区间和的个数（multiset/线段树/归并）

原题
给定一个整数数组 nums，返回区间和在 [lower, upper] 之间的个数，包含 lower 和 upper。
区间和 S(i, j) 表示在 nums 中，位置从 i 到 j 的元素之和，包含 i 和 j (i ≤ j)。
参考

multiset

class Solution {
public:
    int countRangeSum(vector<int>& nums, int lower, int upper) {
        int n = nums.size();
        multiset<long long> st;
        st.insert(0);
        long long sum = 0;
        int ans = 0,l,r;
        for(auto val:nums) {
            sum += val;
            ans += distance(st.lower_bound(sum-upper),st.upper_bound(sum-lower));
            st.insert(sum);
        }
        return ans;
    }
};

线段树+离散化

class Solution {
public:
    int countRangeSum(vector<int>& nums, int lower, int upper) {
        int n = nums.size(),pos,a,b;
        int ans = 0;
        vector<long long> h(n+1);
        h[0] = 0;
        for(int i = 1;i <= n;++i) {
            h[i] = h[i-1]+nums[i-1];
        }
        sort(h.begin(),h.end());
        h.erase(unique(h.begin(),h.end()),h.end());
        n = h.size();
        long long cur = 0;
        sum.resize(4*(n+1),0);
        pos = lower_bound(h.begin(),h.end(),cur) - h.begin() + 1;
        update(1,1,n,pos,1);
        for(auto val : nums) {
            cur += val;
            a = lower_bound(h.begin(),h.end(),cur-upper) - h.begin() + 1;
            b = upper_bound(h.begin(),h.end(),cur-lower) - h.begin();
            if(a <= b)
                ans += query(1,1,n,a,b);
            pos = lower_bound(h.begin(),h.end(),cur) - h.begin() + 1;
            update(1,1,n,pos,1);
        }
        return ans;
    }
    vector<int> sum;
    void pushup(int rt) {
        sum[rt] = sum[rt<<1] + sum[rt<<1|1];
    }
    void update(int rt,int l,int r,int pos,int val) {
        if(l == r) {
            sum[rt] += val;
            return;
        }
        int m = (l+r)>>1;
        if(pos <= m)
            update(rt<<1,l,m,pos,val);
        else
            update(rt<<1|1,m+1,r,pos,val);
        pushup(rt);
    }
    int query(int rt,int l,int r,int a,int b) {
        if(l >= a && r <= b)
            return sum[rt];
        int m = (l+r)>>1;
        int ans = 0;
        if(a <= m)
            ans += query(rt<<1,l,m,a,b);
        if(m < b)
            ans += query(rt<<1|1,m+1,r,a,b);
        return ans;
    }
    
};

归并排序

常数最小的解法

class Solution {
public:
    int ans;
    int countRangeSum(vector<int>& nums, int lower, int upper) {
        int n = nums.size();
        ans = 0;
        vector<long long> sum(n+1);
        vector<long long> tmp(n+1);
        sum[0] = 0;
        for(int i = 1;i <= n;++i)
            sum[i] = sum[i-1] + nums[i-1];
        mergeSort(sum,tmp,0,n+1,lower,upper);
        return ans;
    }
    void mergeSort(vector<long long> &sum,vector<long long> &tmp,int l,int r,int &lower,int &upper) {
        if(l+1 >= r) return;
        int m = (l+r)>>1;
        mergeSort(sum,tmp,l,m,lower,upper);
        mergeSort(sum,tmp,m,r,lower,upper);
        merge(sum,tmp,l,m,r,lower,upper);
    }
    void merge(vector<long long> &sum,vector<long long> &tmp,int &l,int &m,int &r,int &lower,int &upper) {
        int pos = l;
        int pl = m,pr = m;
        while(pos < m) {
            while(pl < r && sum[pl] - sum[pos] < lower) ++pl;
            while(pr < r && sum[pr] - sum[pos] <= upper) ++pr;
            ans += pr - pl;
            ++pos;
        }
        pos = l;
        pl = l;pr = m;
        while(pl < m && pr < r) {
            if(sum[pl] < sum[pr]) tmp[pos++] = sum[pl++];
            else tmp[pos++] = sum[pr++];
        }
        while(pl < m) tmp[pos++] = sum[pl++];
        while(pr < r) tmp[pos++] = sum[pr++];
        for(int i = l;i < r;++i) sum[i] = tmp[i];
    }
    
};