Check and sort a translation of the PDFreaders petition.

petition.py

#! /usr/bin/env python

# Copyright (C) 2011 Heiki "Repentinus" Ojasild.

# Copying and distribution of this file, with or without modification,
# are permitted in any medium without royalty provided the copyright
# notice and this notice are preserved. This file is offered as-is,
# without any warranty.

# This is used to check that the translator has not introduced an error to
# the petition country list. It also sorts the countries alphabetically.
# Usage: script petition.??.xhtml petition.en.xhtml locale

# The sorted output will be written into petition.XX.xhtml

from functools import cmp_to_key
from lxml import etree
import locale
import sys

if len(sys.argv) != 4:
	print("This script takes three command line arguments.")
	exit()

parser = etree.XMLParser(remove_blank_text=True)
try:
	translation = etree.parse(sys.argv[1], parser)
	original = etree.parse(sys.argv[2], parser)
except:
	print("An error has occurred while processing input.")
	exit()

for i in translation.iter():
	if i.text:
		i.text = ' '.join(i.text.split())

authoritative = []
for i in original.iter("select"):
	if i.get("id") == "country":
		for j in i.iter("option"):
			if j.get("value"):
				authoritative.append(j.get("value"))
			else:
				authoritative.append("" + j.text)
		break

neu = []

def mcmp(a, b):
	return locale.strcoll(a[0], b[0])

for i in translation.iter("select"):
	if i.get("id") == "country":
		for j in i.iter("option"):
			if j.get("value") in authoritative:
				neu.append((j.text, j))
				i.remove(j)
			else:
				print(j.get("value") + " is missing from the authoritative list.")
		locale.setlocale(locale.LC_ALL, sys.argv[3])
		neu = sorted(neu, key=cmp_to_key(mcmp))
		for j in neu:
			i.append(j[1])

translation.write("petition.XX.xhtml", encoding="UTF-8", xml_declaration=True, method="xml", pretty_print=True)