Finding number of subarrays whose sum equals `k`

We would be given an array of integers and a value k. We need to find the total number of sub-arrays whose sum equals k.

I found some interesting code online (on Leetcode) which is as follows:

public class Solution {
    public int subarraySum(int[] nums, int k) {
        int sum = 0, result = 0;
        Map<Integer, Integer> preSum = new HashMap<>();
        preSum.put(0, 1);

        for (int i = 0; i < nums.length; i++) {
            sum += nums[i];
            if (preSum.containsKey(sum - k)) {
                result += preSum.get(sum - k);
            }
            preSum.put(sum, preSum.getOrDefault(sum, 0) + 1);
        }

        return result;
    }
}

To understand it, I walked through some specific examples like [1,1,1,1,1] with k=3 and [1,2,3,0,3,2,6] with k=6. While the code works perfectly in both the cases, I fail to follow how it actually computes the output.

I have two specific points of confusion:

1) Why does the code continuously add the values in the array, without ever zeroing it out? For example, in case of [1,1,1,1,1] with k=3, once sum=3, don't we need to reset sum to zero? Doesn't not resetting sum interfere with finding later subarrays?

2) Shouldn't we simply do result++ when we find a subarray of sum k? Why do we add preSum.get(sum-k) instead?