locale-escape

#!/usr/bin/env python2

# Disclaimer: I don't remember authoring this script.  Let me know if you want
# to claim credit for it.

import sys
import re

lc_identification = re.compile("^LC_IDENTIFICATION")
lc_end_identification = re.compile("^END\s+?LC_IDENTIFICATION")
lc_all = re.compile("^LC_")
lc_end_all= re.compile("^END\s+?LC_")
blank = re.compile("^\s*$")
string_segment = re.compile('"\;"|"')

comment = re.compile("^(%|#)")
copy_entity = re.compile("^copy")

isLC_IDENTIFICATION=False
process=False

for line in sys.stdin:
  if comment.search(line):
    print line,
    continue
  if lc_identification.search(line):
    isLC_IDENTIFICATION=True
    print line,
    continue
  if lc_end_identification.search(line):
    isLC_IDENTIFICATION=False
    print line,
    continue
  if isLC_IDENTIFICATION:
    print line,
    continue
  if lc_all.search(line):
    process=True
    print line,
    continue
  if lc_end_all.search(line):
    process=False
    print line,
    continue
  if copy_entity.search(line):
    print line,
    continue
  if process and not blank.search(line) and not re.match("^\w+?\s*\d*(;\d+)*$", line):
    # Phew now we can encode
    for string in string_segment.split(re.sub('^[^"]*?"', "", re.sub('"[^"]*?$', "", re.sub('/$|";/$', '"', line)))):
      string = re.sub('\n', "", string)
      newstring = ""
      if not blank.search(string):
        newword = string.decode('utf-8')
        for char in re.sub("<U[\dA-F]{4,4}>", "", newword):
          encode = "<U%04X>" % (ord(char))
          newword = re.sub(char, encode, newword)
        newstring = newstring + newword.encode('utf-8')
      line = re.sub(string, newstring, line)
      #print "'%s' %s\n%s\n" % (string, newstring, line)
    print line,
  else:
    print line,