DOCS: Update import statements in docstrings

[alphaCTzo7G]

xref #21731 (review)

• ./pandas/plotting/_converter.py|134 col 17| ">>> from pandas.plotting import register_matplotlib_converters"
• pandas/plotting/_misc.py|501 col 5| >>> from pandas import read_csv
• pandas/plotting/_misc.py|502 col 5| >>> from pandas.tools.plotting import parallel_coordinates
• ./pandas/util/testing.py|602 col 5| >>> from pandas.util.testing import capture_stdout DOC: updates documentations Closes #21749 #42979
• ./pandas/util/testing.py|650 col 5| >>> from pandas.util.testing import capture_stderr DOC: updates documentations Closes #21749 #42979
• ./pandas/util/testing.py|2213 col 7| >>> from pandas.util.testing import network DOC: updates documentations Closes #21749 #42979
• ./pandas/util/testing.py|2214 col 7| >>> from pandas.io.common import urlopen DOC: updates documentations Closes #21749 #42979
• ./pandas/tseries/holiday.py|145 col 9| >>> from pandas.tseries.holiday import Holiday, nearest_workday DOC: Update import statement #45485
• ./pandas/tseries/holiday.py|146 col 9| >>> from pandas import DateOffset
• ./pandas/core/sparse/series.py|739 col 9| >>> from numpy import nan
• pandas/core/frame.py|2844 col 9| >>> from numpy.random import randn
• pandas/core/frame.py|2845 col 9| >>> from pandas import DataFrame
• ./pandas/core/algorithms.py|1519 col 5| >>> from pandas.api.extensions import take
• ./pandas/core/dtypes/concat.py|285 col 5| >>> from pandas.api.types import union_categoricals Ko ya346/update import doc #45346
• ./pandas/io/pytables.py|457 col 5| >>> from pandas import DataFrame
• ./pandas/io/pytables.py|458 col 5| >>> from numpy.random import randn
• ./pandas/io/json/normalize.py|138 col 5| >>> from pandas.io.json import json_normalize
• pandas/io/stata.py|2713 col 5| >>> from pandas.io.stata import StataWriter117 Ko ya346/update import doc #45346

[TomAugspurger]

What's this for?

[alphaCTzo7G]

This is related to this.. #21731 (comment)

[gfyoung]

@TomAugspurger : I've updated the issue to clarify the purpose.

[alphaCTzo7G]

I modified the import statements in the following files from the list in the OP and some others:

MODIFIED
    ./pandas/io/pytables.py|457 col 5| >>> from pandas import DataFrame
    ./pandas/io/pytables.py|458 col 5| >>> from numpy.random import randn
    ./pandas/core/sparse/series.py|739 col 9| >>> from numpy import nan
    ./pandas/tseries/holiday.py|146 col 9| >>> from pandas import DateOffset

but kept these intact..

./pandas/plotting/_converter.py|134 col 17| ">>> from pandas.plotting import register_matplotlib_converters"

    pandas/plotting/_misc.py|502 col 5| >>> from pandas.tools.plotting import parallel_coordinates
    ./pandas/util/testing.py|602 col 5| >>> from pandas.util.testing import capture_stdout
    ./pandas/util/testing.py|650 col 5| >>> from pandas.util.testing import capture_stderr
    ./pandas/util/testing.py|2213 col 7| >>> from pandas.util.testing import network
    ./pandas/util/testing.py|2214 col 7| >>> from pandas.io.common import urlopen
    ./pandas/tseries/holiday.py|145 col 9| >>> from pandas.tseries.holiday import Holiday, nearest_workday
    ./pandas/core/algorithms.py|1519 col 5| >>> from pandas.api.extensions import take
    ./pandas/core/dtypes/concat.py|285 col 5| >>> from pandas.api.types import union_categoricals
    ./pandas/io/json/normalize.py|138 col 5| >>> from pandas.io.json import json_normalize
    pandas/io/stata.py|2713 col 5| >>> from pandas.io.stata import StataWriter117

Do you think we should
change these as well to the form import pandas as pd; pd.... If we do, it wil
result in long function names as in module.class.function(args)

These were modified in a different PR

    pandas/core/frame.py|2844 col 9| >>> from numpy.random import randn
    pandas/core/frame.py|2845 col 9| >>> from pandas import Dataframe

[jbrockmendel]

Do we have an idea which of these are still relevant?

[salomondush]

Take

[salomondush]

Hello, I was wondering if it's okay to take on this issue because I noticed that some statements still use the old form