yt_dlp/extractor/lbry.py

   1 import functools
   2 import json
   3 import re
   4 import urllib.parse
   5
   6 from .common import InfoExtractor
   7 from ..networking import HEADRequest
   8 from ..utils import (
   9     ExtractorError,
  10     OnDemandPagedList,
  11     UnsupportedError,
  12     determine_ext,
  13     int_or_none,
  14     mimetype2ext,
  15     parse_qs,
  16     traverse_obj,
  17     try_get,
  18     url_or_none,
  19     urlhandle_detect_ext,
  20     urljoin,
  21 )
  22
  23
  24 class LBRYBaseIE(InfoExtractor):
  25     _BASE_URL_REGEX = r'(?x)(?:https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/|lbry://)'
  26     _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
  27     _OPT_CLAIM_ID = f'[^$@:/?#&]+(?:[:#]{_CLAIM_ID_REGEX})?'
  28     _SUPPORTED_STREAM_TYPES = ['video', 'audio']
  29     _PAGE_SIZE = 50
  30
  31     def _call_api_proxy(self, method, display_id, params, resource):
  32         headers = {'Content-Type': 'application/json-rpc'}
  33         token = try_get(self._get_cookies('https://odysee.com'), lambda x: x['auth_token'].value)
  34         if token:
  35             headers['x-lbry-auth-token'] = token
  36         response = self._download_json(
  37             'https://api.lbry.tv/api/v1/proxy',
  38             display_id, f'Downloading {resource} JSON metadata',
  39             headers=headers,
  40             data=json.dumps({
  41                 'method': method,
  42                 'params': params,
  43             }).encode())
  44         err = response.get('error')
  45         if err:
  46             raise ExtractorError(
  47                 f'{self.IE_NAME} said: {err.get("code")} - {err.get("message")}', expected=True)
  48         return response['result']
  49
  50     def _resolve_url(self, url, display_id, resource):
  51         return self._call_api_proxy(
  52             'resolve', display_id, {'urls': url}, resource)[url]
  53
  54     def _permanent_url(self, url, claim_name, claim_id):
  55         return urljoin(
  56             url.replace('lbry://', 'https://lbry.tv/'),
  57             f'/{claim_name}:{claim_id}')
  58
  59     def _parse_stream(self, stream, url):
  60         stream_type = traverse_obj(stream, ('value', 'stream_type', {str}))
  61
  62         info = traverse_obj(stream, {
  63             'title': ('value', 'title', {str}),
  64             'thumbnail': ('value', 'thumbnail', 'url', {url_or_none}),
  65             'description': ('value', 'description', {str}),
  66             'license': ('value', 'license', {str}),
  67             'timestamp': ('timestamp', {int_or_none}),
  68             'release_timestamp': ('value', 'release_time', {int_or_none}),
  69             'tags': ('value', 'tags', ..., filter),
  70             'duration': ('value', stream_type, 'duration', {int_or_none}),
  71             'channel': ('signing_channel', 'value', 'title', {str}),
  72             'channel_id': ('signing_channel', 'claim_id', {str}),
  73             'uploader_id': ('signing_channel', 'name', {str}),
  74         })
  75
  76         if info.get('uploader_id') and info.get('channel_id'):
  77             info['channel_url'] = self._permanent_url(url, info['uploader_id'], info['channel_id'])
  78
  79         return info
  80
  81     def _fetch_page(self, display_id, url, params, page):
  82         page += 1
  83         page_params = {
  84             'no_totals': True,
  85             'page': page,
  86             'page_size': self._PAGE_SIZE,
  87             **params,
  88         }
  89         result = self._call_api_proxy(
  90             'claim_search', display_id, page_params, f'page {page}')
  91         for item in traverse_obj(result, ('items', lambda _, v: v['name'] and v['claim_id'])):
  92             yield {
  93                 **self._parse_stream(item, url),
  94                 '_type': 'url',
  95                 'id': item['claim_id'],
  96                 'url': self._permanent_url(url, item['name'], item['claim_id']),
  97             }
  98
  99     def _playlist_entries(self, url, display_id, claim_param, metadata):
 100         qs = parse_qs(url)
 101         content = qs.get('content', [None])[0]
 102         params = {
 103             'fee_amount': qs.get('fee_amount', ['>=0'])[0],
 104             'order_by': {
 105                 'new': ['release_time'],
 106                 'top': ['effective_amount'],
 107                 'trending': ['trending_group', 'trending_mixed'],
 108             }[qs.get('order', ['new'])[0]],
 109             'claim_type': 'stream',
 110             'stream_types': [content] if content in ['audio', 'video'] else self._SUPPORTED_STREAM_TYPES,
 111             **claim_param,
 112         }
 113         duration = qs.get('duration', [None])[0]
 114         if duration:
 115             params['duration'] = {
 116                 'long': '>=1200',
 117                 'short': '<=240',
 118             }[duration]
 119         language = qs.get('language', ['all'])[0]
 120         if language != 'all':
 121             languages = [language]
 122             if language == 'en':
 123                 languages.append('none')
 124             params['any_languages'] = languages
 125
 126         entries = OnDemandPagedList(
 127             functools.partial(self._fetch_page, display_id, url, params),
 128             self._PAGE_SIZE)
 129
 130         return self.playlist_result(
 131             entries, display_id, **traverse_obj(metadata, ('value', {
 132                 'title': 'title',
 133                 'description': 'description',
 134             })))
 135
 136
 137 class LBRYIE(LBRYBaseIE):
 138     IE_NAME = 'lbry'
 139     IE_DESC = 'odysee.com'
 140     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + rf'''
 141         (?:\$/(?:download|embed)/)?
