yt_dlp/extractor/mainstreaming.py

   1 import re
   2
   3 from .common import InfoExtractor
   4
   5 from ..utils import (
   6     int_or_none,
   7     js_to_json,
   8     parse_duration,
   9     traverse_obj,
  10     try_get,
  11     urljoin
  12 )
  13
  14
  15 class MainStreamingIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:webtools-?)?(?P<host>[A-Za-z0-9-]*\.msvdn.net)/(?:embed|amp_embed|content)/(?P<id>\w+)'
  17     IE_DESC = 'MainStreaming Player'
  18
  19     _TESTS = [
  20         {
  21             # Live stream offline, has alternative content id
  22             'url': 'https://webtools-e18da6642b684f8aa9ae449862783a56.msvdn.net/embed/53EN6GxbWaJC',
  23             'info_dict': {
  24                 'id': '53EN6GxbWaJC',
  25                 'title': 'Diretta homepage 2021-12-31 12:00',
  26                 'description': '',
  27                 'live_status': 'was_live',
  28                 'ext': 'mp4',
  29                 'thumbnail': r're:https?://[A-Za-z0-9-]*\.msvdn.net/image/\w+/poster',
  30             },
  31             'expected_warnings': [
  32                 'Ignoring alternative content ID: WDAF1KOWUpH3',
  33                 'MainStreaming said: Live event is OFFLINE'
  34             ],
  35             'skip': 'live stream offline'
  36         }, {
  37             # playlist
  38             'url': 'https://webtools-e18da6642b684f8aa9ae449862783a56.msvdn.net/embed/WDAF1KOWUpH3',
  39             'info_dict': {
  40                 'id': 'WDAF1KOWUpH3',
  41                 'title': 'Playlist homepage',
  42             },
  43             'playlist_mincount': 2
  44         }, {
  45             # livestream
  46             'url': 'https://webtools-859c1818ed614cc5b0047439470927b0.msvdn.net/embed/tDoFkZD3T1Lw',
  47             'info_dict': {
  48                 'id': 'tDoFkZD3T1Lw',
  49                 'title': r're:Class CNBC Live \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
  50                 'live_status': 'is_live',
  51                 'ext': 'mp4',
  52                 'thumbnail': r're:https?://[A-Za-z0-9-]*\.msvdn.net/image/\w+/poster',
  53             },
  54             'skip': 'live stream'
  55         }, {
  56             'url': 'https://webtools-f5842579ff984c1c98d63b8d789673eb.msvdn.net/embed/EUlZfGWkGpOd?autoPlay=false',
  57             'info_dict': {
  58                 'id': 'EUlZfGWkGpOd',
  59                 'title': 'La Settimana ',
  60                 'description': '03 Ottobre ore 02:00',
  61                 'ext': 'mp4',
  62                 'live_status': 'not_live',
  63                 'thumbnail': r're:https?://[A-Za-z0-9-]*\.msvdn.net/image/\w+/poster',
  64                 'duration': 1512
  65             }
  66         }, {
  67             # video without webtools- prefix
  68             'url': 'https://f5842579ff984c1c98d63b8d789673eb.msvdn.net/embed/MfuWmzL2lGkA?autoplay=false&T=1635860445',
  69             'info_dict': {
  70                 'id': 'MfuWmzL2lGkA',
  71                 'title': 'TG Mattina',
  72                 'description': '06 Ottobre ore 08:00',
  73                 'ext': 'mp4',
  74                 'live_status': 'not_live',
  75                 'thumbnail': r're:https?://[A-Za-z0-9-]*\.msvdn.net/image/\w+/poster',
  76                 'duration': 789.04
  77             }
  78         }, {
  79             # always-on livestream with DVR
  80             'url': 'https://webtools-f5842579ff984c1c98d63b8d789673eb.msvdn.net/embed/HVvPMzy',
  81             'info_dict': {
  82                 'id': 'HVvPMzy',
  83                 'title': r're:^Diretta LaC News24 \d{4}-\d{2}-\d{2} \d{2}:\d{2}$',
  84                 'description': 'canale all news',
  85                 'live_status': 'is_live',
  86                 'ext': 'mp4',
  87                 'thumbnail': r're:https?://[A-Za-z0-9-]*\.msvdn.net/image/\w+/poster',
  88             },
  89             'params': {
  90                 'skip_download': True,
  91             },
  92         }, {
  93             # no host
  94             'url': 'https://webtools.msvdn.net/embed/MfuWmzL2lGkA',
  95             'only_matching': True
  96         }, {
  97             'url': 'https://859c1818ed614cc5b0047439470927b0.msvdn.net/amp_embed/tDoFkZD3T1Lw',
  98             'only_matching': True
  99         }, {
 100             'url': 'https://859c1818ed614cc5b0047439470927b0.msvdn.net/content/tDoFkZD3T1Lw#',
 101             'only_matching': True
 102         }
 103     ]
 104
 105     @staticmethod
 106     def _extract_urls(webpage):
 107         mobj = re.findall(
 108             r'<iframe[^>]+?src=["\']?(?P<url>%s)["\']?' % MainStreamingIE._VALID_URL, webpage)
 109         if mobj:
 110             return [group[0] for group in mobj]
 111
 112     def _playlist_entries(self, host, playlist_content):
 113         for entry in playlist_content:
 114             content_id = entry.get('contentID')
 115             yield {
 116                 '_type': 'url',
 117                 'ie_key': MainStreamingIE.ie_key(),
 118                 'id': content_id,
 119                 'duration': int_or_none(traverse_obj(entry, ('duration', 'totalSeconds'))),
 120                 'title': entry.get('title'),
