[BUG] Reshaping a slice creates an empty array

[furbrain]

Describe the bug Calling reshape on a slice of an array can lead to an empty array. ulab version 6.4.1-2D

To Reproduce

>>> a = np.array(range(36)).reshape((6,6))
>>> a
array([[0.0, 1.0, 2.0, 3.0, 4.0, 5.0],
       [6.0, 7.0, 8.0, 9.0, 10.0, 11.0],
       [12.0, 13.0, 14.0, 15.0, 16.0, 17.0],
       [18.0, 19.0, 20.0, 21.0, 22.0, 23.0],
       [24.0, 25.0, 26.0, 27.0, 28.0, 29.0],
       [30.0, 31.0, 32.0, 33.0, 34.0, 35.0]], dtype=float32)
>>> b = a[0,:]
>>> b.reshape((1,6))
array([[0.0, 1.0, 2.0, 3.0, 4.0, 5.0]], dtype=float32)
>>> c = a[:,0]
>>> c.reshape((1,6))
array([], shape=(0,6), dtype=float32)

Expected behavior

c should be [[0.0, 6.0, 12.0, 18.0, 24.0, 30.0]], not []

Additional context

This seems to be a result of recent changes. Ulab (6.0.12) that came with circuitpython 8.1 did not have this issue

[v923z]

I get

c.reshape((1,6))
array([[0.0],
       [6.0],
       [12.0],
       [18.0],
       [24.0],
       [30.0]], dtype=float64)

which is also wrong.

[furbrain]

I think that is correct: c is looking at the first column, b is looking at the first row. My demo code was running on a nrf52840 in circuitpython

Phil

[furbrain]

Sorry, didn't look at your result properly. I see the issue...

[furbrain]

Just had a look at numpy docs for reshape: "it is not always possible to change the shape of an array without copying the data"

[v923z]

Yes, that's correct, but your particular example should still be doable without copying. I'll try to sort out the issue.

[v923z]

Fixed in https://github.com/v923z/micropython-ulab/pull/660