結果

問題 No.3078 Difference Sum Query
ユーザー apricity
提出日時 2025-03-28 21:48:03
言語 C++17
(gcc 13.3.0 + boost 1.87.0)
結果
AC  
実行時間 479 ms / 2,000 ms
コード長 13,202 bytes
コンパイル時間 1,709 ms
コンパイル使用メモリ 145,656 KB
実行使用メモリ 35,528 KB
最終ジャッジ日時 2025-03-28 21:48:15
合計ジャッジ時間 11,504 ms
ジャッジサーバーID
(参考情報)
judge5 / judge4
このコードへのチャレンジ
(要ログイン)
ファイルパターン 結果
sample AC * 2
other AC * 26
権限があれば一括ダウンロードができます

ソースコード

diff #

#ifdef LOCAL
#include "template.hpp"
#else
#include<iostream>
#include<string>
#include<vector>
#include<algorithm>
#include<numeric>
#include<cmath>
#include<utility>
#include<tuple>
#include<array>
#include<cstdint>
#include<cstdio>
#include<iomanip>
#include<map>
#include<set>
#include<unordered_map>
#include<unordered_set>
#include<queue>
#include<stack>
#include<deque>
#include<bitset>
#include<cctype>
#include<chrono>
#include<random>
#include<cassert>
#include<cstddef>
#include<iterator>
#include<string_view>
#include<type_traits>
#include<functional>

using namespace std;

namespace io {

template <typename T, typename U>
istream &operator>>(istream &is, pair<T, U> &p) {
    is >> p.first >> p.second;
    return is;
}
template <typename T>
istream &operator>>(istream &is, vector<T> &v) {
    for (auto &x : v) is >> x;
    return is;
}
template <typename T, size_t N = 0>
istream &operator>>(istream &is, array<T, N> &v) {
    for (auto &x : v) is >> x;
    return is;
}
template <size_t N = 0, typename T>
istream& cin_tuple_impl(istream &is, T &t) {
    if constexpr (N < std::tuple_size<T>::value) {
        auto &x = std::get<N>(t);
        is >> x;
        cin_tuple_impl<N + 1>(is, t);
    }
    return is;
}
template <class... T>
istream &operator>>(istream &is, tuple<T...> &t) {
    return cin_tuple_impl(is, t);
}

template<typename T, typename U>
ostream &operator<<(ostream &os, const pair<T, U> &p) {
    os << p.first << " " << p.second;
    return os;
}
template<typename T>
ostream &operator<<(ostream &os, const vector<T> &v) {
    int s = (int)v.size();
    for (int i = 0; i < s; i++) os << (i ? " " : "") << v[i];
    return os;
}
template<typename T, size_t N>
ostream &operator<<(ostream &os, const array<T, N> &v) {
    size_t n = v.size();
    for (size_t i = 0; i < n; i++) {
        if (i) os << " ";
        os << v[i];
    }
    return os;
}
template <size_t N = 0, typename T>
ostream& cout_tuple_impl(ostream &os, const T &t) {
    if constexpr (N < std::tuple_size<T>::value) {
        if constexpr (N > 0) os << " ";
        const auto &x = std::get<N>(t);
        os << x;
        cout_tuple_impl<N + 1>(os, t);
    }
    return os;
}
template <class... T>
ostream &operator<<(ostream &os, const tuple<T...> &t) {
    return cout_tuple_impl(os, t);
}

void in() {}
template<typename T, class... U>
void in(T &t, U &...u) {
    cin >> t;
    in(u...);
}
void out() { cout << "\n"; }
template<typename T, class... U, char sep = ' '>
void out(const T &t, const U &...u) {
    cout << t;
    if (sizeof...(u)) cout << sep;
    out(u...);
}
void outr() {}
template<typename T, class... U, char sep = ' '>
void outr(const T &t, const U &...u) {
    cout << t;
    outr(u...);
}

void __attribute__((constructor)) _c() {
    ios_base::sync_with_stdio(false);
    cin.tie(nullptr);
    cout << fixed << setprecision(15);
}
} // namespace io

using io::in;
using io::out;
using io::outr;

