SlicedLowLevelWCS.pixel_to_world_values() fails when object initialized with np.int64

Description

This bug came through the following code using NDCube, a sunpy affiliate project.

# Plot map of Stokes I in core
I_mean = np.mean(cube[0].data, axis=(1,2)) # mean profile over map
ix_I_min = np.argmin(I_mean) # index of minimum of mean profile over map
level1_cube[0,ix_I_min, :, :].plot()

Expected behavior

I expected to see a plot.

Actual behavior

The result was a long stack trace ending with:

...
/opt/local/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/astropy/wcs/wcsapi/wrappers/sliced_wcs.py in _pixel_to_world_values_all(self, *pixel_arrays)
    219             else:
    220                 ipix_curr += 1
--> 221                 if self._slices_pixel[ipix].start is not None:
    222                     pixel_arrays_new.append(pixel_arrays[ipix_curr] + self._slices_pixel[ipix].start)
    223                 else:

AttributeError: 'numpy.int64' object has no attribute 'start'

If I cast ix_I_min to int, there is no problem.

Steps to Reproduce

This code is a minimal example to reproduce what I think is the same bug.

import numpy as np
import astropy.wcs
from astropy.wcs.wcsapi import SlicedLowLevelWCS

wcs_dict = {
'CTYPE1': 'WAVE    ', 'CUNIT1': 'Angstrom', 'CDELT1': 0.2, 'CRPIX1': 0, 'CRVAL1': 10, 'NAXIS1': 5,
'CTYPE2': 'HPLT-TAN', 'CUNIT2': 'deg', 'CDELT2': 0.5, 'CRPIX2': 2, 'CRVAL2': 0.5, 'NAXIS2': 4,
'CTYPE3': 'HPLN-TAN', 'CUNIT3': 'deg', 'CDELT3': 0.4, 'CRPIX3': 2, 'CRVAL3': 1, 'NAXIS3': 3}
input_wcs = astropy.wcs.WCS(wcs_dict)
low_level = SlicedLowLevelWCS(input_wcs, [np.int64(0), 0])
low_level.pixel_to_world_values(0)

this results in:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-415-f0f7b290cda9> in <module>
      9 input_wcs = astropy.wcs.WCS(wcs_dict)
     10 low_level = SlicedLowLevelWCS(input_wcs, [np.int64(0), 0])
---> 11 low_level.pixel_to_world_values(0)

/opt/local/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/astropy/wcs/wcsapi/wrappers/sliced_wcs.py in pixel_to_world_values(self, *pixel_arrays)
    228 
    229     def pixel_to_world_values(self, *pixel_arrays):
--> 230         world_arrays = self._pixel_to_world_values_all(*pixel_arrays)
    231 
    232         # Detect the case of a length 0 array

/opt/local/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/astropy/wcs/wcsapi/wrappers/sliced_wcs.py in _pixel_to_world_values_all(self, *pixel_arrays)
    219             else:
    220                 ipix_curr += 1
--> 221                 if self._slices_pixel[ipix].start is not None:
    222                     pixel_arrays_new.append(pixel_arrays[ipix_curr] + self._slices_pixel[ipix].start)
    223                 else:

AttributeError: 'numpy.int64' object has no attribute 'start'

System Details

macOS-10.15.7-x86_64-i386-64bit Python 3.8.8 (default, Feb 25 2021, 17:03:48) [Clang 12.0.0 (clang-1200.0.26.2)] Numpy 1.20.1 astropy 4.2 Scipy 1.6.0 Matplotlib 3.3.4

[Cadair]

I didn’t try.

[pllim]

@Cadair , were you able to track down which PR actually fixed it? Thanks!