题意:有一个长为\(n\)的排列\(p\),设\(S_i=\sum_{j=1}^{i-1}p_j\cdot[p_j<p_i]\),给出\(S\),要求还原出\(p\)。保证有解,\(n\leq 2\times 10^5\)。
考虑倒序将\(S\)还原为全\(0\)的序列,从小到大依次考虑插入每个数的影响。假设在位置\(x\)插入\(i\),显然此时\(S_x=0\),且会使得位置\(x\)右侧的每一个未插入数字的\(S_y\)都减去\(i\)。因此对于第\(i\)个数,唯一合法的位置就是所有\(S_x=0\)的位置中最右侧的那个。由于保证有解,维护最小值,在线段树上二分即可。
Code:
#include <cstdio>
#include <cctype>
#include <cassert>
#include <cstring>
#include <iostream>
#include <algorithm>
#define R register
#define ll long long
using namespace std;
const int N = 210000, K = N << 2;
const ll inf = 1e18;
int n, lt[K], rt[K], p[N];
ll a[N], minVal[K], tag[K];
inline void pushdown(int);
template <class T> inline void read(T &x) {
x = 0;
char ch = getchar(), w = 0;
while (!isdigit(ch))
w = (ch == '-'), ch = getchar();
while (isdigit(ch))
x = (x << 1) + (x << 3) + (ch ^ 48), ch = getchar();
x = w ? -x : x;
return;
}
void build(int k) {
if (lt[k] == rt[k]) {
minVal[k] = a[lt[k]];
return;
}
int mid = (lt[k] + rt[k]) >> 1, l = k << 1, r = (k << 1) + 1;
lt[l] = lt[k], rt[l] = mid, lt[r] = mid + 1, rt[r] = rt[k];
build(l), build(r), minVal[k] = min(minVal[l], minVal[r]);
return;
}
void modify(int k, int x, int y, ll w) {
if (lt[k] >= x && rt[k] <= y) {
minVal[k] += w, tag[k] += w;
return;
}
pushdown(k);
int mid = (lt[k] + rt[k]) >> 1, l = k << 1, r = (k << 1) + 1;
if (x <= mid) modify(l, x, y, w);
if (y > mid) modify(r, x, y, w);
minVal[k] = min(minVal[l], minVal[r]);
return;
}
int query(int k) {
if (lt[k] == rt[k]) {
minVal[k] = inf;
return lt[k];
}
int l = k << 1, r = (k << 1) + 1, ret;
pushdown(k), ret = query(minVal[r] ? l : r);
minVal[k] = min(minVal[l], minVal[r]);
return ret;
}
inline void pushdown(int k) {
int l = k << 1, r = (k << 1) + 1;
if (tag[k]) {
modify(l, lt[k], rt[k], tag[k]);
modify(r, lt[k], rt[k], tag[k]);
tag[k] = 0;
}
return;
}
int main() {
read(n);
for (R int i = 1; i <= n; ++i)
read(a[i]);
lt[1] = 1, rt[1] = n, build(1);
for (R int i = 1; i <= n; ++i) {
int pos = query(1);
p[pos] = i, modify(1, pos, n, -i);
}
for (R int i = 1; i <= n; ++i)
printf("%d ", p[i]);
return 0;
}
手机扫一扫
移动阅读更方便
你可能感兴趣的文章