namedtuple_reader.py

"""Read columns of a delimited file into a namedtuple.

Supports column type parsers, e.g.

def parser(colname, data):
    if colname == "foo":
        return map(float, data)
    return data

"""
import csv
import collections as c


def read_namedtuple(csvfile, column_names=None, parser=None, **kwargs):
    """Read columns of a delimited file into nested tuples in a named tuple [1].

    Optionally specify column names. If None, attempt to parse column
    names from the first line of the file.

    Optionally specify a column parser. The callable must take two
    positional arguments, which are the column name and the column data,
    respectively. The callable must return an iterable of the processed
    column data.
    
    Pass any additional keyword arguments to `csv.reader` [2].

    [1]: <https://docs.python.org/3/library/collections.html#collections.namedtuple>
    [2]: <https://docs.python.org/3/library/csv.html#csv.reader>
    
    """
    with open(csvfile) as file:
        reader = csv.reader(file, **kwargs)
        if column_names is None:
            cols = [s.replace(' ', '_') for s in next(reader)]
        else:
            cols = column_names

        Columns = c.namedtuple("Columns", cols)
        if parser is None:
            return Columns._make([tuple(x) for x in zip(*list(reader))])
        return Columns._make([tuple(parser(c, x)) for c, x in zip(cols, zip(*list(reader)))])