youtube_dl/extractor/europa.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_urlparse
   6 from ..utils import (
   7     int_or_none,
   8     orderedSet,
   9     parse_duration,
  10     qualities,
  11     unified_strdate,
  12     xpath_text
  13 )
  14
  15
  16 class EuropaIE(InfoExtractor):
  17     _VALID_URL = r'https?://ec\.europa\.eu/avservices/(?:video/player|audio/audioDetails)\.cfm\?.*?\bref=(?P<id>[A-Za-z0-9-]+)'
  18     _TESTS = [{
  19         'url': 'http://ec.europa.eu/avservices/video/player.cfm?ref=I107758',
  20         'md5': '574f080699ddd1e19a675b0ddf010371',
  21         'info_dict': {
  22             'id': 'I107758',
  23             'ext': 'mp4',
  24             'title': 'TRADE - Wikileaks on TTIP',
  25             'description': 'NEW  LIVE EC Midday press briefing of 11/08/2015',
  26             'thumbnail': 're:^https?://.*\.jpg$',
  27             'upload_date': '20150811',
  28             'duration': 34,
  29             'view_count': int,
  30             'formats': 'mincount:3',
  31         }
  32     }, {
  33         'url': 'http://ec.europa.eu/avservices/video/player.cfm?sitelang=en&ref=I107786',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://ec.europa.eu/avservices/audio/audioDetails.cfm?ref=I-109295&sitelang=en',
  37         'only_matching': True,
  38     }]
  39
  40     def _real_extract(self, url):
  41         video_id = self._match_id(url)
  42
  43         playlist = self._download_xml(
  44             'http://ec.europa.eu/avservices/video/player/playlist.cfm?ID=%s' % video_id, video_id)
  45
  46         def get_item(type_, preference):
  47             items = {}
  48             for item in playlist.findall('./info/%s/item' % type_):
  49                 lang, label = xpath_text(item, 'lg', default=None), xpath_text(item, 'label', default=None)
  50                 if lang and label:
  51                     items[lang] = label.strip()
  52             for p in preference:
  53                 if items.get(p):
  54                     return items[p]
  55
  56         query = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
  57         preferred_lang = query.get('sitelang', ('en', ))[0]
  58
  59         preferred_langs = orderedSet((preferred_lang, 'en', 'int'))
  60
  61         title = get_item('title', preferred_langs) or video_id
  62         description = get_item('description', preferred_langs)
  63         thumbnmail = xpath_text(playlist, './info/thumburl', 'thumbnail')
  64         upload_date = unified_strdate(xpath_text(playlist, './info/date', 'upload date'))
  65         duration = parse_duration(xpath_text(playlist, './info/duration', 'duration'))
  66         view_count = int_or_none(xpath_text(playlist, './info/views', 'views'))
  67
  68         language_preference = qualities(preferred_langs[::-1])
  69
  70         formats = []
  71         for file_ in playlist.findall('./files/file'):
  72             video_url = xpath_text(file_, './url')
  73             if not video_url:
  74                 continue
  75             lang = xpath_text(file_, './lg')
  76             formats.append({
  77                 'url': video_url,
  78                 'format_id': lang,
  79                 'format_note': xpath_text(file_, './lglabel'),
  80                 'language_preference': language_preference(lang)
  81             })
  82         self._sort_formats(formats)
  83
  84         return {
  85             'id': video_id,
  86             'title': title,
  87             'description': description,
  88             'thumbnail': thumbnmail,
  89             'upload_date': upload_date,
  90             'duration': duration,
  91             'view_count': view_count,
  92             'formats': formats
  93         }