Process Localise XLIFF exports for importing back into WPML by inserting translation job IDs from source files

translation-localize-to-wpml.py

# coding=utf-8

import os
import argparse
from glob import glob
import xml.etree.ElementTree as ET
import zipfile

namespace = 'urn:oasis:names:tc:xliff:document:1.2'
namespaces = {'xliff': namespace}
errors = 0

def get_current_language(filename):
  tree = ET.parse(filename)
  root = tree.getroot()

  target_lang = None

  # find current language
  for type_tag in root.findall('xliff:file', namespaces):
    target_lang = type_tag.get('target-language')
    #print('Language detected: {0}'.format(target_lang))
  return target_lang

def get_original(path, language): 
  global errors
  matches = 0
  attrib_original = ''

  print('Searching for source file with language {0}'.format(language))

  # search for files recursively
  result = [y for x in os.walk(path) for y in glob(os.path.join(x[0], '*.xliff'))]

  for file in result:
    if os.path.isdir(file):
      continue

    tree = ET.parse(file)
    root = tree.getroot()

    for type_tag in root.findall('xliff:file', namespaces):
      job = type_tag.get('original')
      target_lang = type_tag.get('target-language')

    #print(file, target_lang)
    if target_lang == language:
      matches += 1
      attrib_original = job
      print('Matching file found: {0} ({1}, {2})'.format(file, target_lang, attrib_original))

  if matches > 1:
    print('Error: Multiple matches ({0}), please make sure there are no duplicate languages (are you processing more than 1 page)?'.format(matches))
    errors += 1
    return None

  if matches == 1:
    return attrib_original

  print('Error: No matches found')
  errors += 1
  return None

def set_original(source_filename, target_filename, value):
  tree = ET.parse(source_filename)
  root = tree.getroot()

  target_lang = None

  # find and set original attribute
  for type_tag in root.findall('xliff:file', namespaces):
    job = type_tag.get('original')
    #print(job)
    type_tag.set('original', value)
  
  print('Writing {0}'.format(target_filename))
  tree.write(target_filename, encoding="utf-8", xml_declaration=True)

def create_archive(path, prefix):
  os.chdir(path)
  zip_filename = 'archive.zip'

  print('---')
  print('Writing {0}'.format(os.path.join(path, zip_filename)))

  zf = zipfile.ZipFile(zip_filename, mode='w')
  pattern = '*.xliff'
  if prefix != '':
    pattern = prefix + '-*.xliff'

  try:
    # search for files recursively
    result = [y for x in os.walk('./') for y in glob(pattern)]

    for file in result:
      if os.path.isdir(file):
        continue

      print('adding {0}'.format(file))
      zf.write(file)
  finally:
    zf.close()

def main():
  parser = argparse.ArgumentParser(description="Prepares XLIFF export from Lokalise for importing into WPML. Looks for XLIFF files in paths and generates an archive at the output path.")
  parser.add_argument('wpml_path', help='Directory containing XLIFF files exported using WPML')
  parser.add_argument('lokalise_path', help='Directory containing XLIFF files exported using Lokalise')
  parser.add_argument('output_path', nargs='?', default=os.getcwd(), help='XLIFF files and archive will be stored here')
  parser.add_argument('--output_prefix', default='REPLACED')
  args = parser.parse_args()

  # setup XML namespaces for XLIFF
  namespace = 'urn:oasis:names:tc:xliff:document:1.2'
  ET.register_namespace('', namespace)
  namespaces = {'xliff': namespace}

  # search for files recursively
  result = [y for x in os.walk(args.lokalise_path) for y in glob(os.path.join(x[0], '*.xliff'))]

  for file in result:
    if os.path.isdir(file):
      continue

    print('---')
    print('Processing {0}'.format(file))

    # get language of current file
    current_language = get_current_language(file)
    if current_language == None:
      continue

    # find original file for current language
    attrib_original = get_original(args.wpml_path, current_language)
    if attrib_original == None:
      continue

    # write out new file
    job = attrib_original.split("-")
    job_id = job[0]
    prefix = ''
    if args.output_prefix != '':
      prefix = args.output_prefix + '-'
    output_filename = '{0}translation-job-{1}.xliff'.format(prefix, job_id)
    set_original(file, os.path.join(args.output_path, output_filename), attrib_original)

  # create ZIP in output path
  create_archive(args.output_path, args.output_prefix)

  if errors == 0:
    print('Done.')
  else:
    print('Done with {0} errors, see above.'.format(errors))

if __name__ == '__main__':
  main()