BUG/DOC: Add documentation in types/common.py #15941
Conversation
gfyoung
force-pushed
the
pandas-types-doc
branch
2 times, most recently
from
3ac9d08 to
986d8b6
Compare
| dtype = _get_dtype(arr_or_dtype) | ||
| return dtype.kind in ('O', 'S', 'U') and not is_period_dtype(dtype) | ||
| """ | ||
| Check whether the provided array or dtype is of the string dtype. |
There was a problem hiding this comment.
see comments here: #15585
maybe add a reference / comments
gfyoung
changed the title
gfyoung
force-pushed
the
pandas-types-doc
branch
from
986d8b6 to
5fcc3cf
Compare
Codecov Report
@@ Coverage Diff @@
## master #15941 +/- ##
==========================================
+ Coverage 90.98% 90.99% +<.01%
==========================================
Files 145 145
Lines 49556 49565 +9
==========================================
+ Hits 45090 45101 +11
+ Misses 4466 4464 -2
Continue to review full report at Codecov.
|
pandas/types/common.py
Outdated
|
|
||
| Returns | ||
| ------- | ||
| is_orderable_exc : Whether or not the exception raised is an |
There was a problem hiding this comment.
I think better to use say:
Returns
--------
boolean : .......
(for all of these)
pandas/tests/types/test_common.py
Outdated
| # | ||
| # No exception should be raised. | ||
|
|
||
| assert not com.is_string_dtype(None) |
There was a problem hiding this comment.
can you add all of these (use parametrize / construct them)
gfyoung
force-pushed
the
pandas-types-doc
branch
from
5fcc3cf to
9777867
Compare
pandas/tests/types/test_common.py
Outdated
|
|
||
|
|
||
| get_dtype_funcs = [ | ||
| (com.is_dtype_equal, 2), |
There was a problem hiding this comment.
what I meant was to add all of the validation functions.
(In [3]: [ f for f in dir(com) if f.startswith('is_') and f.endswith('dtype')]
Out[3]:
['is_any_int_dtype',
'is_bool_dtype',
'is_categorical_dtype',
'is_complex_dtype',
'is_datetime64_any_dtype',
'is_datetime64_dtype',
'is_datetime64_ns_dtype',
'is_datetime64tz_dtype',
'is_datetime_or_timedelta_dtype',
'is_float_dtype',
'is_floating_dtype',
'is_int64_dtype',
'is_int_or_datetime_dtype',
'is_integer_dtype',
'is_numeric_dtype',
'is_object_dtype',
'is_period_dtype',
'is_signed_integer_dtype',
'is_string_dtype',
'is_string_like_dtype',
'is_timedelta64_dtype',
'is_timedelta64_ns_dtype',
'is_unsigned_integer_dtype']
then is_dtype_equal should be a separate function to test (as its different)
There was a problem hiding this comment.
could make this a nice fixture :>
There was a problem hiding this comment.
Well the whole point was to check all of the functions that _get_dtype. The ones in the list currently are the only ones that do.
There was a problem hiding this comment.
could make this a nice fixture :>
Perhaps, if it's needed 😄
There was a problem hiding this comment.
hmm, well we should check them all anyhow :>
There was a problem hiding this comment.
Fair enough. Will do then.
Partially addresses pandas-devgh-15895.
The following functions were not catching the TypeError raised by _get_dtype: 1) is_string_dtype 2) is_string_like_dtype 3) is_timedelta64_ns_dtype Thus, when "None" was passed in, an Exception was raised instead of returning False, as other functions did.
gfyoung
force-pushed
the
pandas-types-doc
branch
from
9777867 to
327687c
Compare
|
thanks @gfyoung |
Adds documentation for all internal functions in
types/common.pyIn addition, caught a bug in which some functions calling
_get_dtypewere not catching theTypeError. Documented those functions along the way too.Partially addresses #15895.