pandas select

from functools import reduce
from operator import and_, or_

def select(df, **kwargs):
    '''Builds a boolean array where columns indicated by keys in kwargs are tested for equality to their values.
    In the case where a value is a list, a logical or is performed between the list of resulting boolean arrays.
    Finally, a logical and is performed between all boolean arrays.
    '''

    res = []
    for k, v in kwargs.items():

        # TODO: if iterable, expand with list(iter)

        # multiple column selection with logical or
        if isinstance(v, list):
            res_or = []
            for w in v:
                res_or.append(df[k] == w)
            res_or = reduce(lambda x, y: or_(x,y), res_or)
            res.append(res_or)

        # single column selection
        else:
            res.append(df[k] == v)

    # logical and
    if res:
        res = reduce(lambda x, y: and_(x,y), res)
        res = df[res]
    else:
        res = df

    return res