special-characters.py

#!/usr/bin/env python
import re
import sys

with open(sys.argv[1]) as f:
  lines = f.readlines()

rx = re.compile(r"[\w\s]")

d = {}
for l in lines:
  if l.startswith('#'): continue
  alpha = len(rx.findall(l))
  nonAlpha = len(l) - alpha
  d[l] = nonAlpha / (1.0 * len(l))

xs = sorted(d.items(), key=lambda x:x[1])
xs.reverse()
for i in range(20):
  print(xs[i])