BUG: read_csv not applying dtype to index col

[phofl]

• closes index_col in read_csv and read_table ignores dtype argument #9435
• tests added / passed
• Ensure all linting tests pass, see here for how to run them
• whatsnew entry

_agg_index and _convert_to_ndarryays need a refactor in the future, they both contain more or less the same casting logic now

[jreback]

_agg_index and _convert_to_ndarryays need a refactor in the future, they both contain more or less the same casting logic now

sgtm

[gaow]

Unfortunately, it does not work for cases where dtype is specified globally:

>>> import pandas as pd
>>> from io import StringIO
>>> data = "1,a\n2,b"
>>> df = pd.read_csv(StringIO(data), index_col=0, dtype=str, header=None)
>>> df.index
Int64Index([1, 2], dtype='int64', name=0)

whereas this works

>>> df = pd.read_csv(StringIO(data), index_col=0, dtype={0:str}, header=None)
>>> df.index
Index(['1', '2'], dtype='object', name=0)

However i need to set all columns str globally. So I expect the first chunk of code works and turns index into str.

[phofl]

Hm I think I prefer the current behavior. But you could open an issue to gather feedback from others.

[phofl]

One additional note: If we should decide to change that, we need a deprecation cycle.