Get rid of np.int

[Didou09]

Main changes:

New strategy for int/ float type checks

# Useful scalar types
_NINT = (np.int32, np.int64)
_INT = (int,) + _NINT
_NFLOAT = (np.float32, np.float64)
_FLOAT = (float,) + _NFLOAT
_NUMB = _INT + _FLOAT

And always run isinstance()

Because the behaviour of np.int, np.int_ or np.integer is too dependent on the version of numpy and on the plateform
Also shows different results depending is using type() or dtype or isinstance()

• isintance() allows multiple types and subclasses
• Better to use tuple of all explicit numpy types that resort to generic types

Issues:

Fixes, in devel, issue #983

[pep8speaks]

Hello @Didou09! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

• In the file tofu/imas2tofu/_core.py:

Line 376:80: E501 line too long (83 > 79 characters)
Line 1500:71: E231 missing whitespace after ','
Line 1500:80: E501 line too long (83 > 79 characters)

• In the file tofu/imas2tofu/_mat2ids2calc.py:

Line 208:16: W292 no newline at end of file

Comment last updated at 2024-11-13 18:37:09 UTC