BUG: algorithms._count_generic?

[dsm054]

algorithms._count_generic returns a Series built from counts, which isn't defined:

def _count_generic(values, table_type, type_caster):
    from pandas.core.series import Series

    values = type_caster(values)
    table = table_type(min(len(values), 1000000))
    uniques, labels = table.factorize(values)

    return Series(counts, index=uniques)

but this function doesn't seem to be used anywhere. The only reference I can find is in algorithms.count, or what would be algorithms.count, but that's commented out:

# def count(values, uniques=None):
#     f = lambda htype, caster: _count_generic(values, htype, caster)

#     if uniques is not None:
#         raise NotImplementedError
#     else:
#         return _hashtable_algo(f, values.dtype)

[jreback]

this I elieve was cythonized (as all the count function are)
in src/gererated.pyx

it's possible this is an older version

[dsm054]

So would it be okay to remove both _count_generic and #count from algorithms as vestigial?

[jreback]

if they aren't referred and don't break any tests
(it's possible they t not not tested I guess) so be careful

[jreback]

@dsm054 can you do this when you have a chance? if nothng breaks when you remove

[jreback]

@dsm054 ?