v < INT32_MIN || v > INT32_MAX) throw std::overflow_error("32-bit result overflow"); return (int)v; }
static int alg_add(int a,int b) { return alg_checked((int64_t)a+b); }
static int alg_sub(int a,int b) { return alg_checked((int64_t)a-b); }
static int alg_mul(int a,int b) { return alg_checked((int64_t)a*b); }
static size_t alg_size(int n) { alg_require(n>=0 && n<=4000000); return (size_t)n; }

int z_array(const std::vector<int>& a, std::vector<int>& output) {
    alg_require(a.size() <= 1000000);
    alg_require(output.size() <= 1000000);
    int n = (int)a.size();
    alg_require(((n >= 0) && (n <= 1000000)));
    alg_require(((int)a.size() == n));
    alg_require(((int)output.size() == n));
    int left = 0;
    int right = 0;
    int i = 0;
    while ((i < n)) {
        output[i] = 0;
        i = alg_add(i, 1);
    }
    i = 1;
    while ((i < n)) {
        if ((i <= right)) {
            output[i] = alg_min(alg_add(alg_sub(right, i), 1), output[alg_sub(i, left)]);
        }
        while (((alg_add(i, output[i]) < n) && (a[output[i]] == a[alg_add(i, output[i])]))) {
            output[i] = alg_add(output[i], 1);
        }
        if ((alg_sub(alg_add(i, output[i]), 1) > right)) {
            left = i;
            right = alg_sub(alg_add(i, output[i]), 1);
        }
        i = alg_add(i, 1);
    }
    return 0;
}

int main(void) {
    std::vector<int> a = {1, 1, 1, 2, 1};
    std::vector<int> output = {0, 0, 0, 0, 0};
    int result = z_array(a, output);
    printf("%d\n", result);
    // Expected output: [0, 2, 1, 0, 1]
    for (int alg_print = 0; alg_print < 5; alg_print++) printf("%d ", output[alg_print]);
    printf("\n");
    return 0;
}

using System;
// Z Algorithm in C#
//
// PURPOSE
// The Z array records how many tokens match the prefix at every position.
//
// METHOD
// Maintain the rightmost matching interval. Reuse a previously computed Z value when inside
// it, then extend only beyond its boundary.
//
// COMPLEXITY
// O(n) time and O(1) auxiliary space beyond caller-owned output.
//
// CONTRACT
// Integer token input; output has n entries. Defines Z[0]=0; returns 0.
// 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
// C# / .NET 10: ReadOnlySpan<int> inputs, Span<int> outputs, primitive workspace arrays,
// Array.BinarySearch for LIS, and checked integer arithmetic. Invalid arguments and overflow
// throw exceptions.
//
// EXAMPLE
// Expected return: 0.
// Expected output after call: [0, 2, 1, 0, 1].
// Run: Use Program.cs in a net10.0 console project; dotnet run.
// Tests for this exact source artifact run before publication. The accompanying source
// archive includes unit cases, runtime results, and source hashes.

class Program {
    static void alg_require(bool valid) { if (!valid) throw new ArgumentException("invalid algorithm input"); }
    static int alg_size(int n) { alg_require(n>=0 && n<=4000000); return n; }

    static int z_array(ReadOnlySpan<int> a, Span<int> output) {
        checked {
            int n = a.Length;
            alg_require(((n >= 0) && (n <= 1000000)));
            alg_require((a.Length == n));
            alg_require((output.Length == n));
            int left = 0;
            int right = 0;
            int i = 0;
            while ((i < n)) {
                output[i] = 0;
                i = (i + 1);
            }
            i = 1;
            while ((i < n)) {
                if ((i <= right)) {
                    output[i] = Math.Min(((right - i) + 1), output[(i - left)]);
                }
                while ((((i + output[i]) < n) && (a[output[i]] == a[(i + output[i])]))) {
                    output[i] = (output[i] + 1);
                }
                if ((((i + output[i]) - 1) > right)) {
                    left = i;
                    right = ((i + output[i]) - 1);
                }
                i = (i + 1);
            }
            return 0;
        }
    }

    static void Main() {
        int[] a = new int[] {1, 1, 1, 2, 1};
        int[] output = new int[] {0, 0, 0, 0, 0};
        int result = z_array(a, output);
        Console.WriteLine(result);
        // Expected output: [0, 2, 1, 0, 1]
        Console.WriteLine("[" + string.Join(", ", output) + "]");
    }
}