Support str-like values for str_to_int

[jongbinjung]

Since numpy>=2 (?), str values are not native str but numpy.str_.

Without the explicit conversion to str, the test fails with:

================================================== FAILURES ===================================================
_______________________________________________ test_numpy_str ________________________________________________

    def test_numpy_str():
        expected = [
            617700169957507071,
            617700169957769215,
            617700169958031359,
            617700169958293503,
            617700169961177087,
            617700169964847103,
            617700169965109247,
        ]
        cells = np.array([h3.int_to_str(h) for h in expected])
>       got = [h3.str_to_int(c) for c in cells]

tests/test_apis/test_numpy_int.py:50:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
tests/test_apis/test_numpy_int.py:50: in <listcomp>
    got = [h3.str_to_int(c) for c in cells]
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

h = np.str_('89283082803ffff')

    def str_to_int(h):
        """
        Converts a hexadecimal string to an H3 64-bit integer index.

        Parameters
        ----------
        h : str
            Hexadecimal string like ``'89754e64993ffff'``

        Returns
        -------
        int
            Unsigned 64-bit integer
        """
        # return _cy.str_to_int(str(h))
>       return _cy.str_to_int(h)
E       TypeError: Argument 'h' has incorrect type (expected str, got numpy.str_)

env/lib/python3.11/site-packages/h3/api/numpy_int/__init__.py:57: TypeError

• Test at failing state

[CLAassistant]

All committers have signed the CLA.

[ajfriend]

Nice! Can we bump it back to the failing state just to see the issue in CI? I might just need to make sure it is allowed to run.

[jongbinjung]

@ajfriend Force-pushed to the commit at failing state. Waiting for approval to run tests.

[ajfriend]

Hmmm. Not clear why coverage isn't working, especially since it seems to be working on #409

[jongbinjung]

Hmmm. Not clear why coverage isn't working, especially since it seems to be working on #409

Seems like a known issue with codecov action and PRs from a fork ...

[ajfriend]

OK. I think this is good to land, but it sounds like we need to figure out an alternative to codecov going forward: #393