当用户点击X时，我想检查页面中是否有任何未保存的更改。我有isSaved属性告诉我。
现在，如果有任何未保存的更改，我想显示警告窗口（已就绪），以便用户确认是否关闭窗格。
如果用户说是，我想关闭。如果不是我不想关闭。
如何在MVVM中实现这一点？我首先了解了触发的PreviewClose。如果有未保存的更改，我该如何停止关闭窗格？
Here are my interaction trigger.
<i:Interaction.Triggers>
<i:EventTrigger EventName="PreviewClose">
<cmd:EventToCommand Command="{Binding PreviewCloseCommand}" CommandParameter="{Binding ElementName=ContentContainer}" />
</i:EventTrigger>
<i:EventTrigger EventName="Close">
<cmd:EventToCommand Command="{Binding CloseCommand}" CommandParameter="{Binding ElementName=ContentContainer}" />
</i:EventTrigger>
</i:Interaction.Triggers>
答案 0 :(得分：0)
有关几个解决方案，请参阅numpy.roll。
对于取消关闭窗口，您可以更新from itertools import product
import numpy as np
def multiroll(x, shift, axis=None):
"""Roll an array along each axis.
Parameters
----------
x : array_like
Array to be rolled.
shift : sequence of int
Number of indices by which to shift each axis.
axis : sequence of int, optional
The axes to be rolled. If not given, all axes is assumed, and
len(shift) must equal the number of dimensions of x.
Returns
-------
y : numpy array, with the same type and size as x
The rolled array.
Notes
-----
The length of x along each axis must be positive. The function
does not handle arrays that have axes with length 0.
See Also
--------
numpy.roll
Example
-------
Here's a two-dimensional array:
>>> x = np.arange(20).reshape(4,5)
>>> x
array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])
Roll the first axis one step and the second axis three steps:
>>> multiroll(x, [1, 3])
array([[17, 18, 19, 15, 16],
[ 2, 3, 4, 0, 1],
[ 7, 8, 9, 5, 6],
[12, 13, 14, 10, 11]])
That's equivalent to:
>>> np.roll(np.roll(x, 1, axis=0), 3, axis=1)
array([[17, 18, 19, 15, 16],
[ 2, 3, 4, 0, 1],
[ 7, 8, 9, 5, 6],
[12, 13, 14, 10, 11]])
Not all the axes must be rolled. The following uses
the `axis` argument to roll just the second axis:
>>> multiroll(x, [2], axis=[1])
array([[ 3, 4, 0, 1, 2],
[ 8, 9, 5, 6, 7],
[13, 14, 10, 11, 12],
[18, 19, 15, 16, 17]])
which is equivalent to:
>>> np.roll(x, 2, axis=1)
array([[ 3, 4, 0, 1, 2],
[ 8, 9, 5, 6, 7],
[13, 14, 10, 11, 12],
[18, 19, 15, 16, 17]])
"""
x = np.asarray(x)
if axis is None:
if len(shift) != x.ndim:
raise ValueError("The array has %d axes, but len(shift) is only "
"%d. When 'axis' is not given, a shift must be "
"provided for all axes." % (x.ndim, len(shift)))
axis = range(x.ndim)
else:
# axis does not have to contain all the axes. Here we append the
# missing axes to axis, and for each missing axis, append 0 to shift.
missing_axes = set(range(x.ndim)) - set(axis)
num_missing = len(missing_axes)
axis = tuple(axis) + tuple(missing_axes)
shift = tuple(shift) + (0,)*num_missing
# Use mod to convert all shifts to be values between 0 and the length
# of the corresponding axis.
shift = [s % x.shape[ax] for s, ax in zip(shift, axis)]
# Reorder the values in shift to correspond to axes 0, 1, ..., x.ndim-1.
shift = np.take(shift, np.argsort(axis))
# Create the output array, and copy the shifted blocks from x to y.
y = np.empty_like(x)
src_slices = [(slice(n-shft, n), slice(0, n-shft))
for shft, n in zip(shift, x.shape)]
dst_slices = [(slice(0, shft), slice(shft, n))
for shft, n in zip(shift, x.shape)]
src_blks = product(*src_slices)
dst_blks = product(*dst_slices)
for src_blk, dst_blk in zip(src_blks, dst_blks):
y[dst_blk] = x[src_blk]
return y
以取消活动。