Largest subvector and

HZ occasionally fooled non-computer students with some professional questions. Today, after the JOBDU test group was opened, he spoke again: in the Ancient One-dimensional pattern recognition, it is often necessary to calculate the largest sum of continuous subvectors. When all vectors are positive, the problem is well solved. However, if a vector contains a negative number, should it contain a negative number and expect the positive number next to it to compensate? For example: {6,-3,-0th,-15, 1, 3rd}, the maximum sum of continuous subvectors is 8 (starting from ). Will you be fooled by him? Input: multiple groups of data are input. Each group of test data includes two rows. The first act is an integer n (0 <= n <= 100000). When n = 0, the input ends. The next row contains n integers (we guarantee that all integers belong to [-]). Output: For each test case, three integers must be output in a single row, represents the largest continuous subvector and the subscript of the first element of the subvector and the subscript of the last element, respectively. If multiple subvectors exist, the smallest starting element is output. Sample input: 3-1-3-25-8 3 2 0 586-3-2 7-15 1 2 20 sample output: -1 0 0 10 1 4 8 0 3 idea: simple DP problem code AC: [cpp] # include <stdio. h> int main () {int data [100001]; int I, n, sum, max, start_idx, end_idx, now; while (scanf ("% d ", & n )! = EOF & n! = 0) {sum = 0; for (I = 0; I <n; I ++) {scanf ("% d", & data [I]); if (I = 0) {max = data [I]; now = 0; start_idx = 0; end_idx = 0;} if (sum <0) {sum = data [I]; now = I;} else {sum + = data [I];} if (sum> max) {max = sum; start_idx = now; end_idx = I;} else if (sum = max) {if (data [start_idx]> data [now]) {start_idx = now; end_idx = I ;}}} printf ("% d \ n", max, start_idx, end_idx);} return 0 ;}