#define SHOW(x) static_cast<void>(0)

using ll = long long;
using D = double;
using LD = long double;
using P = pair<ll, ll>;
using u8 = uint8_t;
using u16 = uint16_t;
using u32 = uint32_t;
using u64 = uint64_t;
using i128 = __int128;
using u128 = unsigned __int128;
using vi = vector<ll>;
template <class T> using vc = vector<T>;
template <class T> using vvc = vector<vc<T>>;
template <class T> using vvvc = vector<vvc<T>>;
template <class T> using vvvvc = vector<vvvc<T>>;
template <class T> using vvvvvc = vector<vvvvc<T>>;
#define vv(type, name, h, ...) \
  vector<vector<type>> name(h, vector<type>(__VA_ARGS__))
#define vvv(type, name, h, w, ...)   \
  vector<vector<vector<type>>> name( \
      h, vector<vector<type>>(w, vector<type>(__VA_ARGS__)))
#define vvvv(type, name, a, b, c, ...)       \
  vector<vector<vector<vector<type>>>> name( \
      a, vector<vector<vector<type>>>(       \
             b, vector<vector<type>>(c, vector<type>(__VA_ARGS__))))
template<typename T> using PQ = priority_queue<T,vector<T>>;
template<typename T> using minPQ = priority_queue<T, vector<T>, greater<T>>;

#define rep1(a)          for(ll i = 0; i < a; i++)
#define rep2(i, a)       for(ll i = 0; i < a; i++)
#define rep3(i, a, b)    for(ll i = a; i < b; i++)
#define rep4(i, a, b, c) for(ll i = a; i < b; i += c)
#define overload4(a, b, c, d, e, ...) e
#define rep(...) overload4(__VA_ARGS__, rep4, rep3, rep2, rep1)(__VA_ARGS__)
#define rrep1(a)          for(ll i = (a)-1; i >= 0; i--)
#define rrep2(i, a)       for(ll i = (a)-1; i >= 0; i--)
#define rrep3(i, a, b)    for(ll i = (b)-1; i >= a; i--)
#define rrep4(i, a, b, c) for(ll i = (b)-1; i >= a; i -= c)
#define rrep(...) overload4(__VA_ARGS__, rrep4, rrep3, rrep2, rrep1)(__VA_ARGS__)
#define for_subset(t, s) for (ll t = (s); t >= 0; t = (t == 0 ? -1 : (t - 1) & (s)))
#define ALL(v) v.begin(), v.end()
#define RALL(v) v.rbegin(), v.rend()
#define UNIQUE(v) v.erase( unique(v.begin(), v.end()), v.end() )
#define SZ(v) ll(v.size())
#define MIN(v) *min_element(ALL(v))
#define MAX(v) *max_element(ALL(v))
#define LB(c, x) distance((c).begin(), lower_bound(ALL(c), (x)))
#define UB(c, x) distance((c).begin(), upper_bound(ALL(c), (x)))
template <typename T, typename U>
T SUM(const vector<U> &v) {
    T res = 0;
    for(auto &&a : v) res += a;
    return res;
}
template <typename T>
vector<pair<T,int>> RLE(const vector<T> &v) {
    if (v.empty()) return {};
    T cur = v.front();
    int cnt = 1;
    vector<pair<T,int>> res;
    for (int i = 1; i < (int)v.size(); i++) {
        if (cur == v[i]) cnt++;
        else {
            res.emplace_back(cur, cnt);
            cnt = 1; cur = v[i];
        }
    }
    res.emplace_back(cur, cnt);
    return res;
}
template<class T, class S>
inline bool chmax(T &a, const S &b) { return (a < b ? a = b, true : false); }
template<class T, class S>
inline bool chmin(T &a, const S &b) { return (a > b ? a = b, true : false); }
void YESNO(bool flag) { out(flag ? "YES" : "NO"); }
void yesno(bool flag) { out(flag ? "Yes" : "No"); }

