Problem Author: | Mohammad Nasser (MOAGEZA) |
Prepared by: | Mohammad Nasser (MOAGEZA) |
Edited by: | Hasan Saleh (MathModel) |
Solution
General Idea: We can only increase elements by using operations on same-parity subarrays. If any element a_i > x, it's impossible since we can only increase values.
Key Observation:
When we increment an element, we can pair it with a neighbor to minimize total operations. The strategy is to greedily process elements from smallest to largest.
Algorithm:
Sort elements by value while tracking their original positions.
For each element (starting from smallest):
Look at its left and right neighbors in the original array.
Choose the neighbor with smaller value (costs less to increment together).
Add the difference: (\text{neighbor value} - \text{current value}) to the answer.
Remove the current element from consideration.
Finally, add (x - \max(a)) to bring the maximum element to .
The minimum number of operations is:
The solution runs in \mathcal{O}(n \log n) time per test case due to sorting and set operations.
#include <bits/stdc++.h>
#define int long long
using namespace std;
void solve(){
int n, x, mx = 0, ans = 0;
cin >> n >> x;
vector<int> a(n);
set<int> s;
vector<pair<int, int>> v;
for(int i = 0; i < n; i++){
cin >> a[i];
mx = max(mx, a[i]);
s.insert(i);
v.push_back({a[i], i});
}
if(mx > x){
cout << -1;
return;
}
sort(v.begin(), v.end());
for(int i = 0; i < n - 1; i++){
int val = v[i].first, idx = v[i].second;
int res1 = 2e9, res2 = 2e9;
s.erase(idx);
auto it = s.upper_bound(idx);
if(it != s.end()){
res1 = a[*it];
}
if(it != s.begin()){
it--;
res2 = a[*it];
}
ans += min(res1, res2) - val;
}
ans += x - mx;
cout << ans;
}
signed main(){
ios::sync_with_stdio(false);
cin.tie(NULL);
int t;
cin >> t;
while(t--){
solve();
cout << '\n';
}
return 0;
}