Fix failure to convert string "uint64" to NaN #32541
Conversation
Including regression test
roberthdevries
force-pushed
the
fix-32394-to-numeric-fail-string-uint64
branch
from
955550d to
0240f9c
Compare
|
Great thanks @roberthdevries - keep the PRs coming! |
| result = lib.maybe_convert_numeric( | ||
| np.array(["uint64"], dtype=object), set(), coerce_numeric=True | ||
| ) | ||
| assert np.isnan(result) |
There was a problem hiding this comment.
this is not a good test - does it replicate what the OP was? this is not user facing at all
pls follow what is going on in other tests
There was a problem hiding this comment.
The OP currently takes this path, though it'd be good to test at the pd.to_numeric level too.
|
@WillAyd this has a number of issues in testing pls revert the OP is not tested at all |
| result = lib.maybe_convert_numeric( | ||
| np.array(["uint64"], dtype=object), set(), coerce_numeric=True | ||
| ) | ||
| assert np.isnan(result) |
There was a problem hiding this comment.
The OP currently takes this path, though it'd be good to test at the pd.to_numeric level too.
|
right just need another PR; @roberthdevries can u push a test for to_numeric @WillAyd remember to set the version in the PR and issue before merging |
|
@jreback I will add another PR with a test for to_numeric referencing the original issue |
Including regression test
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diff