Code Sample, a copy-pastable example if possible
import pandas as pd
empty_frame=pd.read_excel('pandas/tests/io/data/test1.xlsx', parse_cols=['A','B','C'])
print(empty_frame)
useful_frame=pd.read_csv('pandas/tests/io/data/test1.csv', usecols=['A','B','C'])
print(useful_frame)
Problem description
based on making the argument naming consistent for various read_*
functions (#4988), the functionality should also be consistent as well. ideally, keeping the read_excel ability to parse a string as well. a continuation of #16488.
Expected Output
expected output to using parse_col=
/usecols=['A','B','C']
is a dataFrame containing the data in excel columns A, B, & C.
Output of pd.show_versions()
Comment From: jreback
how is this different than #4988 ?
Comment From: abarber4gh
in #4988, @jreback requested "needs to be in a separate PR from the deprecation change." my original PR for #4988 was to change the keyword (from parse_cols
to usecols
) to match other read_*
functions and implement this functionality. this issue was created to separate the kwarg change from the functionality change.
Comment From: jreback
@abarber4gh but that is what the original issue is about. So unless you have something different, this is a duplicate issue.
Comment From: gfyoung
@jreback : I agree. Closing as such unless otherwise noted.