Knuth Morris Pratt Pattern Matching in C++

#include <stdio.h>
#include <vector>
#include <algorithm>
#include <stdexcept>
#include <cstdint>
#include <limits>
// Knuth Morris Pratt Pattern Matching in C++
//
// PURPOSE
// KMP avoids rechecking text tokens by remembering which matched pattern suffix is also a
// prefix. The prefix table tells where to resume after a mismatch.
//
// METHOD
// Build the longest proper prefix/suffix table for the pattern. During matching, fall back
// through that table without moving the text position backward. A full matched prefix gives
// the first occurrence.
//
// COMPLEXITY
// O(n+m) time and O(m) auxiliary space.
//
// CONTRACT
// Inputs are integer token sequences. Empty pattern matches at 0. Returns first matching
// index or -1. KMP does not define Unicode tokenization.
// 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: 2.
// 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 kmp_match(const std::vector<int>& text, const std::vector<int>& pattern) {
    alg_require(text.size() <= 1000000);
    alg_require(pattern.size() <= 1000000);
    int n = (int)text.size();
    int m = (int)pattern.size();
    alg_require(((n >= 0) && (n <= 1000000)));
    alg_require(((m >= 0) && (m <= 1000000)));
    alg_require(((int)pattern.size() == m));
    alg_require(((int)text.size() == n));
    if ((m == 0)) {
        return 0;
    }
    std::vector<int> prefix(alg_size(alg_add(m, 1)), 0);
    int i = 1;
    int j = 0;
    while ((i < m)) {
        while (((j > 0) && (pattern[i] != pattern[j]))) {
            j = prefix[alg_sub(j, 1)];
        }
        if ((pattern[i] == pattern[j])) {
            j = alg_add(j, 1);
        }
        prefix[i] = j;
        i = alg_add(i, 1);
    }
    i = 0;
    j = 0;
    while ((i < n)) {
        while (((j > 0) && (text[i] != pattern[j]))) {
            j = prefix[alg_sub(j, 1)];
        }
        if ((text[i] == pattern[j])) {
            j = alg_add(j, 1);
        }
        if ((j == m)) {
            return alg_add(alg_sub(i, m), 1);
        }
        i = alg_add(i, 1);
    }
    return alg_sub(0, 1);
}

int main(void) {
    std::vector<int> text = {1, 2, 1, 2, 3};
    std::vector<int> pattern = {1, 2, 3};
    int result = kmp_match(text, pattern);
    printf("%d\n", result);
    return 0;
}