Golang solution using priority queue

  • 1

    First, fetch all 0 th elements of each array in matrix and initialize priority queue, whose priority is a value in matrix.
    Then we can run heap.Pop one by one. And after every call of Pop, if popped element is matrix[i][j] add matrix[i][j+1] to the priority queue. By doing so, we can ensure always [elements in priority queue] < [elements remaining in matrix],
    therefore k th popped element is what we want.

    func kthSmallest(matrix [][]int, k int) int {
    	mlen, nlen := len(matrix), len(matrix[0])
    	var pq Elems
    	for i := 0; i < mlen; i++ {
    		pq = append(pq, Elem{m: i, n: 0, val: matrix[i][0]})
    	var res int
    	for i := 0; i < k; i++ {
    		e, _ := heap.Pop(&pq).(Elem)
    		if e.n < nlen-1 {
    			heap.Push(&pq, Elem{m: e.m, n: e.n + 1, val: matrix[e.m][e.n+1]})
    		res = e.val
    	return res
    // Below code is Golang boiler plate to suffice priority queue interface.
    // Refer https://golang.org/pkg/container/heap/
    type Elem struct {
    	m   int
    	n   int
    	val int
    type Elems []Elem
    func (pq Elems) Swap(i, j int) {
    	pq[i], pq[j] = pq[j], pq[i]
    func (pq Elems) Len() int {
    	return len(pq)
    func (pq Elems) Less(i, j int) bool {
    	return pq[i].val < pq[j].val
    func (pq *Elems) Pop() interface{} {
    	vpq := *pq
    	x := vpq[len(vpq)-1]
    	*pq = vpq[:len(vpq)-1]
    	return x
    func (pq *Elems) Push(num interface{}) {
    	numElm, _ := num.(Elem)
    	*pq = append(*pq, numElm)

Log in to reply

Looks like your connection to LeetCode Discuss was lost, please wait while we try to reconnect.