Script to correct to manipulate CBCL data file

fix_cbcl_data.py

#!/usr/bin/env python

'''
usage: %s input_file output_file

'''

import collections
import csv
import sys


def split_column(row, col, name_format):
	for n, c in enumerate(row[col][1:]):
		try:
			name = name_format % (int(n) + 1)
			row[name] = c
		except:
			# Ignore errors from trying to cast to an int, to skip any
			# stray characters like the leading '.
			pass
	del row[col]


def expand_row(row):
	row = collections.defaultdict(lambda: '', row)
	split_column(row, 'bpitems', 'bpitem%03d')
	split_column(row, 'compitems', 'compitem%02d')
	return row


def main(rf, wf):
	reader = csv.DictReader(rf)
	writer = csv.writer(wf)
	columns = None

	for row in reader:
		expanded_row = expand_row(row)

		if columns is None:
			columns = sorted(expanded_row.keys())
			# Move id to first column
			columns.remove('id')
			columns.insert(0, 'id')

			# Write a header line
			writer.writerow(columns)

		try:
			values = [expanded_row[c] for c in columns]
			writer.writerow(values)
		except:
			pass


if __name__ == '__main__':
    if len(sys.argv) < 3:
        sys.stderr.write(__doc__ % sys.argv[0])
        sys.exit(1)

    with open(sys.argv[1], 'rU') as rf, open(sys.argv[2], 'w') as wf:
        main(rf, wf)