int popcnt(int x) { return __builtin_popcount(x); }
int popcnt(u32 x) { return __builtin_popcount(x); }
int popcnt(ll x) { return __builtin_popcountll(x); }
int popcnt(u64 x) { return __builtin_popcountll(x); }
int popcnt_sgn(int x) { return (__builtin_parity(x) & 1 ? -1 : 1); }
int popcnt_sgn(u32 x) { return (__builtin_parity(x) & 1 ? -1 : 1); }
int popcnt_sgn(ll x) { return (__builtin_parityl(x) & 1 ? -1 : 1); }
int popcnt_sgn(u64 x) { return (__builtin_parityl(x) & 1 ? -1 : 1); }
int highbit(int x) { return (x == 0 ? -1 : 31 - __builtin_clz(x)); }
int highbit(u32 x) { return (x == 0 ? -1 : 31 - __builtin_clz(x)); }
int highbit(ll x) { return (x == 0 ? -1 : 63 - __builtin_clzll(x)); }
int highbit(u64 x) { return (x == 0 ? -1 : 63 - __builtin_clzll(x)); }
int lowbit(int x) { return (x == 0 ? -1 : __builtin_ctz(x)); }
int lowbit(u32 x) { return (x == 0 ? -1 : __builtin_ctz(x)); }
int lowbit(ll x) { return (x == 0 ? -1 : __builtin_ctzll(x)); }
int lowbit(u64 x) { return (x == 0 ? -1 : __builtin_ctzll(x)); }

template <typename T>
T get_bit(T x, int k) { return x >> k & 1; }
template <typename T>
T set_bit(T x, int k) { return x | T(1) << k; }
template <typename T>
T reset_bit(T x, int k) { return x & ~(T(1) << k); }
template <typename T>
T flip_bit(T x, int k) { return x ^ T(1) << k; }

template <typename T>
T popf(deque<T> &que) { T a = que.front(); que.pop_front(); return a; }
template <typename T>
T popb(deque<T> &que) { T a = que.back(); que.pop_back(); return a; }
template <typename T>
T pop(queue<T> &que) { T a = que.front(); que.pop(); return a; }
template <typename T>
T pop(stack<T> &que) { T a = que.top(); que.pop(); return a; }
template <typename T>
T pop(PQ<T> &que) { T a = que.top(); que.pop(); return a; }
template <typename T>
T pop(minPQ<T> &que) { T a = que.top(); que.pop(); return a; }

template <typename F>
ll binary_search(F check, ll ok, ll ng, bool check_ok = true) {
    if (check_ok) assert(check(ok));
    while (abs(ok -  ng) > 1) {
        ll mid = (ok + ng) / 2;
        (check(mid) ? ok : ng) = mid;
    }
    return ok;
}
template <typename F>
double binary_search_real(F check, double ok, double ng, int iter = 60) {
    for (int _ = 0; _ < iter; _++) {
        double mid = (ok + ng) / 2;
        (check(mid) ? ok : ng) = mid;
    }
    return (ok + ng) / 2;
}

// max x s.t. b*x <= a
ll div_floor(ll a, ll b) {
    assert(b != 0);
    if (b < 0) a = -a, b = -b;
    return a / b - (a % b < 0);
}
// max x s.t. b*x < a
ll div_under(ll a, ll b) {
    assert(b != 0);
    if (b < 0) a = -a, b = -b;
    return a / b - (a % b <= 0);
}
// min x s.t. b*x >= a
ll div_ceil(ll a, ll b) {
    assert(b != 0);
    if (b < 0) a = -a, b = -b;
    return a / b + (a % b > 0);
}
// min x s.t. b*x > a
ll div_over(ll a, ll b) {
    assert(b != 0);
    if (b < 0) a = -a, b = -b;
    return a / b + (a % b >= 0);
}
// x = a mod b (b > 0), 0 <= x < b
ll modulo(ll a, ll b) {
    assert(b > 0);
    ll c = a % b;
    return c < 0 ? c + b : c;
}
// (q,r) s.t. a = b*q + r, 0 <= r < b (b > 0)
// div_floor(a,b), modulo(a,b)
pair<ll,ll> divmod(ll a, ll b) {
    ll q = div_floor(a,b);
    return {q, a - b*q};
}
#endif

#line 1 "structure/wavelet/succinct-indexable-dictionary.hpp"
/**
 * @brief Succinct Indexable Dictionary(完備辞書)
 */
struct SuccinctIndexableDictionary {
  size_t length;
  size_t blocks;
  vector<unsigned> bit, sum;

  SuccinctIndexableDictionary() = default;

  SuccinctIndexableDictionary(size_t length)
      : length(length), blocks((length + 31) >> 5) {
    bit.assign(blocks, 0U);
    sum.assign(blocks, 0U);
  }

  void set(int k) { bit[k >> 5] |= 1U << (k & 31); }

  void build() {
    sum[0] = 0U;
    for (int i = 1; i < blocks; i++) {
      sum[i] = sum[i - 1] + __builtin_popcount(bit[i - 1]);
    }
  }

  bool operator[](int k) { return (bool((bit[k >> 5] >> (k & 31)) & 1)); }

  int rank(int k) {
    return (sum[k >> 5] +
            __builtin_popcount(bit[k >> 5] & ((1U << (k & 31)) - 1)));
  }

  int rank(bool val, int k) { return (val ? rank(k) : k - rank(k)); }
};
#line 2 "structure/wavelet/wavelet-matrix-rectangle-sum.hpp"

