yt_dlp/extractor/lbry.py

   1 import functools
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str, compat_urllib_parse_unquote
   6 from ..utils import (
   7     ExtractorError,
   8     HEADRequest,
   9     OnDemandPagedList,
  10     UnsupportedError,
  11     determine_ext,
  12     int_or_none,
  13     mimetype2ext,
  14     parse_qs,
  15     try_get,
  16     urljoin,
  17 )
  18
  19
  20 class LBRYBaseIE(InfoExtractor):
  21     _BASE_URL_REGEX = r'(?:https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/|lbry://)'
  22     _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
  23     _OPT_CLAIM_ID = '[^:/?#&]+(?:[:#]%s)?' % _CLAIM_ID_REGEX
  24     _SUPPORTED_STREAM_TYPES = ['video', 'audio']
  25
  26     def _call_api_proxy(self, method, display_id, params, resource):
  27         headers = {'Content-Type': 'application/json-rpc'}
  28         token = try_get(self._get_cookies('https://odysee.com'), lambda x: x['auth_token'].value)
  29         if token:
  30             headers['x-lbry-auth-token'] = token
  31         response = self._download_json(
  32             'https://api.lbry.tv/api/v1/proxy',
  33             display_id, 'Downloading %s JSON metadata' % resource,
  34             headers=headers,
  35             data=json.dumps({
  36                 'method': method,
  37                 'params': params,
  38             }).encode())
  39         err = response.get('error')
  40         if err:
  41             raise ExtractorError(
  42                 f'{self.IE_NAME} said: {err.get("code")} - {err.get("message")}', expected=True)
  43         return response['result']
  44
  45     def _resolve_url(self, url, display_id, resource):
  46         return self._call_api_proxy(
  47             'resolve', display_id, {'urls': url}, resource)[url]
  48
  49     def _permanent_url(self, url, claim_name, claim_id):
  50         return urljoin(
  51             url.replace('lbry://', 'https://lbry.tv/'),
  52             '/%s:%s' % (claim_name, claim_id))
  53
  54     def _parse_stream(self, stream, url):
  55         stream_value = stream.get('value') or {}
  56         stream_type = stream_value.get('stream_type')
  57         source = stream_value.get('source') or {}
  58         media = stream_value.get(stream_type) or {}
  59         signing_channel = stream.get('signing_channel') or {}
  60         channel_name = signing_channel.get('name')
  61         channel_claim_id = signing_channel.get('claim_id')
  62         channel_url = None
  63         if channel_name and channel_claim_id:
  64             channel_url = self._permanent_url(url, channel_name, channel_claim_id)
  65
  66         info = {
  67             'thumbnail': try_get(stream_value, lambda x: x['thumbnail']['url'], compat_str),
  68             'description': stream_value.get('description'),
  69             'license': stream_value.get('license'),
  70             'timestamp': int_or_none(stream.get('timestamp')),
  71             'release_timestamp': int_or_none(stream_value.get('release_time')),
  72             'tags': stream_value.get('tags'),
  73             'duration': int_or_none(media.get('duration')),
  74             'channel': try_get(signing_channel, lambda x: x['value']['title']),
  75             'channel_id': channel_claim_id,
  76             'channel_url': channel_url,
  77             'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
  78             'filesize': int_or_none(source.get('size')),
  79         }
  80         if stream_type == 'audio':
  81             info['vcodec'] = 'none'
  82         else:
  83             info.update({
  84                 'width': int_or_none(media.get('width')),
  85                 'height': int_or_none(media.get('height')),
  86             })
  87         return info
  88
  89
  90 class LBRYIE(LBRYBaseIE):
  91     IE_NAME = 'lbry'
  92     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>\$/[^/]+/[^/]+/{1}|@{0}/{0}|(?!@){0})'.format(LBRYBaseIE._OPT_CLAIM_ID, LBRYBaseIE._CLAIM_ID_REGEX)
  93     _TESTS = [{
  94         # Video
  95         'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
  96         'md5': 'fffd15d76062e9a985c22c7c7f2f4805',
  97         'info_dict': {
  98             'id': '17f983b61f53091fb8ea58a9c56804e4ff8cff4d',
  99             'ext': 'mp4',
 100             'title': 'First day in LBRY? Start HERE!',
 101             'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
 102             'timestamp': 1595694354,
 103             'upload_date': '20200725',
 104             'release_timestamp': 1595340697,
 105             'release_date': '20200721',
 106             'width': 1280,
 107             'height': 720,
 108             'thumbnail': 'https://spee.ch/7/67f2d809c263288c.png',
 109             'license': 'None',
 110             'duration': 346,
 111             'channel': 'LBRY/Odysee rats united!!!',
 112             'channel_id': '1c8ad6a2ab4e889a71146ae4deeb23bb92dab627',
 113             'channel_url': 'https://lbry.tv/@Mantega:1c8ad6a2ab4e889a71146ae4deeb23bb92dab627',
 114             'tags': [
 115                 'first day in lbry',
 116                 'lbc',
