""" 
 utilties.py

 Jim Mahoney | cs.marlboro.college | March 2020
"""
import csv
from scratch.statistics import mean, standard_deviation

def read_housing(filename='housing.data'):
    """ Return housing table as a list of lists of floats """
    data = []
    with open(filename) as datafile:
        reader = csv.reader(datafile, delimiter=' ')
        for raw_row in reader:
            row = [float(x) for x in raw_row if x != '']
            #print(row)
            data.append(row)
    return data

def try_float(x):
    """ Return float(x) or x """
    try:
        result = float(x)
    except:
        result = x
    return result

def read_abalone(filename='abalone.data'):
    """ Return abalone data as a list of list of floats or strings """
    data = []
    with open(filename) as datafile:
        reader = csv.reader(datafile, delimiter=',')
        for raw_row in reader:
            row = [try_float(x) for x in raw_row]
            data.append(row)
    return data    

def normalize(xs):
    """ Given xs=(list of floats) i.e. "plural of x"
        subtract mean, divide by standard deviation, 
        return list of floats """
    average = mean(xs)
    sigma = standard_deviation(xs)
    return [(x - average)/sigma for x in xs]

def normalize_column(data, col):
    """ modify data in place by normalizing one of its columns """
    n_rows = len(data)
    xs = [data[row][col] for row in range(n_rows)]
    xs_norm = normalize(xs)
    for row in range(n_rows):
        data[row][col] = xs_norm[row]

def get_features(data, row, col_features):
    """ return freatures for given row, values [x1, x2, ...] """
    return [data[row][col] for col in col_features]

def tests():
    data = read_data()
    for row in data:
        row_string = [f'{x:6.2f}' for x in row]
        print(' '.join(row_string))

if __name__ == '__main__':
    tests()