python实现算法导论伪代码——求最大连续子数组

给定一个数组，求出和最大的连续子数组。返回子数组的开始位，结束位，和。

思路：分治策略，将数组一分为二，最大子数组可能在左半，或右半，或横跨中间。不断分割，直到子问题可解，求出子数组，再逐步合并。

def Max_subarray(a,low,high): #初始输入的low和high应为0，len(a)-1
    if high == low:
        return [low, high, a[low]]
    else:
        mid = int((high+low)/2)
        L = Max_subarray(a,low,mid)
        R = Max_subarray(a,mid+1,high)
        M = Max_cross_subarray(a,low,mid,high)
        if L[2]>=R[2] and L[2]>=M[2]:
            return L
        if R[2]>=L[2] and R[2]>=M[2]:
            return R
        else:
            return M

def Max_cross_subarray(a,low,mid,high):
    max_left = mid
    max_right = mid
    left_sum = -float("inf")
    right_sum = -float("inf")
    sum = 0
    i = mid
    while i>= low:
        sum = sum + a[i]
        if sum >= left_sum:
            left_sum = sum
            max_left = i
        i=i-1
    sum = 0
    j = mid+1
    while j<= high:
        sum = sum + a[j]
        if sum >= right_sum:
            right_sum = sum
            max_right = j
        j=j+1
    return [max_left, max_right, left_sum+right_sum]