From 31ddd23e1d91cafa1b57f2ac22795d2185027ead Mon Sep 17 00:00:00 2001
From: hitorig <86967184+hitorig@users.noreply.github.com>
Date: Fri, 6 Aug 2021 18:03:29 +0800
Subject: [PATCH] =?UTF-8?q?Update=200047.=E5=85=A8=E6=8E=92=E5=88=97II.md?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 problems/0047.全排列II.md | 48 +++++++++++++++++-------------------
 1 file changed, 22 insertions(+), 26 deletions(-)

diff --git a/problems/0047.全排列II.md b/problems/0047.全排列II.md
index 66107c95..c8764241 100644
--- a/problems/0047.全排列II.md
+++ b/problems/0047.全排列II.md
@@ -228,35 +228,31 @@ Go：
 ```go
 var res [][]int
 func permute(nums []int) [][]int {
-    res = [][]int{}
-    sort.Ints(nums)
-    dfs(nums, make([]int, 0), make([]bool, len(nums)))
-    return res
+	res = [][]int{}
+	backTrack(nums,len(nums),[]int{})
+	return res
 }
+func backTrack(nums []int,numsLen int,path []int)  {
+	if len(nums)==0{
+		p:=make([]int,len(path))
+		copy(p,path)
+		res = append(res,p)
+	}
+	used := [21]int{}//跟前一题唯一的区别，同一层不使用重复的数。关于used的思想carl在递增子序列那一题中提到过
+	for i:=0;i<numsLen;i++{
+		if used[nums[i]+10]==1{
+			continue
+		}
+		cur:=nums[i]
+		path = append(path,cur)
+		used[nums[i]+10]=1
+		nums = append(nums[:i],nums[i+1:]...)
+		backTrack(nums,len(nums),path)
+		nums = append(nums[:i],append([]int{cur},nums[i:]...)...)
+		path = path[:len(path)-1]
 
-func dfs(nums, path []int, used []bool) {
-    if len(path) == len(nums) {
-        res = append(res, append([]int{}, path...))
-        return
-    }
+	}
 
-    m := make(map[int]bool)
-    for i := 0; i < len(nums); i++ {
-        // used 从剩余 nums 中选
-        if used[i] {
-            continue
-        }
-        // m 集合间去重
-        if _, ok := m[nums[i]]; ok {
-            continue
-        }
-        m[nums[i]] = true
-        path = append(path, nums[i])
-        used[i] = true
-        dfs(nums, path, used)
-        used[i] = false
-        path = path[:len(path)-1]
-    }
 }
 ```