#!/usr/bin/env python3

import re
import sys
import json
from locale import getpreferredencoding
from urllib.request import urlopen, Request
from urllib.parse import urlencode
from argparse import ArgumentParser


def die(msg, *args):
	sys.stderr.write((msg % args) + '\n')
	sys.exit(1)


def translate(text, dst_lang, src_lang='auto'):
	params = dict(
			client   = 't',
			hl       = 'en',
			multires = 1,
			otf      = 2,
			pc       = 1,
			ssel     = 0,
			tsel     = 0,
			sc       = 1,
			text     = text,
			sl       = src_lang,
			tl       = dst_lang,
		)
	url = 'http://translate.google.com/translate_a/t'
	data = ('client=t&' + urlencode(dict(text=text)) + '&hl=en&' +
			urlencode(dict(sl=src_lang)) + '&' +
			urlencode(dict(tl=dst_lang)) +
			'&multires=1&otf=2&pc=1&ssel=0&tsel=0&sc=1')
	headers = { 'User-agent': 'Mozilla/5.0' }

	req = Request(url=url, data=data.encode('utf-8'),
			headers=headers)
	page = urlopen(req).read().decode('utf-8')
	# attempt to fix invalid json (will break translated text that had ,,+)
	page = re.sub(',,+', ',', page)

	try:
		result = json.loads(page)
	except Exception as e:
		die("Error parsing results: %s", e)

	return ''.join((c[0] for c in result[0]))


def parse_args():
	parser = ArgumentParser(description='Translate text using '
			'Google Translate.')
	parser.add_argument('--from','-f', dest='from_', metavar='LANG',
			default='auto',
			help='Translate from LANG language '
				'(e.g. en, de, es, default: auto)')
	parser.add_argument('--to', '-t', metavar='LANG', default='en',
			help='Translate to LANG language '
				'(e.g. en, de, es, default: en)')
	parser.add_argument('--input-file', '-i', metavar='FILE',
			help='Get text to translate from FILE instead of stdin')
	parser.add_argument('--output-file', '-o', metavar='FILE',
			help='Output translated text to FILE instead of stdout')
	parser.add_argument('--input-encoding', '-I', metavar='ENC',
			default='LOCALE,utf-8,iso-8859-15',
			help='Use ENC caracter encoding to read the input, '
				'can be a comma separated list of encodings to '
				'try, LOCALE being a special value for the '
				'user\'s locale-specified preferred encoding '
				'(default: LOCALE,utf-8,iso-8859-15)')
	parser.add_argument('--output-encoding', '-O', metavar='ENC',
			default='LOCALE',
			help='Use ENC caracter encoding to write the output '
				'(default: LOCALE)')

	args = parser.parse_args()

	args.input = sys.stdin
	if args.input_file:
		try:
			args.input = open(args.input_file)
		except Exception as e:
			die("Can't open file %s for reading: %s",
					args.input_file, e)

	args.output = sys.stdout
	if args.output_file:
		try:
			args.output = open(args.output_file, 'w')
		except Exception as e:
			die("Can't open file %s for writing: %s",
					args.output_file, e)

	args.input_encoding = args.input_encoding.split(',')
	if not args.input_encoding:
		args.input_encoding = ['LOCALE']
	args.input_encoding = [getpreferredencoding() if e == 'LOCALE' else e
			for e in args.input_encoding]

	if args.output_encoding == 'LOCALE':
		args.output_encoding = getpreferredencoding()

	return args


args = parse_args()

text = args.input.buffer.read()
err = None
for enc in args.input_encoding:
	try:
		text = text.decode(enc)
		break
	except UnicodeError as e:
		err = e
else:
	die("Can't decode input (tried these encodings: %s). Last error: %s",
			args.input_encoding, err)

trans_text = translate(text, args.to, args.from_) + '\n'

args.output.buffer.write(trans_text.encode(args.output_encoding))