yt_dlp/extractor/tv5mondeplus.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     clean_html,
   6     determine_ext,
   7     extract_attributes,
   8     get_element_by_class,
   9     get_element_html_by_class,
  10     int_or_none,
  11     url_or_none,
  12 )
  13 from ..utils.traversal import traverse_obj
  14
  15
  16 class TV5MondePlusIE(InfoExtractor):
  17     IE_NAME = 'TV5MONDE'
  18     _VALID_URL = r'https?://(?:www\.)?tv5monde\.com/tv/video/(?P<id>[^/?#]+)'
  19     _TESTS = [{
  20         # documentary
  21         'url': 'https://www.tv5monde.com/tv/video/65931-baudouin-l-heritage-d-un-roi-baudouin-l-heritage-d-un-roi',
  22         'md5': 'd2a708902d3df230a357c99701aece05',
  23         'info_dict': {
  24             'id': '3FPa7JMu21_6D4BA7b',
  25             'display_id': '65931-baudouin-l-heritage-d-un-roi-baudouin-l-heritage-d-un-roi',
  26             'ext': 'mp4',
  27             'title': "Baudouin, l'héritage d'un roi",
  28             'thumbnail': 'https://psi.tv5monde.com/upsilon-images/960x540/6f/baudouin-f49c6b0e.jpg',
  29             'duration': 4842,
  30             'upload_date': '20240130',
  31             'timestamp': 1706641242,
  32             'episode': "BAUDOUIN, L'HERITAGE D'UN ROI",
  33             'description': 'md5:78125c74a5cac06d7743a2d09126edad',
  34             'series': "Baudouin, l'héritage d'un roi",
  35         },
  36     }, {
  37         # series episode
  38         'url': 'https://www.tv5monde.com/tv/video/52952-toute-la-vie-mardi-23-mars-2021',
  39         'md5': 'f5e09637cadd55639c05874e22eb56bf',
  40         'info_dict': {
  41             'id': 'obRRZ8m6g9_6D4BA7b',
  42             'display_id': '52952-toute-la-vie-mardi-23-mars-2021',
  43             'ext': 'mp4',
  44             'title': 'Toute la vie',
  45             'description': 'md5:a824a2e1dfd94cf45fa379a1fb43ce65',
  46             'thumbnail': 'https://psi.tv5monde.com/media/image/960px/5880553.jpg',
  47             'duration': 2526,
  48             'upload_date': '20230721',
  49             'timestamp': 1689971646,
  50             'series': 'Toute la vie',
  51             'episode': 'Mardi 23 mars 2021',
  52         },
  53     }, {
  54         # movie
  55         'url': 'https://www.tv5monde.com/tv/video/8771-ce-fleuve-qui-nous-charrie-ce-fleuve-qui-nous-charrie-p001-ce-fleuve-qui-nous-charrie',
  56         'md5': '87cefc34e10a6bf4f7823cccd7b36eb2',
  57         'info_dict': {
  58             'id': 'DOcfvdLKXL_6D4BA7b',
  59             'display_id': '8771-ce-fleuve-qui-nous-charrie-ce-fleuve-qui-nous-charrie-p001-ce-fleuve-qui-nous-charrie',
  60             'ext': 'mp4',
  61             'title': 'Ce fleuve qui nous charrie',
  62             'description': 'md5:62ba3f875343c7fc4082bdfbbc1be992',
  63             'thumbnail': 'https://psi.tv5monde.com/media/image/960px/5476617.jpg',
  64             'duration': 5300,
  65             'upload_date': '20210822',
  66             'timestamp': 1629594105,
  67             'episode': 'CE FLEUVE QUI NOUS CHARRIE-P001-CE FLEUVE QUI NOUS CHARRIE',
  68             'series': 'Ce fleuve qui nous charrie',
  69         },
  70     }, {
  71         # news
  72         'url': 'https://www.tv5monde.com/tv/video/70402-tv5monde-le-journal-edition-du-08-05-24-11h',
  73         'md5': 'c62977d6d10754a2ecebba70ad370479',
  74         'info_dict': {
  75             'id': 'LgQFrOCNsc_6D4BA7b',
  76             'display_id': '70402-tv5monde-le-journal-edition-du-08-05-24-11h',
  77             'ext': 'mp4',
  78             'title': 'TV5MONDE, le journal',
  79             'description': 'md5:777dc209eaa4423b678477c36b0b04a8',
  80             'thumbnail': 'https://psi.tv5monde.com/media/image/960px/6184105.jpg',
  81             'duration': 854,
  82             'upload_date': '20240508',
  83             'timestamp': 1715159640,
  84             'series': 'TV5MONDE, le journal',
  85             'episode': 'EDITION DU 08/05/24 - 11H',
  86         },
  87     }]
  88     _GEO_BYPASS = False
  89
  90     @staticmethod
  91     def _extract_subtitles(data_captions):
  92         subtitles = {}
  93         for f in traverse_obj(data_captions, ('files', lambda _, v: url_or_none(v['file']))):
  94             subtitles.setdefault(f.get('label') or 'fra', []).append({'url': f['file']})
