BUG: _validate_setitem_value fails to raise for PandasArray #51044 #54575
Conversation
|
I would like to know if I'm going in the right direction/if the logic needs to be changed |
pandas/tests/arrays/test_array.py
Outdated
| @@ -444,3 +444,11 @@ def test_array_to_numpy_na(): | |||
| result = arr.to_numpy(na_value=True, dtype=bool) | |||
| expected = np.array([True, True]) | |||
| tm.assert_numpy_array_equal(result, expected) | |||
|
|
|||
|
|
|||
| def test_array_validate_setitem_value(): | |||
There was a problem hiding this comment.
this can go in tests.arrays.numpy_.test_numpy.py
pandas/tests/arrays/test_array.py
Outdated
| def test_array_validate_setitem_value(): | ||
| # Issue# 51044 | ||
| arr = pd.Series(range(5)).array | ||
| with pytest.raises(TypeError, match="bad"): |
There was a problem hiding this comment.
can you put pytest.raises around each of the bad calls below. otherwise when the first one raises, the second one doesnt get tested
pandas/tests/arrays/test_array.py
Outdated
| arr = pd.Series(range(5)).array | ||
| with pytest.raises(TypeError, match="bad"): | ||
| arr._validate_setitem_value("foo") | ||
| arr._validate_setitem_value(1.5) |
There was a problem hiding this comment.
can you test that 1.5 doesn't raise on float/complex-dtype
pandas/core/arrays/numpy_.py
Outdated
| if lib.is_integer(value) or lib.is_float(value): | ||
| return value | ||
| else: | ||
| if lib.is_integer(value) or (lib.is_float(value) and value.is_integer()): |
There was a problem hiding this comment.
this is only for integer-dtype. also you'll need to check itemsize to be sure it fits. there is probably something in core.dtypes.cast (maybe np_can_hold_element?) that can be reused here
pandas/core/arrays/numpy_.py
Outdated
|
|
||
| elif kind == "f": | ||
| if lib.is_integer(value) or lib.is_float(value): | ||
| return value |
There was a problem hiding this comment.
these checks are correct for scalar value, not sure if validate_setitem_value needs to handle listlike?
There was a problem hiding this comment.
The TypeError still occurs if listlike values are entered in to validate causing tests to fail, even if it's supposed to work, see pandas/tests/arrays/numpy_/test_numpy.py test_ufunc_unary
A bunch of tests are failing on the CI. This is the first place to look for potential problems. |
mroeschke
added
Error Reporting
pandas/core/arrays/numpy_.py
Outdated
| else: | ||
| if self.dtype == NumpyEADtype("object") or self.dtype is None: | ||
| return value | ||
| if NumpyEADtype(type(value)) != self.dtype: |
There was a problem hiding this comment.
I think you can just add a check for NumpyEADtype(type(value)) == self.dtype as an or when you do NumpyEADtype(type(value)) == NumpyEADtype(self.dtype)
There was a problem hiding this comment.
Then you can just raise a TypeError without any more checks
|
Hi @mroeschke are you able to provide any guidance here? |
|
This pull request is stale because it has been open for thirty days with no activity. Please update and respond to this comment if you're still interested in working on this. |
|
Thanks for the pull request, but it appears to have gone stale. If interested in continuing, please merge in the main branch, address any review comments and/or failing tests, and we can reopen. |
doc/source/whatsnew/vX.X.X.rstfile if fixing a bug or adding a new feature.