wrong index name during read_csv if using usecols

[tesla1060]

Hi,

If I have a test.csv file looks like below

SecId    Time               Price    P2    P3
10000   2013-5-11       100      10     1
500       2013-5-12     101      11     1

if I read in the above file with the following code

df =pd.read_csv('test.csv', usecols=['Time', 'Price'], parse_dates=True, index_col=0)

then df.index.name will be SecId instead of Time

[hayd]

Why not use index_col='Time' ?

What pandas version are you using? I get:

Exception: usecols not supported with engine='python' or multicharacter separators (yet).

#2733 and #3192

[cpcloud]

if you replace the spaces with , shouldn't this work? maybe close since this is a dupe?

[paulgb]

This is not a dupe of #2733, it also occurs when the separator is a comma. I've added a test case to my fork (using tesla's data).

I believe the issue is related to the order in which the index_col and usecols operations take place, but I'm not familiar enough with the code yet to fix it.