 142         (?P<id>
 143             [^$@:/?#]+/{LBRYBaseIE._CLAIM_ID_REGEX}
 144             |(?:@{LBRYBaseIE._OPT_CLAIM_ID}/)?{LBRYBaseIE._OPT_CLAIM_ID}
 145         )'''
 146     _TESTS = [{
 147         # Video
 148         'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
 149         'md5': '65bd7ec1f6744ada55da8e4c48a2edf9',
 150         'info_dict': {
 151             'id': '17f983b61f53091fb8ea58a9c56804e4ff8cff4d',
 152             'ext': 'mp4',
 153             'title': 'First day in LBRY? Start HERE!',
 154             'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
 155             'timestamp': 1595694354,
 156             'upload_date': '20200725',
 157             'release_timestamp': 1595340697,
 158             'release_date': '20200721',
 159             'width': 1280,
 160             'height': 720,
 161             'thumbnail': 'https://spee.ch/7/67f2d809c263288c.png',
 162             'license': 'None',
 163             'uploader_id': '@Mantega',
 164             'duration': 346,
 165             'channel': 'LBRY/Odysee rats united!!!',
 166             'channel_id': '1c8ad6a2ab4e889a71146ae4deeb23bb92dab627',
 167             'channel_url': 'https://lbry.tv/@Mantega:1c8ad6a2ab4e889a71146ae4deeb23bb92dab627',
 168             'tags': [
 169                 'first day in lbry',
 170                 'lbc',
 171                 'lbry',
 172                 'start',
 173                 'tutorial',
 174             ],
 175         },
 176     }, {
 177         # Audio
 178         'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
 179         'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
 180         'info_dict': {
 181             'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 182             'ext': 'mp3',
 183             'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
 184             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
 185             'timestamp': 1591312601,
 186             'upload_date': '20200604',
 187             'release_timestamp': 1591312421,
 188             'release_date': '20200604',
 189             'tags': list,
 190             'duration': 2570,
 191             'channel': 'The LBRY Foundation',
 192             'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 193             'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
 194             'vcodec': 'none',
 195             'thumbnail': 'https://spee.ch/d/0bc63b0e6bf1492d.png',
 196             'license': 'None',
 197             'uploader_id': '@LBRYFoundation',
 198         },
 199     }, {
 200         'url': 'https://odysee.com/@gardeningincanada:b/plants-i-will-never-grow-again.-the:e',
 201         'md5': 'c35fac796f62a14274b4dc2addb5d0ba',
 202         'info_dict': {
 203             'id': 'e51671357333fe22ae88aad320bde2f6f96b1410',
 204             'ext': 'mp4',
 205             'title': 'PLANTS I WILL NEVER GROW AGAIN. THE BLACK LIST PLANTS FOR A CANADIAN GARDEN | Gardening in Canada 🍁',
 206             'description': 'md5:9c539c6a03fb843956de61a4d5288d5e',
 207             'timestamp': 1618254123,
 208             'upload_date': '20210412',
 209             'release_timestamp': 1618254002,
 210             'release_date': '20210412',
 211             'tags': list,
 212             'duration': 554,
 213             'channel': 'Gardening In Canada',
 214             'channel_id': 'b8be0e93b423dad221abe29545fbe8ec36e806bc',
 215             'channel_url': 'https://odysee.com/@gardeningincanada:b8be0e93b423dad221abe29545fbe8ec36e806bc',
 216             'uploader_id': '@gardeningincanada',
 217             'formats': 'mincount:3',
 218             'thumbnail': 'https://thumbnails.lbry.com/AgHSc_HzrrE',
 219             'license': 'Copyrighted (contact publisher)',
 220         },
 221     }, {
 222         # HLS live stream (might expire)
 223         'url': 'https://odysee.com/@RT:fd/livestream_RT:d',
 224         'info_dict': {
 225             'id': 'fdd11cb3ab75f95efb7b3bc2d726aa13ac915b66',
 226             'ext': 'mp4',
 227             'live_status': 'is_live',
 228             'title': 'startswith:RT News | Livestream 24/7',
 229             'description': 'md5:fe68d0056dfe79c1a6b8ce8c34d5f6fa',
 230             'timestamp': int,
 231             'upload_date': str,
 232             'release_timestamp': int,
 233             'release_date': str,
 234             'tags': list,
 235             'channel': 'RT',
 236             'channel_id': 'fdd11cb3ab75f95efb7b3bc2d726aa13ac915b66',
 237             'channel_url': 'https://odysee.com/@RT:fdd11cb3ab75f95efb7b3bc2d726aa13ac915b66',
