usecols is not respected when the file is empty

[VelizarVESSELINOV]

xref #9755

If the file is empty, it is logical to compensate with extra columns.

Code Sample, a copy-pastable example if possible

"""Example of Pandas bug."""
from pandas import read_csv

try:
    from StringIO import StringIO
except ImportError:
    from io import StringIO

name_list = ['Dummy', 'X', 'Dummy_2']
s = StringIO(',,')

df = read_csv(s, names=name_list, usecols=name_list[1:2], header=None)

print(list(df))

s = StringIO('')

df = read_csv(s, names=name_list, usecols=name_list[1:2], header=None)

print(list(df))

Expected Output

Current output:

['X']
['Dummy', 'X', 'Dummy_2']

Expected output:

['X']
['X']

output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.5.1.final.0
python-bits: 64
OS: Darwin
OS-release: 15.3.0
machine: x86_64
processor: i386
byteorder: little
LC_ALL: None
LANG: None

pandas: 0.17.1
nose: None
pip: 7.1.2
setuptools: 18.3.2
Cython: None
numpy: 1.10.1
scipy: 0.16.1
statsmodels: None
IPython: 4.0.1
sphinx: None
patsy: None
dateutil: 2.4.2
pytz: 2015.7
blosc: None
bottleneck: None
tables: None
numexpr: None
matplotlib: 1.5.0
openpyxl: 2.3.2
xlrd: None
xlwt: None
xlsxwriter: None
lxml: None
bs4: 4.4.1
html5lib: None
httplib2: None
apiclient: None
sqlalchemy: None
pymysql: None
psycopg2: None
Jinja2: 2.8
None

[jreback]

I agree, since the following is true. IOW an empty with names only produces an empty frame with the names; usecols just sub-selects.

In [19]: read_csv(StringIO(''), names=name_list, header=None)
Out[19]: 
Empty DataFrame
Columns: [Dummy, X, Dummy_2]
Index: []

[gfyoung]

FYI, if you try running that using engine='python', you get a StopIteration error when s = StringIO(''). Is that a bug as well?

[gfyoung]

I'll answer my own question: yes. If the C engine can do it, the Python one should be able to do it as well.

[VelizarVESSELINOV]

Thanks, much appreciated :)