yt_dlp/extractor/europa.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     orderedSet,
   5     parse_duration,
   6     parse_qs,
   7     qualities,
   8     unified_strdate,
   9     xpath_text
  10 )
  11
  12
  13 class EuropaIE(InfoExtractor):
  14     _VALID_URL = r'https?://ec\.europa\.eu/avservices/(?:video/player|audio/audioDetails)\.cfm\?.*?\bref=(?P<id>[A-Za-z0-9-]+)'
  15     _TESTS = [{
  16         'url': 'http://ec.europa.eu/avservices/video/player.cfm?ref=I107758',
  17         'md5': '574f080699ddd1e19a675b0ddf010371',
  18         'info_dict': {
  19             'id': 'I107758',
  20             'ext': 'mp4',
  21             'title': 'TRADE - Wikileaks on TTIP',
  22             'description': 'NEW  LIVE EC Midday press briefing of 11/08/2015',
  23             'thumbnail': r're:^https?://.*\.jpg$',
  24             'upload_date': '20150811',
  25             'duration': 34,
  26             'view_count': int,
  27             'formats': 'mincount:3',
  28         }
  29     }, {
  30         'url': 'http://ec.europa.eu/avservices/video/player.cfm?sitelang=en&ref=I107786',
  31         'only_matching': True,
  32     }, {
  33         'url': 'http://ec.europa.eu/avservices/audio/audioDetails.cfm?ref=I-109295&sitelang=en',
  34         'only_matching': True,
  35     }]
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         playlist = self._download_xml(
  41             'http://ec.europa.eu/avservices/video/player/playlist.cfm?ID=%s' % video_id, video_id)
  42
  43         def get_item(type_, preference):
  44             items = {}
  45             for item in playlist.findall('./info/%s/item' % type_):
  46                 lang, label = xpath_text(item, 'lg', default=None), xpath_text(item, 'label', default=None)
  47                 if lang and label:
  48                     items[lang] = label.strip()
  49             for p in preference:
  50                 if items.get(p):
  51                     return items[p]
  52
  53         query = parse_qs(url)
  54         preferred_lang = query.get('sitelang', ('en', ))[0]
  55
  56         preferred_langs = orderedSet((preferred_lang, 'en', 'int'))
  57
  58         title = get_item('title', preferred_langs) or video_id
  59         description = get_item('description', preferred_langs)
  60         thumbnail = xpath_text(playlist, './info/thumburl', 'thumbnail')
  61         upload_date = unified_strdate(xpath_text(playlist, './info/date', 'upload date'))
  62         duration = parse_duration(xpath_text(playlist, './info/duration', 'duration'))
  63         view_count = int_or_none(xpath_text(playlist, './info/views', 'views'))
  64
  65         language_preference = qualities(preferred_langs[::-1])
  66
  67         formats = []
  68         for file_ in playlist.findall('./files/file'):
  69             video_url = xpath_text(file_, './url')
  70             if not video_url:
  71                 continue
  72             lang = xpath_text(file_, './lg')
  73             formats.append({
  74                 'url': video_url,
  75                 'format_id': lang,
  76                 'format_note': xpath_text(file_, './lglabel'),
  77                 'language_preference': language_preference(lang)
  78             })
  79
  80         return {
  81             'id': video_id,
  82             'title': title,
  83             'description': description,
  84             'thumbnail': thumbnail,
  85             'upload_date': upload_date,
  86             'duration': duration,
  87             'view_count': view_count,
  88             'formats': formats
  89         }