 238             'formats': 'mincount:1',
 239             'thumbnail': 'startswith:https://thumb',
 240             'license': 'None',
 241             'uploader_id': '@RT',
 242         },
 243         'params': {'skip_download': True},
 244     }, {
 245         # original quality format w/higher resolution than HLS formats
 246         'url': 'https://odysee.com/@wickedtruths:2/Biotechnological-Invasion-of-Skin-(April-2023):4',
 247         'md5': '305b0b3b369bde1b984961f005b67193',
 248         'info_dict': {
 249             'id': '41fbfe805eb73c8d3012c0c49faa0f563274f634',
 250             'ext': 'mp4',
 251             'title': 'Biotechnological Invasion of Skin (April 2023)',
 252             'description': 'md5:fe28689db2cb7ba3436d819ac3ffc378',
 253             'channel': 'Wicked Truths',
 254             'channel_id': '23d2bbf856b0ceed5b1d7c5960bcc72da5a20cb0',
 255             'channel_url': 'https://odysee.com/@wickedtruths:23d2bbf856b0ceed5b1d7c5960bcc72da5a20cb0',
 256             'uploader_id': '@wickedtruths',
 257             'timestamp': 1695114347,
 258             'upload_date': '20230919',
 259             'release_timestamp': 1685617473,
 260             'release_date': '20230601',
 261             'duration': 1063,
 262             'thumbnail': 'https://thumbs.odycdn.com/4e6d39da4df0cfdad45f64e253a15959.webp',
 263             'tags': ['smart skin surveillance', 'biotechnology invasion of skin', 'morgellons'],
 264             'license': 'None',
 265             'protocol': 'https',  # test for direct mp4 download
 266         },
 267     }, {
 268         'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
 269         'only_matching': True,
 270     }, {
 271         'url': 'https://odysee.com/@ScammerRevolts:b0/I-SYSKEY\'D-THE-SAME-SCAMMERS-3-TIMES!:b',
 272         'only_matching': True,
 273     }, {
 274         'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 275         'only_matching': True,
 276     }, {
 277         'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 278         'only_matching': True,
 279     }, {
 280         'url': 'https://lbry.tv/Episode-1:e7',
 281         'only_matching': True,
 282     }, {
 283         'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
 284         'only_matching': True,
 285     }, {
 286         'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 287         'only_matching': True,
 288     }, {
 289         'url': 'https://lbry.tv/@lacajadepandora:a/TRUMP-EST%C3%81-BIEN-PUESTO-con-Pilar-Baselga,-Carlos-Senra,-Luis-Palacios-(720p_30fps_H264-192kbit_AAC):1',
 290         'only_matching': True,
 291     }, {
 292         'url': 'lbry://@lbry#3f/odysee#7',
 293         'only_matching': True,
 294     }]
 295
 296     def _real_extract(self, url):
 297         display_id = self._match_id(url)
 298         if display_id.startswith('@'):
 299             display_id = display_id.replace(':', '#')
 300         else:
 301             display_id = display_id.replace('/', ':')
 302         display_id = urllib.parse.unquote(display_id)
 303         uri = 'lbry://' + display_id
 304         result = self._resolve_url(uri, display_id, 'stream')
 305         headers = {'Referer': 'https://odysee.com/'}
 306
 307         formats = []
 308         stream_type = traverse_obj(result, ('value', 'stream_type', {str}))
 309
 310         if stream_type in self._SUPPORTED_STREAM_TYPES:
 311             claim_id, is_live = result['claim_id'], False
 312             streaming_url = self._call_api_proxy(
 313                 'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
 314
 315             # GET request to v3 API returns original video/audio file if available
 316             direct_url = re.sub(r'/api/v\d+/', '/api/v3/', streaming_url)
 317             urlh = self._request_webpage(
 318                 direct_url, display_id, 'Checking for original quality', headers=headers, fatal=False)
 319             if urlh and urlhandle_detect_ext(urlh) != 'm3u8':
 320                 formats.append({
 321                     'url': direct_url,
 322                     'format_id': 'original',
 323                     'quality': 1,
 324                     **traverse_obj(result, ('value', {
 325                         'ext': ('source', (('name', {determine_ext}), ('media_type', {mimetype2ext}))),
 326                         'filesize': ('source', 'size', {int_or_none}),
 327                         'width': ('video', 'width', {int_or_none}),
 328                         'height': ('video', 'height', {int_or_none}),
 329                     }), get_all=False),
 330                     'vcodec': 'none' if stream_type == 'audio' else None,
 331                 })
 332
 333             # HEAD request returns redirect response to m3u8 URL if available
 334             final_url = self._request_webpage(
 335                 HEADRequest(streaming_url), display_id, headers=headers,
 336                 note='Downloading streaming redirect url info').url
 337
 338         elif result.get('value_type') == 'stream':
 339             claim_id, is_live = result['signing_channel']['claim_id'], True
 340             live_data = self._download_json(
 341                 'https://api.odysee.live/livestream/is_live', claim_id,
 342                 query={'channel_claim_id': claim_id},
 343                 note='Downloading livestream JSON metadata')['data']
 344             final_url = live_data.get('VideoURL')
 345             # Upcoming videos may still give VideoURL
 346             if not live_data.get('Live'):
 347                 final_url = None
 348                 self.raise_no_formats('This stream is not live', True, claim_id)
 349
 350         else:
 351             raise UnsupportedError(url)
 352
 353         if determine_ext(final_url) == 'm3u8':
 354             formats.extend(self._extract_m3u8_formats(
 355                 final_url, display_id, 'mp4', m3u8_id='hls', live=is_live, headers=headers))
 356
 357         return {
 358             **self._parse_stream(result, url),
 359             'id': claim_id,
 360             'formats': formats,
 361             'is_live': is_live,
 362             'http_headers': headers,
 363         }
 364
 365
 366 class LBRYChannelIE(LBRYBaseIE):
 367     IE_NAME = 'lbry:channel'
 368     IE_DESC = 'odysee.com channels'
 369     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + rf'(?P<id>@{LBRYBaseIE._OPT_CLAIM_ID})/?(?:[?&]|$)'
 370     _TESTS = [{
 371         'url': 'https://lbry.tv/@LBRYFoundation:0',
 372         'info_dict': {
 373             'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 374             'title': 'The LBRY Foundation',
 375             'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
 376         },
 377         'playlist_mincount': 29,
 378     }, {
 379         'url': 'https://lbry.tv/@LBRYFoundation',
 380         'only_matching': True,
 381     }, {
 382         'url': 'lbry://@lbry#3f',
 383         'only_matching': True,
 384     }]
 385
 386     def _real_extract(self, url):
 387         display_id = self._match_id(url).replace(':', '#')
 388         result = self._resolve_url(f'lbry://{display_id}', display_id, 'channel')
 389         claim_id = result['claim_id']
 390
 391         return self._playlist_entries(url, claim_id, {'channel_ids': [claim_id]}, result)
 392
 393
 394 class LBRYPlaylistIE(LBRYBaseIE):
 395     IE_NAME = 'lbry:playlist'
 396     IE_DESC = 'odysee.com playlists'
 397     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'\$/(?:play)?list/(?P<id>[0-9a-f-]+)'
 398     _TESTS = [{
 399         'url': 'https://odysee.com/$/playlist/ffef782f27486f0ac138bde8777f72ebdd0548c2',
 400         'info_dict': {
 401             'id': 'ffef782f27486f0ac138bde8777f72ebdd0548c2',
 402             'title': 'Théâtre Classique',
 403             'description': 'Théâtre Classique',
 404         },
 405         'playlist_mincount': 4,
 406     }, {
 407         'url': 'https://odysee.com/$/list/9c6658b3dd21e4f2a0602d523a13150e2b48b770',
 408         'info_dict': {
 409             'id': '9c6658b3dd21e4f2a0602d523a13150e2b48b770',
 410             'title': 'Social Media Exposed',
 411             'description': 'md5:98af97317aacd5b85d595775ea37d80e',
 412         },
 413         'playlist_mincount': 34,
 414     }, {
 415         'url': 'https://odysee.com/$/playlist/938fb11d-215f-4d1c-ad64-723954df2184',
 416         'info_dict': {
 417             'id': '938fb11d-215f-4d1c-ad64-723954df2184',
 418         },
 419         'playlist_mincount': 1000,
 420     }]
 421
 422     def _real_extract(self, url):
 423         display_id = self._match_id(url)
 424         result = traverse_obj(self._call_api_proxy('claim_search', display_id, {
 425             'claim_ids': [display_id],
 426             'no_totals': True,
 427             'page': 1,
 428             'page_size': self._PAGE_SIZE,
 429         }, 'playlist'), ('items', 0))
 430         claim_param = {'claim_ids': traverse_obj(result, ('value', 'claims', ..., {str}))}
 431
 432         return self._playlist_entries(url, display_id, claim_param, result)