## Beautiful Triplets

Warning: HackerRank has recently changed the constraints of the problem by allowing duplicates. This change invalidated a couple of our solutions. This is discussed in the post. Solutions We found many solutions to this challenge. Naive solution 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 bool isBeautiful(int i, int j, int k, int d, const vector<int>& arr) { int a = arr[j] - arr[i]; int b = arr[k] - arr[j]; return a == b && a == d; } int beautifulTriplets(int d, vector<int> arr) { int n = arr. [Read More]

Solutions This problem has two main solutions. The first fully exploits the structure of the constraints, the second is more general. Tradeoffs of both: we remove all the duplicates. In a different context this could be forbidden and then we would have to use extra space or another logic; we don’t really change the score table. Instead, in a real game it’s very likely that we need to update the current scores (e. [Read More]

## Find Kth Zero

Consider an array of integers A = [a1, a2, …, aN]. We can perform two types of queries on A: 1 k: For query type 1, find and print an integer denoting the index of the kth zero in array A on a new line; if no such index exists (i.e., there is no zero), print NO instead. p x: For query type 2, replace the element positioned at index p with integer x (i. [Read More]

## Impress the Interviewer

Alan has finally reached the final round of interviews at Gugol. He is one step from catching the job he loves and he has to impress the interviewer with a killer answer to the last question. The problem looks like very easy: given an array where all elements appear even number of times except one. All repeating occurrences of elements appear in pairs and these pairs are not adjacent (there cannot be more than two consecutive occurrences of any element). [Read More]

## Minimum Loss

Solutions The naive solution is quadratic and it’s too slow for a few test cases. We show it here, for completeness: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 #include <iterator>#include <climits>#include <vector>#include <iostream>#include <algorithm>using namespace std; int main() { int n; cin >> n; vector<int> v(n); copy_n(istream_iterator<int>(cin), n, begin(v)); auto minLoss = INT_MAX; for (auto i=0; i<n; ++i) { for (auto j=i+1; j<n; ++j) { if (v[i] > v[j]) { minLoss = min(minLoss, v[i] - v[j]); } } } cout << minLoss; } Set-based solution We need to amortize the internal loop. [Read More]

## Student Sections

The school term starts soon, and the students need to be sorted into their respective sections. There are $$n$$ students numbered $$1$$ to $$n$$, and $$m$$ sections numbered $$1$$ to $$m$$. Section needs to have exactly $$a_i$$ students. To simplify the process, the school has decided to assign students to sections in increasing student number, which means the first $$a_1$$ students will be assigned section $$1$$, the next $$a_2$$ students will be assigned section $$2$$, and so on. [Read More]

## The Crazy Broker

Solutions This problem seem complicated if you try not using additional data structures. This is a typical case where pre-processing is needed and, in particular, we use two special kind of prefix sums. To answer to the first type of queries, we need to cache the running maximums of the prices: 1 2 3 4 5 6 7 8 vector<int> prefixMax(n); auto runningMax = p.front(); prefixMax[0] = runningMax; for (auto i=1; i<n; ++i) { prefixMax[i] = max(runningMax, p[i]); runningMax = prefixMax[i]; } In C++, we have a very useful standard algorithm to perform the same routine: [Read More]

## Triple Sum

Solutions Consider this test case: Visually, for each element of B we should mark the elements in A and C that are less than or equal to that element: And, for 3: Performing the operation on the second 3 is redundant, because triplets must be unique. Along the way, we could multiply the count of A’s marked elements with the count of B’s marked elements. All the results will be then accumulated. [Read More]

## Weighted Uniform String

Solutions A C++ solution: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 string s; cin >> s; int q, i; cin >> q; vector<int> w; char last = 0; for (auto c : s) { if (c == last) w.push_back(w.back()+c-'a'+1); else w.push_back(c-'a'+1); last = c; } sort(begin(w), end(w)); while (q--) { cin >> i; cout << (binary_search(begin(w), end(w), i) ? [Read More]