LeetCode 39 && 40 Combination Sum I && II 关键在于剪枝，剪枝讨论在末尾

Given a set of candidate numbers (C) and a target number (T), find all unique combinations in C where the candidate numbers sums to T.

The same repeated number may be chosen from C unlimited number of times.

Note:

• All numbers (including target) will be positive integers.
• Elements in a combination (a1, a2, … , ak) must be in non-descending order. (ie, a1 ≤ a2 ≤ … ≤ ak).
• The solution set must not contain duplicate combinations.

For example, given candidate set 2,3,6,7 and target 7, 
A solution set is: 
[7] 

[2, 2, 3] 

Combination Sum II

Given a collection of candidate numbers (C) and a target number (T), find all unique combinations in C where the candidate numbers sums to T.

Each number in C may only be used once in the combination.

Note:

• All numbers (including target) will be positive integers.
• Elements in a combination (a1, a2, … , ak) must be in non-descending order. (ie, a1 ≤ a2 ≤ … ≤ ak).
• The solution set must not contain duplicate combinations.

For example, given candidate set 10,1,2,7,6,1,5 and target 8, 
A solution set is: 
[1, 7] 
[1, 2, 5] 
[2, 6] 
[1, 1, 6] 

解题思路：

这道题是一道典型的搜索算法计算题，其中LeetCode 39为支持重复的求和计算，而LeetCode 40为不支持重复求和的计算题。这两道题的关键都在于深度检索的时候的边界的设定，递归边界为如果结果大于target则相关路径不用继续遍历。LeetCode 40与LeetCode 39的不同在于不支持重复元素，因此，递归函数对应参数level为+1.另外，对于单变量的情况要单独考虑。

代码如下：

Combination Sum：

public class Solution {
     List<List<Integer>> result;
		List<Integer> solu;
		public List<List<Integer>> combinationSum(int[] candidates, int target) {
	        result = new ArrayList<>();
	        solu = new ArrayList<>();
	        Arrays.sort(candidates);
	        getCombination(candidates, target, 0, 0);
	        return result;
	    }
		public void getCombination(int[] candidates, int target, int sum, int level){
			if(sum>target) return;
			if(sum==target){
				result.add(new ArrayList<>(solu));
				return;
			}
			for(int i=level;i<candidates.length;i++){
				sum+=candidates[i];
				solu.add(candidates[i]);
				getCombination(candidates, target, sum, i);
				solu.remove(solu.size()-1);
				sum-=candidates[i];
			}
		}
}

Combination Sum II：

public class Solution {
    	    List<List<Integer>> result;
		List<Integer> solu;
    public List<List<Integer>> combinationSum2(int[] candidates, int target) {
         result = new ArrayList<>();
	        solu = new ArrayList<>();
	        Arrays.sort(candidates);
	        getCombination(candidates, target, 0, 0);
	        return result;
    }
    
    public void getCombination(int[] candidates, int target, int sum, int level){
	if(sum>target) return;
			if(sum==target){
				if(!result.contains(new ArrayList<>(solu)))
				result.add(new ArrayList<>(solu));
				return;
			}
			if(candidates.length==1){//添加的条件
				if(target==candidates[0]){
				solu.add(candidates[0]);
				result.add(solu);
                }
				return ;
			}
			for(int i=level;i<candidates.length;i++){
				sum+=candidates[i];// 
				solu.add(candidates[i]);
				getCombination(candidates, target, sum, i+1);//新的递归参数
				solu.remove(solu.size()-1);
				sum-=candidates[i];
			}
		}
}

实际上，该算法可以优化处理，对剪枝进行优化，我们在判断出当前节点符合或者大于target的时候，对同一个层次之后的节点便不再需要遍历，修改之后的代码如下：

public int getCombination(int[] candidates, int target, int sum, int level){
			if(sum>target) return 1;//剪枝
			if(sum==target){
				if(!result.contains(new ArrayList<>(solu)))
				result.add(new ArrayList<>(solu));
				return 1;//剪枝
			}
			if(candidates.length==1){
				if(target==candidates[0]){
				solu.add(candidates[0]);
				result.add(solu);
                }
				return 0;
			}
			for(int i=level;i<candidates.length;i++){
				sum+=candidates[i];// 
				solu.add(candidates[i]);
				int j=getCombination(candidates, target, sum, i+1);
				solu.remove(solu.size()-1);
				sum-=candidates[i];
				if(j==1){//剪枝
					i=candidates.length;
				}
			}
			return 0;
		}