Detect and wrap roman numerals with a special "small-caps" span tag

roman-numerals.py

#!/usr/bin/env python

import sys
import re

re_full = r'(MMM|MM|M)?(CM|CD|DCCC|DCC|DC|D|CCC|CC|C)?(XC|XL|LXXX|LXX|LX|L|XXX|XX|X)?(IX|IV|VIII|VII|VI|V|III|II|I)?'
re_appr = r'\b[MDCLXVI]+\b'

def wrap_roman(matchobj):
    s = matchobj.group(0)
    m = re.match(re_full, s)
    g = m.group(0) if m != None else ''
    if len(s) > 0 and len(g) == len(s):
        return '<span class="small-caps">' + g + '</span>'
    return s

for line in sys.stdin:
  print re.sub(re_appr, wrap_roman, line),