Python CSV filter: replaces repeating values in columns with spaces. Originally, used to clean csv files generated with multiindex pandas dataframe to_csv() method.

clean-csv.py

# Clean csv with multiindices
from sys import argv

# Command line arguments:
#   original csv file
#   number of columns to check (optional)

with open(argv[1], 'r') as f:
    if len(argv) > 2:
        n = int(argv[2])
        i0 = ('', ) * n
    else:
        n = None
        i0 = None
    for l in f:
        if n is None:
            # define n from the 1-st line
            i1 = l.split(',')
            n = len(i1)
            i0 = ('', ) * n
        i1 = l.split(',', n)
        nl = ()
        for s0, s1 in zip(i0, i1):
            if s0 == s1:
                e = ' ' * len(s0)
            else:
                e = s1
            nl += (e, )
        nl += tuple(i1[n:])
        print ','.join(nl),
        i0 = i1[:n]