 121                 'url': f'https://{host}/embed/{content_id}'
 122             }
 123
 124     @staticmethod
 125     def _get_webtools_host(host):
 126         if not host.startswith('webtools'):
 127             host = 'webtools' + ('-' if not host.startswith('.') else '') + host
 128         return host
 129
 130     def _get_webtools_base_url(self, host):
 131         return f'{self.http_scheme()}//{self._get_webtools_host(host)}'
 132
 133     def _call_api(self, host: str, path: str, item_id: str, query=None, note='Downloading API JSON', fatal=False):
 134         # JSON API, does not appear to be documented
 135         return self._call_webtools_api(host, '/api/v2/' + path, item_id, query, note, fatal)
 136
 137     def _call_webtools_api(self, host: str, path: str, item_id: str, query=None, note='Downloading webtools API JSON', fatal=False):
 138         # webtools docs: https://webtools.msvdn.net/
 139         return self._download_json(
 140             urljoin(self._get_webtools_base_url(host), path), item_id, query=query, note=note, fatal=fatal)
 141
 142     def _real_extract(self, url):
 143         host, video_id = self._match_valid_url(url).groups()
 144         content_info = try_get(
 145             self._call_api(
 146                 host, f'content/{video_id}', video_id, note='Downloading content info API JSON'), lambda x: x['playerContentInfo'])
 147         # Fallback
 148         if not content_info:
 149             webpage = self._download_webpage(url, video_id)
 150             player_config = self._parse_json(
 151                 self._search_regex(
 152                     r'config\s*=\s*({.+?})\s*;', webpage, 'mainstreaming player config',
 153                     default='{}', flags=re.DOTALL),
 154                 video_id, transform_source=js_to_json, fatal=False) or {}
 155             content_info = player_config['contentInfo']
 156
 157         host = content_info.get('host') or host
 158         video_id = content_info.get('contentID') or video_id
 159         title = content_info.get('title')
 160         description = traverse_obj(content_info, 'longDescription', 'shortDescription', expected_type=str)
 161         live_status = 'not_live'
 162         if content_info.get('drmEnabled'):
 163             self.report_drm(video_id)
 164
 165         alternative_content_id = content_info.get('alternativeContentID')
 166         if alternative_content_id:
 167             self.report_warning(f'Ignoring alternative content ID: {alternative_content_id}')
 168
 169         content_type = int_or_none(content_info.get('contentType'))
 170         format_base_url = None
 171         formats = []
 172         subtitles = {}
 173         # Live content
 174         if content_type == 20:
 175             dvr_enabled = traverse_obj(content_info, ('playerSettings', 'dvrEnabled'), expected_type=bool)
 176             format_base_url = f"https://{host}/live/{content_info['liveSourceID']}/{video_id}/%s{'?DVR' if dvr_enabled else ''}"
 177             live_status = 'is_live'
 178             heartbeat = self._call_api(host, f'heartbeat/{video_id}', video_id, note='Checking stream status') or {}
 179             if heartbeat.get('heartBeatUp') is False:
 180                 self.raise_no_formats(f'MainStreaming said: {heartbeat.get("responseMessage")}', expected=True)
 181                 live_status = 'was_live'
 182
 183         # Playlist
 184         elif content_type == 31:
 185             return self.playlist_result(
 186                 self._playlist_entries(host, content_info.get('playlistContents')), video_id, title, description)
 187         # Normal video content?
 188         elif content_type == 10:
 189             format_base_url = f'https://{host}/vod/{video_id}/%s'
 190             # Progressive format
 191             # Note: in https://webtools.msvdn.net/loader/playerV2.js there is mention of original.mp3 format,
 192             # however it seems to be the same as original.mp4?
 193             formats.append({'url': format_base_url % 'original.mp4', 'format_note': 'original', 'quality': 1})
 194         else:
 195             self.raise_no_formats(f'Unknown content type {content_type}')
 196
 197         if format_base_url:
 198             m3u8_formats, m3u8_subs = self._extract_m3u8_formats_and_subtitles(
 199                 format_base_url % 'playlist.m3u8', video_id=video_id, fatal=False)
 200             mpd_formats, mpd_subs = self._extract_mpd_formats_and_subtitles(
 201                 format_base_url % 'manifest.mpd', video_id=video_id, fatal=False)
 202
 203             subtitles = self._merge_subtitles(m3u8_subs, mpd_subs)
 204             formats.extend(m3u8_formats + mpd_formats)
 205
 206         self._sort_formats(formats)
 207
 208         return {
 209             'id': video_id,
 210             'title': title,
 211             'description': description,
 212             'formats': formats,
 213             'live_status': live_status,
 214             'duration': parse_duration(content_info.get('duration')),
 215             'tags': content_info.get('tags'),
 216             'subtitles': subtitles,
 217             'thumbnail': urljoin(self._get_webtools_base_url(host), f'image/{video_id}/poster')
 218         }