Fenwick Prefix Sum in C++

#include <stdio.h>
#include <vector>
#include <algorithm>
#include <stdexcept>
#include <cstdint>
#include <limits>
// Fenwick Prefix Sum in C++
//
// PURPOSE
// A Fenwick prefix query combines disjoint stored partial sums covering the requested
// prefix.
//
// METHOD
// Starting from the exclusive prefix endpoint, add the current partial sum and clear its
// lowest set bit to move to the preceding block.
//
// COMPLEXITY
// O(log n) time, O(1) auxiliary space.
//
// CONTRACT
// tree length is n+1, 0<=end<=n. Returns sum of original values in [0,end). Empty prefix
// returns zero.
// 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: 6.
// 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 fenwick_sum(const std::vector<int>& tree, int end) {
    alg_require(tree.size() <= 1000000);
    int n = alg_sub((int)tree.size(), 1);
    alg_require(((n >= 0) && (n <= 1000000)));
    alg_require(((int)tree.size() == alg_add(n, 1)));
    alg_require(((end >= 0) && (end <= n)));
    int total = 0;
    while ((end > 0)) {
        total = alg_add(total, tree[end]);
        end = alg_sub(end, (end & alg_sub(0, end)));
    }
    return total;
}

int main(void) {
    std::vector<int> tree = {0, 1, 3, 3, 10};
    int end = 3;
    int result = fenwick_sum(tree, end);
    printf("%d\n", result);
    return 0;
}