可迭代解包评估顺序 [英] Iterable Unpacking Evaluation Order
问题描述
我最近回答了一个问题 用户遇到麻烦,因为他们将一个多维数组附加到另一个数组,并且在我的回答中引起了我的注意,可以使用可迭代解包来填充 x
和 y
值并分配给同一行的 board[x][y]
.
我原以为这会引发错误,因为 x
和 y
当时尚未定义,即使在 iterable-unpacking代码> 标记为:
迭代的元素同时分配给多个值
这可以在以下示例中看到:
<预><代码>>>>板 = [[0, 0], [0, 0]]>>>移动 = [0, 1, 2]>>>x, y, board[x][y] = 移动>>>木板[[0, 2], [0, 0]]与以下相同:
<预><代码>>>>板 = [[0, 0], [0, 0]]>>>移动 = [0, 1, 2]>>>x = 移动[0]>>>y = 移动 [1]>>>板[x][y] = 移动[2]>>>木板[[0, 2], [0, 0]]然而在使用以下方法计算斐波那契数列时:
a, b = b, a + b
它不会评估为:
a = bb = a + b
当交换值时:
a, b = b, a
它不会评估为:
a = bb = a
那么为什么这在第一个示例中有效?
=
的右侧总是首先被评估,在这种情况下它正在打包一个元组.然后在解释左侧时解压缩该元组.左侧和右侧不共享变量知识.RHS 成为一个值,然后 LHS 使用该值分配给变量(标签).
在您的示例中,x
和 y
的值是在评估 RHS 之后确定的.然后从左到右解包,以便 board[x][y]
具有有效索引.
切换顺序演示解包顺序:
<预><代码>>>>板[x][y],x,y = 移动[2],移动[0],移动[1]---------------------------------------------------------------------------NameError Traceback(最近一次调用最后一次)<ipython-input-7-a984ef3168f8>在 <module>()---->1 板[x][y], x, y = 移动[2], 移动[0], 移动[1]NameError: 名称 'x' 未定义I recently answered a question where a user was having trouble because they were appending a multi-dimensional array to another array, and it was brought to my attention in my answer that it is possible to use iterable unpacking to populate an x
and y
value and assign to board[x][y]
on the same line.
I had expected this to throw an error as x
and y
had at the time not been defined, as, even in the iterable-unpacking
tag it reads:
elements of an iterable are simultaneously assigned to multiple values
This can be seen as working in the following example:
>>> board = [[0, 0], [0, 0]]
>>> move = [0, 1, 2]
>>> x, y, board[x][y] = move
>>> board
[[0, 2], [0, 0]]
Which is the same as:
>>> board = [[0, 0], [0, 0]]
>>> move = [0, 1, 2]
>>> x = move[0]
>>> y = move[1]
>>> board[x][y] = move[2]
>>> board
[[0, 2], [0, 0]]
And yet when calculating the Fibonacci sequence using:
a, b = b, a + b
It doesn't evaluate as:
a = b
b = a + b
And when swapping values with:
a, b = b, a
It doesn't evaluate as:
a = b
b = a
So why does this work in the first example?
The right side of the =
is always evaluated first, in this case it is packing a tuple. That tuple is then unpacked when interpreting the left hand side. The left and right sides do not share knowledge of variables. The RHS becomes a value and then the LHS uses that value to assign to the variables (labels).
In your example the values of x
and y
are determined after the RHS is evaluated. The unpacking then occurs left to right, so that board[x][y]
has valid indices.
Switching the order demonstrates the unpacking sequence:
>>> board[x][y], x, y = move[2], move[0], move[1]
---------------------------------------------------------------------------
NameError Traceback (most recent call last)
<ipython-input-7-a984ef3168f8> in <module>()
----> 1 board[x][y], x, y = move[2], move[0], move[1]
NameError: name 'x' is not defined
这篇关于可迭代解包评估顺序的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!