array_equivalent has disappeared from pandas.core.common in version 0.19, is this a mistake?

[cleemesser]

A small, complete example of the issue

# Your code here
import pandas
import pandas.core.common as com
print com.array_equivalent

Expected Output

# expect something like:
# <function array_equivalent at 0x000000000445A198>
# But I get

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-2-fc6fc07906df> in <module>()
----> 1 print com.array_equivalent

AttributeError: 'module' object has no attribute 'array_equivalent'

Output of pd.show_versions()

# Paste the output here pd.show_versions() here INSTALLED VERSIONS ------------------ commit: None python: 2.7.11.final.0 python-bits: 64 OS: Windows OS-release: 7 machine: AMD64 processor: Intel64 Family 6 Model 60 Stepping 3, GenuineIntel byteorder: little LC_ALL: None LANG: en_US.UTF-8 LOCALE: None.None

pandas: 0.19.0
nose: 1.3.7
pip: 8.1.2
setuptools: 23.0.0
Cython: 0.24.1
numpy: 1.11.1
scipy: 0.17.1
statsmodels: 0.6.1
xarray: None
IPython: 4.1.2
sphinx: 1.3.5
patsy: 0.4.0
dateutil: 2.5.1
pytz: 2016.2
blosc: None
bottleneck: 1.0.0
tables: 3.2.2
numexpr: 2.6.1
matplotlib: 1.5.1
openpyxl: 2.3.2
xlrd: 0.9.4
xlwt: 1.0.0
xlsxwriter: 0.8.4
lxml: 3.6.0
bs4: 4.4.1
html5lib: None
httplib2: None
apiclient: None
sqlalchemy: 1.0.12
pymysql: None
psycopg2: None
jinja2: 2.8
boto: 2.39.0
pandas_datareader: None

[jreback]

you can import it from pandas.types.missing. its not really a public method. though @jorisvandenbossche and I did have this discussion.

[jorisvandenbossche]

Well, there are two aspect about this:

First, that we didn't add a deprecation warning for array_equivalent (as we did for the other functions in pd.core.common. This can be easily fixed.

Secondly about the fact that we currently don't plan to make this public (opposed to other functions in pd.core.common), and so it is not recommended to use it (although you can now import it from pd.types.missing, we don't guarantee this will keep working).
But if it is regarded as useful functionality, we can consider to making it public.

[tomspur]

array_equivalent is already used in the docs, which fails now with a traceback:

[jorisvandenbossche]

@tomspur yes, we are aware of that (and I forgot to remove it from the docs for the 0.19.0 release)

[jorisvandenbossche]

This functionality should actually be rather available from numpy. The problem is, however, that np.array_equal or np.array_equiv do not handle NaNs well.
But I see that numpy 1.10 added a equal_nan keyword to np.allclose, so you can do:

>>> np.allclose([1.0, np.nan], [1.0, np.nan], equal_nan=True)
True

https://docs.scipy.org/doc/numpy/reference/generated/numpy.allclose.html

We can probably recommend that as alternative

[jreback]

i don't think that handles object dtypes (with or w/o nans)

[jorisvandenbossche]

Ah, yes, and also no datetime64 values. OK, clearly not a full replacement