BUG: dtype datetime64[ns, UTC] not interpreted by issubdtype #28946
Description
Describe the issue:
This is more or less the same as reported in #19187 and allegedly fixed. But it does still occur in the version combinations I tested with (see below).
Reproduce the code example:
import pandas as pd
import numpy as np
dtype = pd.date_range("2016-01-01", periods=2, tz="UTC").dtype
np.issubdtype(dtype, np.number)Error message:
Traceback (most recent call last):
File "/tmp/datetimetest.py", line 6, in <module>
np.issubdtype(dtype, np.number)
File "/home/****/.cache/uv/archive-v0/pa2nJmRNwJrYZLooIE9pq/lib/python3.12/site-packages/numpy/_core/numerictypes.py", line 530, in issubdtype
arg1 = dtype(arg1).type
^^^^^^^^^^^
TypeError: Cannot interpret 'datetime64[ns, UTC]' as a data typePython and NumPy Versions:
Tested with the following combinations:
1.21.0
3.9.21 (main, Mar 17 2025, 21:01:54)
1.21.5
3.9.21 (main, Mar 17 2025, 21:01:54)
pandas 1.2.3, 1.2.4. 1.2.5, (current when #19187 was reported) and 2.2.3
1.22.0
3.10.12 (main, Feb 4 2025, 14:57:36) [GCC 11.4.0]
1.26.4
3.9.21 (main, Mar 17 2025, 21:01:54)
2.0.2
3.12.8 (main, Jan 14 2025, 22:49:14) [Clang 19.1.6 ]
2.2.5
3.12.8 (main, Jan 14 2025, 22:49:14) [Clang 19.1.6 ]
2.2.5
3.13.2 (main, Mar 17 2025, 21:02:54) [Clang 20.1.0 ]
Runtime Environment:
[{'numpy_version': '2.2.5',
'python': '3.13.2 (main, Mar 17 2025, 21:02:54) [Clang 20.1.0 ]',
'uname': uname_result(system='Linux', node='RNG-C-003M3', release='6.8.0-59-generic', version='#61~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 15 17:03:15 UTC 2', machine='x86_64')},
{'simd_extensions': {'baseline': ['SSE', 'SSE2', 'SSE3'],
'found': ['SSSE3',
'SSE41',
'POPCNT',
'SSE42',
'AVX',
'F16C',
'FMA3',
'AVX2'],
'not_found': ['AVX512F',
'AVX512CD',
'AVX512_KNL',
'AVX512_KNM',
'AVX512_SKX',
'AVX512_CLX',
'AVX512_CNL',
'AVX512_ICL']}},
{'architecture': 'Haswell',
'filepath': '/home/****/.cache/uv/archive-v0/7PdzZC8-2kfVkjeY8DQGF/lib/python3.13/site-packages/numpy.libs/libscipy_openblas64_-6bb31eeb.so',
'internal_api': 'openblas',
'num_threads': 20,
'prefix': 'libscipy_openblas',
'threading_layer': 'pthreads',
'user_api': 'blas',
'version': '0.3.28'}]
Context for the issue:
I am converting data and depending on if it's a numeric type or not (hence the use of np.issubdtype(..., np.number)) I need to treat the conversion differently.
I can of course work around it by checking beforehand if it's a datetime, but that would be kind of ugly and against any clean code principles.