 117                 'lbry',
 118                 'start',
 119                 'tutorial'
 120             ],
 121         }
 122     }, {
 123         # Audio
 124         'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
 125         'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
 126         'info_dict': {
 127             'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 128             'ext': 'mp3',
 129             'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
 130             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
 131             'timestamp': 1591312601,
 132             'upload_date': '20200604',
 133             'release_timestamp': 1591312421,
 134             'release_date': '20200604',
 135             'tags': list,
 136             'duration': 2570,
 137             'channel': 'The LBRY Foundation',
 138             'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 139             'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
 140             'vcodec': 'none',
 141             'thumbnail': 'https://spee.ch/d/0bc63b0e6bf1492d.png',
 142             'license': 'None',
 143         }
 144     }, {
 145         # HLS
 146         'url': 'https://odysee.com/@gardeningincanada:b/plants-i-will-never-grow-again.-the:e',
 147         'md5': '25049011f3c8bc2f8b60ad88a031837e',
 148         'info_dict': {
 149             'id': 'e51671357333fe22ae88aad320bde2f6f96b1410',
 150             'ext': 'mp4',
 151             'title': 'PLANTS I WILL NEVER GROW AGAIN. THE BLACK LIST PLANTS FOR A CANADIAN GARDEN | Gardening in Canada 🍁',
 152             'description': 'md5:9c539c6a03fb843956de61a4d5288d5e',
 153             'timestamp': 1618254123,
 154             'upload_date': '20210412',
 155             'release_timestamp': 1618254002,
 156             'release_date': '20210412',
 157             'tags': list,
 158             'duration': 554,
 159             'channel': 'Gardening In Canada',
 160             'channel_id': 'b8be0e93b423dad221abe29545fbe8ec36e806bc',
 161             'channel_url': 'https://odysee.com/@gardeningincanada:b8be0e93b423dad221abe29545fbe8ec36e806bc',
 162             'formats': 'mincount:3',
 163             'thumbnail': 'https://thumbnails.lbry.com/AgHSc_HzrrE',
 164             'license': 'Copyrighted (contact publisher)',
 165         }
 166     }, {
 167         # HLS live stream (might expire)
 168         'url': 'https://odysee.com/@RT:fd/livestream_RT:d',
 169         'info_dict': {
 170             'id': 'fdd11cb3ab75f95efb7b3bc2d726aa13ac915b66',
 171             'ext': 'mp4',
 172             'live_status': 'is_live',
 173             'title': 'startswith:RT News | Livestream 24/7',
 174             'description': 'md5:fe68d0056dfe79c1a6b8ce8c34d5f6fa',
 175             'timestamp': int,
 176             'upload_date': str,
 177             'release_timestamp': int,
 178             'release_date': str,
 179             'tags': list,
 180             'duration': None,
 181             'channel': 'RT',
 182             'channel_id': 'fdd11cb3ab75f95efb7b3bc2d726aa13ac915b66',
 183             'channel_url': 'https://odysee.com/@RT:fdd11cb3ab75f95efb7b3bc2d726aa13ac915b66',
 184             'formats': 'mincount:1',
 185             'thumbnail': 'startswith:https://thumb',
 186             'license': 'None',
 187         },
 188         'params': {'skip_download': True}
 189     }, {
 190         'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
 191         'only_matching': True,
 192     }, {
 193         'url': 'https://odysee.com/@ScammerRevolts:b0/I-SYSKEY\'D-THE-SAME-SCAMMERS-3-TIMES!:b',
 194         'only_matching': True,
 195     }, {
 196         'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 197         'only_matching': True,
 198     }, {
 199         'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 200         'only_matching': True,
 201     }, {
 202         'url': 'https://lbry.tv/Episode-1:e7',
 203         'only_matching': True,
 204     }, {
 205         'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
 206         'only_matching': True,
 207     }, {
 208         'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 209         'only_matching': True,
 210     }, {
 211         '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',
 212         'only_matching': True,
 213     }, {
 214         'url': 'lbry://@lbry#3f/odysee#7',
 215         'only_matching': True,
 216     }]
 217
 218     def _real_extract(self, url):
 219         display_id = self._match_id(url)
 220         if display_id.startswith('$/'):
 221             display_id = display_id.split('/', 2)[-1].replace('/', ':')
 222         else:
 223             display_id = display_id.replace(':', '#')
 224         display_id = compat_urllib_parse_unquote(display_id)
 225         uri = 'lbry://' + display_id
 226         result = self._resolve_url(uri, display_id, 'stream')
 227         headers = {'Referer': 'https://odysee.com/'}
 228         if result['value'].get('stream_type') in self._SUPPORTED_STREAM_TYPES:
 229             claim_id, is_live = result['claim_id'], False
