# Hotfix potentially wrong ISO codes; we are using ISO-639-3. iso_639_2_to_3_mapping = {'cze': 'ces', 'fre': 'fra', 'ger': 'deu'} for part2_code, part3_code in iso_639_2_to_3_mapping.items(): language_pair = language_pair.replace(part2_code, part3_code) try: _total = _total + 1 _hit_xml = tostring(_child, encoding="utf-8").decode('utf-8') if args.dry_run_enabled: _ = HIT(block_id=block_id, hit_xml=_hit_xml, language_pair=language_pair, mturk_only=args.mturk_only) else: # Use get_or_create() to avoid exact duplicates. We do allow # them for WMT13 to measure intra-annotator agreement... h = HIT(block_id=block_id, hit_xml=_hit_xml, language_pair=language_pair, mturk_only=args.mturk_only) h.save() # pylint: disable-msg=W0703 except Exception, msg: print msg _errors = _errors + 1 print print 'Successfully imported {0} HITs, encountered errors for ' \ '{1} HITs.'.format(_total, _errors) print
try: _total = _total + 1 _hit_xml = tostring(_child, encoding="utf-8").decode('utf-8') if args.dry_run_enabled: _ = HIT(block_id=block_id, hit_xml=_hit_xml, language_pair=language_pair, mturk_only=args.mturk_only) else: # Use get_or_create() to avoid exact duplicates. We do allow # them for WMT13 to measure intra-annotator agreement... h = HIT(block_id=block_id, hit_xml=_hit_xml, language_pair=language_pair, mturk_only=args.mturk_only) h.save() # pylint: disable-msg=W0703 except Exception, msg: print msg _errors = _errors + 1 print print '[{0}]'.format(_hits_file) print 'Successfully imported {0} HITs, encountered errors for ' \ '{1} HITs.'.format(_total, _errors) print