#!/usr/bin/env python3.1
#
# $HeadURL: https://svn.spodhuis.org/ksvn/pdp-bincommon/puny $
# $Id: puny 458 2011-02-06 23:43:49Z pdp@SPODHUIS.ORG $

"""
puny: convert domain-names to and from punycode

Converts multiple domains from argv.

By default is verbose, takes the input and shows encoding and decoding.
Given -t/--to-puny will just show the results of converting to punycode.
Given -f/--from-puny will show just the results of converting from punycode.

Offer translations too.
"""

__author__ = 'syscomet@gmail.com (Phil Pennock)'

import codecs
import getopt
import json
import re
import sys
import urllib.parse
import urllib.request

GOOGLE_TRANSLATE_API_URL = 'https://www.googleapis.com/language/translate/v2'
GOOGLE_TRANSLATE_API_KEY = 'AIzaSyDv3RiIxCRSTJE7-NVroL6m6kK9OV0UHLQ'
HUMAN_TARGET_LANGUAGE = 'en'

IDN_PREFIX = 'xn--'
SAFE_LABEL_RE = re.compile(
# Python uses \Z, not \z
    r'^ (?: [A-Za-z0-9] (?: [A-Za-z0-9] | (?: - (?! -) ) )* )? [A-Za-z0-9] \Z',
    flags=re.VERBOSE)

puny_codec = codecs.lookup('punycode')
utf8_codec = codecs.lookup('utf-8')

def ToPuny(domain):
  components = domain.split('.')
  results = []
  for label in components:
    if SAFE_LABEL_RE.match(label):
      results.append(label)
    elif label.startswith(IDN_PREFIX):
      results.append(label)
    else:
      results.append( IDN_PREFIX + puny_codec.encode(label)[0].decode('ascii') )
  return '.'.join(results)

def FromPuny(domain):
  components = domain.split('.')
  results = []
  for label in components:
    if not label.startswith(IDN_PREFIX):
      results.append(label)
    else:
      results.append( puny_codec.decode(label[4:])[0] )
  return '.'.join(results)

def ParseMixedPuny(domain):
  """Domain may contain mixed unicode/punycode.  Parse it all, return tuple.

  Returns: tuple containing:
    0: unicode form
    1: punycode form
  """
  components = domain.split('.')
  if not components[-1]:
    # final dot for root anchor
    components.pop()
  puny_parts = []
  uni_parts = []
  for label in components:
    if label.startswith(IDN_PREFIX):
      l_puny, l_uni = label, puny_codec.decode(label[4:])[0]
    elif SAFE_LABEL_RE.match(label):
      l_puny, l_uni = label, label
    else:
      l_puny, l_uni = IDN_PREFIX + puny_codec.encode(label)[0].decode('ascii'), label
    puny_parts.append(l_puny)
    uni_parts.append(l_uni)
  return '.'.join(uni_parts), '.'.join(puny_parts)

def Translate(text):
  """Use Google Translate to get local language.

  Yields:
    2-tuple:
      translated text as unicode
      source language, or None
  """
  # API uses auto-detect source language if 'source' param not given
  # security of key in transit by using https
  urltext = urllib.parse.quote(utf8_codec.encode(text)[0])
  url = '{url}?key={key}&target={targetlang}&q={q}'.format(
      url=GOOGLE_TRANSLATE_API_URL,
      key=GOOGLE_TRANSLATE_API_KEY,
      targetlang=HUMAN_TARGET_LANGUAGE,
      q=urltext)
  response = urllib.request.urlopen(url)
  decode_charset = 'ASCII'
  m = re.search(r'(?i)\bcharset=([^\s;]+)', response.getheader('content-type'))
  if m:
    decode_charset = m.group(1)
  transblob = json.loads(codecs.lookup(decode_charset).decode(response.read())[0])
  if 'data' in transblob:
    if 'translations' in transblob['data']:
      for d in transblob['data']['translations']:
        lang = None
        if 'detectedSourceLanguage' in d:
          lang = d['detectedSourceLanguage']
        if 'translatedText' in d:
          yield (d['translatedText'], lang)

def VerboseDefault(domains):
  done_one = False
  for arg in domains:
    if done_one:
      print()
    done_one = True
    print('Input: ' + arg)
    res = ParseMixedPuny(arg)
    print('Unicode: {0}\nPunycode: {1}'.format(*res))
    for unilabel in res[0].split('.'):
      if re.search(r'[^A-Za-z0-9_-]', unilabel):
        for trans in Translate(unilabel):
          if trans[1] is not None:
            print('Translation: "{0}" [{2}] -> "{1}"'.format(unilabel, trans[0], trans[1]))
          else:
            print('Translation: "{0}" -> "{1}"'.format(unilabel, trans[0]))

def _main():
  options_list, domains = getopt.getopt(sys.argv[1:], 'ft', ['from-puny', 'to-puny'])
  from_puny, to_puny = False, False
  for opt in options_list:
    if opt[0] in ('-f', '--from-puny'):
      from_puny = True
    elif opt[0] in ('-t', '--to-puny'):
      to_puny = True
    else:
      raise Exception('Unhandled option "{0}"'.format(opt[0]))

  if from_puny and to_puny:
    raise Exception('Converting both from and to punycode is unsupported')

  if from_puny:
    for arg in domains:
      print(FromPuny(arg))
  elif to_puny:
    for arg in domains:
      print(ToPuny(arg))
  else:
    VerboseDefault(domains)

if __name__ == '__main__':
  _main()

# vim: set ft=python sw=2 expandtab: