yt_dlp/extractor/peloton.py

   1 import json
   2 import re
   3 import urllib.parse
   4
   5 from .common import InfoExtractor
   6 from ..networking.exceptions import HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     float_or_none,
  10     str_or_none,
  11     traverse_obj,
  12     url_or_none,
  13 )
  14
  15
  16 class PelotonIE(InfoExtractor):
  17     IE_NAME = 'peloton'
  18     _NETRC_MACHINE = 'peloton'
  19     _VALID_URL = r'https?://members\.onepeloton\.com/classes/player/(?P<id>[a-f0-9]+)'
  20     _TESTS = [{
  21         'url': 'https://members.onepeloton.com/classes/player/0e9653eb53544eeb881298c8d7a87b86',
  22         'info_dict': {
  23             'id': '0e9653eb53544eeb881298c8d7a87b86',
  24             'title': '20 min Chest & Back Strength',
  25             'ext': 'mp4',
  26             'thumbnail': r're:^https?://.+\.jpg',
  27             'description': 'md5:fcd5be9b9eda0194b470e13219050a66',
  28             'creator': 'Chase Tucker',
  29             'release_timestamp': 1556141400,
  30             'timestamp': 1556141400,
  31             'upload_date': '20190424',
  32             'duration': 1389,
  33             'categories': ['Strength'],
  34             'tags': ['Workout Mat', 'Light Weights', 'Medium Weights'],
  35             'is_live': False,
  36             'chapters': 'count:1',
  37             'subtitles': {'en': [{
  38                 'url': r're:^https?://.+',
  39                 'ext': 'vtt',
  40             }]},
  41         }, 'params': {
  42             'skip_download': 'm3u8',
  43         },
  44         'skip': 'Account needed',
  45     }, {
  46         'url': 'https://members.onepeloton.com/classes/player/26603d53d6bb4de1b340514864a6a6a8',
  47         'info_dict': {
  48             'id': '26603d53d6bb4de1b340514864a6a6a8',
  49             'title': '30 min Earth Day Run',
  50             'ext': 'm4a',
  51             'thumbnail': r're:https://.+\.jpg',
  52             'description': 'md5:adc065a073934d7ee0475d217afe0c3d',
  53             'creator': 'Selena Samuela',
  54             'release_timestamp': 1587567600,
  55             'timestamp': 1587567600,
  56             'upload_date': '20200422',
  57             'duration': 1802,
  58             'categories': ['Running'],
  59             'is_live': False,
  60             'chapters': 'count:3',
  61         }, 'params': {
  62             'skip_download': 'm3u8',
  63         },
  64         'skip': 'Account needed',
  65     }]
  66
  67     _MANIFEST_URL_TEMPLATE = '%s?hdnea=%s'
  68
  69     def _start_session(self, video_id):
  70         self._download_webpage('https://api.onepeloton.com/api/started_client_session', video_id, note='Starting session')
  71
  72     def _login(self, video_id):
  73         username, password = self._get_login_info()
  74         if not (username and password):
  75             self.raise_login_required()
  76         try:
  77             self._download_json(
  78                 'https://api.onepeloton.com/auth/login', video_id, note='Logging in',
  79                 data=json.dumps({
  80                     'username_or_email': username,
  81                     'password': password,
  82                     'with_pubsub': False,
  83                 }).encode(),
  84                 headers={'Content-Type': 'application/json', 'User-Agent': 'web'})
  85         except ExtractorError as e:
  86             if isinstance(e.cause, HTTPError) and e.cause.status == 401:
  87                 json_string = self._webpage_read_content(e.cause.response, None, video_id)
  88                 res = self._parse_json(json_string, video_id)
  89                 raise ExtractorError(res['message'], expected=res['message'] == 'Login failed')
  90             else:
  91                 raise
  92
  93     def _get_token(self, video_id):
  94         try:
  95             subscription = self._download_json(
  96                 'https://api.onepeloton.com/api/subscription/stream', video_id, note='Downloading token',
  97                 data=json.dumps({}).encode(), headers={'Content-Type': 'application/json'})
  98         except ExtractorError as e:
  99             if isinstance(e.cause, HTTPError) and e.cause.status == 403:
 100                 json_string = self._webpage_read_content(e.cause.response, None, video_id)
 101                 res = self._parse_json(json_string, video_id)
 102                 raise ExtractorError(res['message'], expected=res['message'] == 'Stream limit reached')
 103             else:
 104                 raise
 105         return subscription['token']
 106
 107     def _real_extract(self, url):
 108         video_id = self._match_id(url)
 109         try:
 110             self._start_session(video_id)
 111         except ExtractorError as e:
 112             if isinstance(e.cause, HTTPError) and e.cause.status == 401:
 113                 self._login(video_id)
 114                 self._start_session(video_id)
 115             else:
 116                 raise
 117
 118         metadata = self._download_json(f'https://api.onepeloton.com/api/ride/{video_id}/details?stream_source=multichannel', video_id)
 119         ride_data = metadata.get('ride')
 120         if not ride_data:
 121             raise ExtractorError('Missing stream metadata')
 122         token = self._get_token(video_id)
 123
 124         is_live = False
 125         if ride_data.get('content_format') == 'audio':
 126             url = self._MANIFEST_URL_TEMPLATE % (ride_data.get('vod_stream_url'), urllib.parse.quote(token))
 127             formats = [{
 128                 'url': url,
 129                 'ext': 'm4a',
 130                 'format_id': 'audio',
 131                 'vcodec': 'none',
 132             }]
 133             subtitles = {}
 134         else:
 135             if ride_data.get('vod_stream_url'):
 136                 url = 'https://members.onepeloton.com/.netlify/functions/m3u8-proxy?displayLanguage=en&acceptedSubtitles={}&url={}?hdnea={}'.format(
 137                     ','.join([re.sub('^([a-z]+)-([A-Z]+)$', r'\1', caption) for caption in ride_data['captions']]),
 138                     ride_data['vod_stream_url'],
 139                     urllib.parse.quote(urllib.parse.quote(token)))
 140             elif ride_data.get('live_stream_url'):
 141                 url = self._MANIFEST_URL_TEMPLATE % (ride_data.get('live_stream_url'), urllib.parse.quote(token))
 142                 is_live = True
 143             else:
 144                 raise ExtractorError('Missing video URL')
 145             formats, subtitles = self._extract_m3u8_formats_and_subtitles(url, video_id, 'mp4')
 146
 147         if metadata.get('instructor_cues'):
 148             subtitles['cues'] = [{
 149                 'data': json.dumps(metadata.get('instructor_cues')),
 150                 'ext': 'json',
 151             }]
 152
 153         category = ride_data.get('fitness_discipline_display_name')
 154         chapters = [{
 155             'start_time': segment.get('start_time_offset'),
 156             'end_time': segment.get('start_time_offset') + segment.get('length'),
 157             'title': segment.get('name'),
 158         } for segment in traverse_obj(metadata, ('segments', 'segment_list'))]
 159
 160         return {
 161             'id': video_id,
 162             'title': ride_data.get('title'),
 163             'formats': formats,
 164             'thumbnail': url_or_none(ride_data.get('image_url')),
 165             'description': str_or_none(ride_data.get('description')),
 166             'creator': traverse_obj(ride_data, ('instructor', 'name')),
 167             'release_timestamp': ride_data.get('original_air_time'),
 168             'timestamp': ride_data.get('original_air_time'),
 169             'subtitles': subtitles,
 170             'duration': float_or_none(ride_data.get('length')),
 171             'categories': [category] if category else None,
 172             'tags': traverse_obj(ride_data, ('equipment_tags', ..., 'name')),
 173             'is_live': is_live,
 174             'chapters': chapters,
 175         }
 176
 177
 178 class PelotonLiveIE(InfoExtractor):
 179     IE_NAME = 'peloton:live'
 180     IE_DESC = 'Peloton Live'
 181     _VALID_URL = r'https?://members\.onepeloton\.com/player/live/(?P<id>[a-f0-9]+)'
 182     _TEST = {
 183         'url': 'https://members.onepeloton.com/player/live/eedee2d19f804a9788f53aa8bd38eb1b',
 184         'info_dict': {
 185             'id': '32edc92d28044be5bf6c7b6f1f8d1cbc',
 186             'title': '30 min HIIT Ride: Live from Home',
 187             'ext': 'mp4',
 188             'thumbnail': r're:^https?://.+\.png',
 189             'description': 'md5:f0d7d8ed3f901b7ee3f62c1671c15817',
 190             'creator': 'Alex Toussaint',
 191             'release_timestamp': 1587736620,
 192             'timestamp': 1587736620,
 193             'upload_date': '20200424',
 194             'duration': 2014,
 195             'categories': ['Cycling'],
 196             'is_live': False,
 197             'chapters': 'count:3',
 198         },
 199         'params': {
 200             'skip_download': 'm3u8',
 201         },
 202         'skip': 'Account needed',
 203     }
 204
 205     def _real_extract(self, url):
 206         workout_id = self._match_id(url)
 207         peloton = self._download_json(f'https://api.onepeloton.com/api/peloton/{workout_id}', workout_id)
 208
 209         if peloton.get('ride_id'):
 210             if not peloton.get('is_live') or peloton.get('is_encore') or peloton.get('status') != 'PRE_START':
 211                 return self.url_result('https://members.onepeloton.com/classes/player/{}'.format(peloton['ride_id']))
 212             else:
 213                 raise ExtractorError('Ride has not started', expected=True)
 214         else:
 215             raise ExtractorError('Missing video ID')