LeetCode: Gas Station
There are N gas stations along a circular route, where the amount of gas at station i is gas[i]
.
You have a car with an unlimited gas tank and it costs cost[i]
of gas to travel from station i to its next station (i+1). You begin the journey with an empty tank at one of the gas stations.
Return the starting gas station's index if you can travel around the circuit once, otherwise return -1.
Note:
The solution is guaranteed to be unique.
地址:https://oj.leetcode.com/problems/gas-station/
算法:一个简单粗暴的方法是,从第0个station开始试,看其是否可以绕一圈回到起点,这样的时间复杂度是$O(n^2)$,如果用这种方法那么肯定是会超时的(当时我也是用这个方法,下面这个巧妙的方法也是参考了网上的答案)。那么,下面我们介绍一种线性时间复杂度的算法,这个算法是对上面一种方法的改进。这个方法最核心的思想是基于如下的事实:假如从第0个station可以到达第$j=1,2,cdots,i-1$个station,但无法到达第i个station的话,那么从第$j=1,2,cdots,i-1$个station出发也肯定无法到达第i个station。原因很简单,假如可以从第j个station到达第i个station,那么我们可以先从第0个到达第j个,然后在从第j个到达第i个,也就是说我们可以从第0个到达第i个,故这个假设是不成立的。这样的话,当我们判断从第0个station无法到达第i个时,可以不用回过头从第1个station判断,我们可以直接继续从第i+1个station开始判断。由于题目说至少会存在一个解,也就是我们不需要去判断不存在解的情况,所以只要循环一遍就足够了。代码:
1 class Solution {
2 public:
3 int canCompleteCircuit(vector<int> &gas, vector<int> &cost) {
4 int n = gas.size();
5 int sum_gas = 0, sum_cost = 0;
6 for(int i = 0; i != n; ++i){
7 sum_gas += gas[i];
8 sum_cost += cost[i];
9 }
10 if (sum_cost > sum_gas){
11 return -1;
12 }
13 sum_gas = 0;
14 sum_cost = 0;
15 int start = 0;
16 for(int i = 0; i != n; ++i){
17 sum_gas += gas[i];
18 sum_cost += cost[i];
19 if(sum_gas >= sum_cost){
20 continue;
21 }else{
22 start = (i + 1) % n;
23 sum_gas = 0;
24 sum_cost = 0;
25 }
26 }
27 return start;
28 }
29 };