/*
 * @brief Wavelet Matrix Rectangle Sum
 *
 */
template <typename T, int MAXLOG, typename D>
struct WaveletMatrixRectangleSum {
  size_t length;
  SuccinctIndexableDictionary matrix[MAXLOG];
  vector<D> ds[MAXLOG];
  int mid[MAXLOG];

  WaveletMatrixRectangleSum() = default;

  WaveletMatrixRectangleSum(const vector<T> &v, const vector<D> &d)
      : length(v.size()) {
    assert(v.size() == d.size());
    vector<int> l(length), r(length), ord(length);
    iota(begin(ord), end(ord), 0);
    for (int level = MAXLOG - 1; level >= 0; level--) {
      matrix[level] = SuccinctIndexableDictionary(length + 1);
      int left = 0, right = 0;
      for (int i = 0; i < length; i++) {
        if (((v[ord[i]] >> level) & 1)) {
          matrix[level].set(i);
          r[right++] = ord[i];
        } else {
          l[left++] = ord[i];
        }
      }
      mid[level] = left;
      matrix[level].build();
      ord.swap(l);
      for (int i = 0; i < right; i++) {
        ord[left + i] = r[i];
      }
      ds[level].resize(length + 1);
      ds[level][0] = D();
      for (int i = 0; i < length; i++) {
        ds[level][i + 1] = ds[level][i] + d[ord[i]];
      }
    }
  }

  pair<int, int> succ(bool f, int l, int r, int level) {
    return {matrix[level].rank(f, l) + mid[level] * f,
            matrix[level].rank(f, r) + mid[level] * f};
  }

  // count d[i] s.t. (l <= i < r) && (v[i] < upper)
  D rect_sum(int l, int r, T upper) {
    D ret = 0;
    for (int level = MAXLOG - 1; level >= 0; level--) {
      bool f = ((upper >> level) & 1);
      if (f)
        ret += ds[level][matrix[level].rank(false, r)] -
               ds[level][matrix[level].rank(false, l)];
      tie(l, r) = succ(f, l, r, level);
    }
    return ret;
  }

  D rect_sum(int l, int r, T lower, T upper) {
    return rect_sum(l, r, upper) - rect_sum(l, r, lower);
  }
};

template <typename T, int MAXLOG, typename D>
struct CompressedWaveletMatrixRectangleSum {
  WaveletMatrixRectangleSum<int, MAXLOG, D> mat;
  vector<T> ys;

  CompressedWaveletMatrixRectangleSum(const vector<T> &v, const vector<D> &d)
      : ys(v) {
    sort(begin(ys), end(ys));
    ys.erase(unique(begin(ys), end(ys)), end(ys));
    vector<int> t(v.size());
    for (int i = 0; i < v.size(); i++) t[i] = get(v[i]);
    mat = WaveletMatrixRectangleSum<int, MAXLOG, D>(t, d);
  }

  inline int get(const T &x) {
    return lower_bound(begin(ys), end(ys), x) - begin(ys);
  }

  D rect_sum(int l, int r, T upper) { return mat.rect_sum(l, r, get(upper)); }

  D rect_sum(int l, int r, T lower, T upper) {
    return mat.rect_sum(l, r, get(lower), get(upper));
  }
};

int main() {
    int n,q; in(n,q);
    vector<ll> a(n); in(a);
    vector<ll> f(n,1);
    CompressedWaveletMatrixRectangleSum<ll,17,ll> w0(a,f);
    CompressedWaveletMatrixRectangleSum<ll,17,ll> w1(a,a);
    vector<ll> p(n+1);
    rep(i,n) p[i+1] = p[i]+a[i];
    while(q--){
        int l,r; ll x; in(l,r,x); l--;
        ll s0 = w0.rect_sum(l,r,0,x);
        ll s1 = w1.rect_sum(l,r,0,x);
        ll ans = x*s0 - s1;
        ans += (p[r]-p[l]-s1) - x*(r-l-s0);
        out(ans);
    }
}
0