Format CSV to work better with unix column command: - Replace commas with custom separator (except commas between double quotes) - Collapse extra whitespace, including newlines between double quotes

csv_formatter.py

import argparse, re

# Arg Parser
parser = argparse.ArgumentParser(description='Prints well-formatted version of supplied CSV file to STDOUT.')
parser.add_argument('filename', metavar='FILE', type=str,
                    help='File to convert.')
parser.add_argument('-s', '--separator', dest='separator', type=str, default='|',
                    help="Character to separate fields. Default: '|' (pipe).")

# Args
args = parser.parse_args()
filename = args.filename
separator = args.separator[0]

# Replace separators
file = open(filename, 'r')
text = re.sub(
    r'((?:[^,"]|"[^"]*")*),',
    r' \1 %c ' % separator,
    file.read(),
)

# Collapse spaces
text = re.sub(r' {2,}', r' ', text)

# Replace quoted newlines
spans = []
for m in re.finditer(r'"[^"]*"', text):
    spans.append(m.span())
text = list(text)
for span in reversed(spans):
    index = span[0];
    while index < span[1]:
        if text[index] == '\n':
            text[index] = ' '
        index = index + 1

print "".join(text)