Calculate matrix difference between files. Used to see which student submissions are line-for-line copies of their peer's work.

submission_diff.py

import difflib
import sys
import csv

output = 'diff_report.csv'
files = sys.argv[1:]

with open(output, 'wb') as csvfile:

  foo = csv.writer(csvfile)
  foo.writerow(['-'] + files)

  def go(f,j):
    if f == j:
      return '-'
    else:
      return round(difflib.SequenceMatcher(None, open(f).read(), open(j).read()).ratio(), 2)

  for f in files:
    foo.writerow([f] + map(lambda x: go(f,x), files))