 230             streaming_url = self._call_api_proxy(
 231                 'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
 232             final_url = self._request_webpage(
 233                 HEADRequest(streaming_url), display_id, headers=headers,
 234                 note='Downloading streaming redirect url info').geturl()
 235         elif result.get('value_type') == 'stream':
 236             claim_id, is_live = result['signing_channel']['claim_id'], True
 237             live_data = self._download_json(
 238                 'https://api.odysee.live/livestream/is_live', claim_id,
 239                 query={'channel_claim_id': claim_id},
 240                 note='Downloading livestream JSON metadata')['data']
 241             streaming_url = final_url = live_data.get('VideoURL')
 242             # Upcoming videos may still give VideoURL
 243             if not live_data.get('Live'):
 244                 streaming_url = final_url = None
 245                 self.raise_no_formats('This stream is not live', True, claim_id)
 246         else:
 247             raise UnsupportedError(url)
 248
 249         info = self._parse_stream(result, url)
 250         if determine_ext(final_url) == 'm3u8':
 251             info['formats'] = self._extract_m3u8_formats(
 252                 final_url, display_id, 'mp4', 'm3u8_native', m3u8_id='hls', live=is_live, headers=headers)
 253         else:
 254             info['url'] = streaming_url
 255         return {
 256             **info,
 257             'id': claim_id,
 258             'title': result['value']['title'],
 259             'is_live': is_live,
 260             'http_headers': headers,
 261         }
 262
 263
 264 class LBRYChannelIE(LBRYBaseIE):
 265     IE_NAME = 'lbry:channel'
 266     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
 267     _TESTS = [{
 268         'url': 'https://lbry.tv/@LBRYFoundation:0',
 269         'info_dict': {
 270             'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 271             'title': 'The LBRY Foundation',
 272             'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
 273         },
 274         'playlist_mincount': 29,
 275     }, {
 276         'url': 'https://lbry.tv/@LBRYFoundation',
 277         'only_matching': True,
 278     }, {
 279         'url': 'lbry://@lbry#3f',
 280         'only_matching': True,
 281     }]
 282     _PAGE_SIZE = 50
 283
 284     def _fetch_page(self, claim_id, url, params, page):
 285         page += 1
 286         page_params = {
 287             'channel_ids': [claim_id],
 288             'claim_type': 'stream',
 289             'no_totals': True,
 290             'page': page,
 291             'page_size': self._PAGE_SIZE,
 292         }
 293         page_params.update(params)
 294         result = self._call_api_proxy(
 295             'claim_search', claim_id, page_params, 'page %d' % page)
 296         for item in (result.get('items') or []):
 297             stream_claim_name = item.get('name')
 298             stream_claim_id = item.get('claim_id')
 299             if not (stream_claim_name and stream_claim_id):
 300                 continue
 301
 302             info = self._parse_stream(item, url)
 303             info.update({
 304                 '_type': 'url',
 305                 'id': stream_claim_id,
 306                 'title': try_get(item, lambda x: x['value']['title']),
 307                 'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
 308             })
 309             yield info
 310
 311     def _real_extract(self, url):
 312         display_id = self._match_id(url).replace(':', '#')
 313         result = self._resolve_url(
 314             'lbry://' + display_id, display_id, 'channel')
 315         claim_id = result['claim_id']
 316         qs = parse_qs(url)
 317         content = qs.get('content', [None])[0]
 318         params = {
 319             'fee_amount': qs.get('fee_amount', ['>=0'])[0],
 320             'order_by': {
 321                 'new': ['release_time'],
 322                 'top': ['effective_amount'],
 323                 'trending': ['trending_group', 'trending_mixed'],
 324             }[qs.get('order', ['new'])[0]],
 325             'stream_types': [content] if content in ['audio', 'video'] else self._SUPPORTED_STREAM_TYPES,
 326         }
 327         duration = qs.get('duration', [None])[0]
 328         if duration:
 329             params['duration'] = {
 330                 'long': '>=1200',
 331                 'short': '<=240',
 332             }[duration]
 333         language = qs.get('language', ['all'])[0]
 334         if language != 'all':
 335             languages = [language]
 336             if language == 'en':
 337                 languages.append('none')
 338             params['any_languages'] = languages
 339         entries = OnDemandPagedList(
 340             functools.partial(self._fetch_page, claim_id, url, params),
 341             self._PAGE_SIZE)
 342         result_value = result.get('value') or {}
 343         return self.playlist_result(
 344             entries, claim_id, result_value.get('title'),
 345             result_value.get('description'))