  95         return subtitles
  96
  97     def _real_extract(self, url):
  98         display_id = self._match_id(url)
  99         webpage = self._download_webpage(url, display_id, impersonate=True)
 100
 101         if ">Ce programme n'est malheureusement pas disponible pour votre zone géographique.<" in webpage:
 102             self.raise_geo_restricted(countries=['FR'])
 103
 104         vpl_data = extract_attributes(self._search_regex(
 105             r'(<[^>]+class="video_player_loader"[^>]+>)',
 106             webpage, 'video player loader'))
 107
 108         video_files = self._parse_json(
 109             vpl_data['data-broadcast'], display_id)
 110         formats = []
 111         video_id = None
 112
 113         def process_video_files(v):
 114             nonlocal video_id
 115             for video_file in v:
 116                 v_url = video_file.get('url')
 117                 if not v_url:
 118                     continue
 119                 if video_file.get('type') == 'application/deferred':
 120                     d_param = urllib.parse.quote(v_url)
 121                     token = video_file.get('token')
 122                     if not token:
 123                         continue
 124                     deferred_json = self._download_json(
 125                         f'https://api.tv5monde.com/player/asset/{d_param}/resolve?condenseKS=true',
 126                         display_id, 'Downloading deferred info', fatal=False, impersonate=True,
 127                         headers={'Authorization': f'Bearer {token}'})
 128                     v_url = traverse_obj(deferred_json, (0, 'url', {url_or_none}))
 129                     if not v_url:
 130                         continue
 131                     # data-guid from the webpage isn't stable, use the material id from the json urls
 132                     video_id = self._search_regex(
 133                         r'materials/([\da-zA-Z]{10}_[\da-fA-F]{7})/', v_url, 'video id', default=None)
 134                     process_video_files(deferred_json)
 135
 136                 video_format = video_file.get('format') or determine_ext(v_url)
 137                 if video_format == 'm3u8':
 138                     formats.extend(self._extract_m3u8_formats(
 139                         v_url, display_id, 'mp4', 'm3u8_native',
 140                         m3u8_id='hls', fatal=False))
 141                 elif video_format == 'mpd':
 142                     formats.extend(self._extract_mpd_formats(
 143                         v_url, display_id, fatal=False))
 144                 else:
 145                     formats.append({
 146                         'url': v_url,
 147                         'format_id': video_format,
 148                     })
 149
 150         process_video_files(video_files)
 151
 152         metadata = self._parse_json(
 153             vpl_data.get('data-metadata') or '{}', display_id, fatal=False)
 154
 155         if not video_id:
 156             video_id = self._search_regex(
 157                 (r'data-guid=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
 158                  r'id_contenu["\']\s:\s*(\d+)'), webpage, 'video id',
 159                 default=display_id)
 160
 161         return {
 162             **traverse_obj(metadata, ('content', {
 163                 'id': ('id', {str}),
 164                 'title': ('title', {str}),
 165                 'episode': ('title', {str}),
 166                 'series': ('series', {str}),
 167                 'timestamp': ('publishDate_ts', {int_or_none}),
 168                 'duration': ('duration', {int_or_none}),
 169             })),
 170             'id': video_id,
 171             'display_id': display_id,
 172             'title': clean_html(get_element_by_class('main-title', webpage)),
 173             'description': clean_html(get_element_by_class('text', get_element_html_by_class('ep-summary', webpage) or '')),
 174             'thumbnail': url_or_none(vpl_data.get('data-image')),
 175             'formats': formats,
 176             'subtitles': self._extract_subtitles(self._parse_json(
 177                 traverse_obj(vpl_data, ('data-captions', {str}), default='{}'), display_id, fatal=False)),
 178         }