Question:

Methods: The violent solution can go through three cycles, but the complexity will reach three; You can reduce the number of cycles by generating a map, and then loop through two layers to get the final result.

package com.eric.leetcode class ThreeSum { fun threeSum(nums: IntArray): List<List<Int>> { val xSet = nums.mapIndexed { index, it -> it to index }.toMap().toMutableMap() val result = mutableSetOf<List<Int>>() for (i in 0.. nums.lastIndex) { for (j in i.. nums.lastIndex) { val x = nums[i] val y = nums[j] if (xSet.containsKey(-x - y)) { val k = xSet[-x - y] if (i ! = j && i ! = k && k ! = j) { result.add(listOf(x, y, -x - y).sorted()) } } } } return result.toList() } } fun main() { val input = intArrayOf(-1, 0, 1, 2, -1, -4) val threeSum = ThreeSum() threeSum.threeSum(input) }Copy the code

Feel free to communicate if you have any questions

Specific code implementation can refer to Github