“[dict（）] * n”和“for循环与.append（dict（））”之间有什么区别？

Question

我正在实现DFS算法，我想创建一个字典列表。但我发现，如果我使用[dict()] * n，我会得到意想不到的结果。

我不确定“[dict()] * n”和“for循环与.append(dict())”之间有什么区别。

任何提示都表示赞赏。

def DFS1(nums, level, target, dp):
    if level == len(nums):
        if target == 0:
            return 1
        else:
            return 0

    if target in dp[level]:
        return dp[level][target]

    cnt1 = DFS1(nums, level + 1, target + nums[level], dp)
    cnt2 = DFS1(nums, level + 1, target - nums[level], dp)
    dp[level][target] = cnt1 + cnt2
    return cnt1 + cnt2

测试代码是：

nums = [1,1,1,1,1]
n = len(nums)
target = -3
#dp = [dict()] * n   #<== This does not work as expected
dp = []
for i in range(0, n):
    dp.append(dict())

resCnt = DFS1(nums, 0, target, dp)
print(resCnt)

Answer 1

基本上发生的是你正在创建引用：

x = [dict()] * 10
print x
>[{}, {}, {}, {}, {}, {}, {}, {}, {}, {}]  #looks fine

x[0]['x'] = 3                              #all references share same memory
print x
[{'x': 3}, {'x': 3}, {'x': 3}, {'x': 3}, {'x': 3}, {'x': 3}, {'x': 3}, {'x': 3}, {'x': 3}, {'x': 3}]

视觉跟随发生：

为了避免referencing，你必须明确告诉python每次都创建新的词典：

x = [{} for x in range(10)]
print x
>[{}, {}, {}, {}, {}, {}, {}, {}, {}, {}]
x[0]['x'] = 3
print x
>[{'x': 3}, {}, {}, {}, {}, {}, {}, {}, {}, {}]

视觉表示：每个存储桶都有dict的自己的副本：