yt_dlp/extractor/lbry.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import functools
   5 import json
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_str,
  10     compat_urllib_parse_unquote,
  11 )
  12 from ..utils import (
  13     determine_ext,
  14     ExtractorError,
  15     int_or_none,
  16     mimetype2ext,
  17     parse_qs,
  18     OnDemandPagedList,
  19     try_get,
  20     urljoin,
  21 )
  22
  23
  24 class LBRYBaseIE(InfoExtractor):
  25     _BASE_URL_REGEX = r'(?:https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/|lbry://)'
  26     _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
  27     _OPT_CLAIM_ID = '[^:/?#&]+(?:[:#]%s)?' % _CLAIM_ID_REGEX
  28     _SUPPORTED_STREAM_TYPES = ['video', 'audio']
  29
  30     def _call_api_proxy(self, method, display_id, params, resource):
  31         response = self._download_json(
  32             'https://api.lbry.tv/api/v1/proxy',
  33             display_id, 'Downloading %s JSON metadata' % resource,
  34             headers={'Content-Type': 'application/json-rpc'},
  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': '65bd7ec1f6744ada55da8e4c48a2edf9',
  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         }
 109     }, {
 110         # Audio
 111         'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
 112         'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
 113         'info_dict': {
 114             'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 115             'ext': 'mp3',
 116             'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
 117             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
 118             'timestamp': 1591312601,
 119             'upload_date': '20200604',
 120             'release_timestamp': 1591312421,
 121             'release_date': '20200604',
 122             'tags': list,
 123             'duration': 2570,
 124             'channel': 'The LBRY Foundation',
 125             'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 126             'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
 127             'vcodec': 'none',
 128         }
 129     }, {
 130         # HLS
 131         'url': 'https://odysee.com/@gardeningincanada:b/plants-i-will-never-grow-again.-the:e',
 132         'md5': 'fc82f45ea54915b1495dd7cb5cc1289f',
 133         'info_dict': {
 134             'id': 'e51671357333fe22ae88aad320bde2f6f96b1410',
 135             'ext': 'mp4',
 136             'title': 'PLANTS I WILL NEVER GROW AGAIN. THE BLACK LIST PLANTS FOR A CANADIAN GARDEN | Gardening in Canada 🍁',
 137             'description': 'md5:9c539c6a03fb843956de61a4d5288d5e',
 138             'timestamp': 1618254123,
 139             'upload_date': '20210412',
 140             'release_timestamp': 1618254002,
 141             'release_date': '20210412',
 142             'tags': list,
 143             'duration': 554,
 144             'channel': 'Gardening In Canada',
 145             'channel_id': 'b8be0e93b423dad221abe29545fbe8ec36e806bc',
 146             'channel_url': 'https://odysee.com/@gardeningincanada:b8be0e93b423dad221abe29545fbe8ec36e806bc',
 147             'formats': 'mincount:3',
 148         }
 149     }, {
 150         'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
 151         'only_matching': True,
 152     }, {
 153         'url': "https://odysee.com/@ScammerRevolts:b0/I-SYSKEY'D-THE-SAME-SCAMMERS-3-TIMES!:b",
 154         'only_matching': True,
 155     }, {
 156         'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 157         'only_matching': True,
 158     }, {
 159         'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 160         'only_matching': True,
 161     }, {
 162         'url': 'https://lbry.tv/Episode-1:e7',
 163         'only_matching': True,
 164     }, {
 165         'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
 166         'only_matching': True,
 167     }, {
 168         'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 169         'only_matching': True,
 170     }, {
 171         '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',
 172         'only_matching': True,
 173     }, {
 174         'url': 'lbry://@lbry#3f/odysee#7',
 175         'only_matching': True,
 176     }]
 177
 178     def _real_extract(self, url):
 179         display_id = self._match_id(url)
 180         if display_id.startswith('$/'):
 181             display_id = display_id.split('/', 2)[-1].replace('/', ':')
 182         else:
 183             display_id = display_id.replace(':', '#')
 184         display_id = compat_urllib_parse_unquote(display_id)
 185         uri = 'lbry://' + display_id
 186         result = self._resolve_url(uri, display_id, 'stream')
 187         if result['value'].get('stream_type') in self._SUPPORTED_STREAM_TYPES:
 188             claim_id, is_live, headers = result['claim_id'], False, None
 189             streaming_url = self._call_api_proxy(
 190                 'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
 191             final_url = self._request_webpage(
 192                 streaming_url, display_id, note='Downloading streaming redirect url info').geturl()
 193         elif result.get('value_type') == 'stream':
 194             claim_id, is_live = result['signing_channel']['claim_id'], True
 195             headers = {'referer': 'https://player.odysee.live/'}
 196             live_data = self._download_json(
 197                 f'https://api.live.odysee.com/v1/odysee/live/{claim_id}', claim_id,
 198                 note='Downloading livestream JSON metadata')['data']
 199             if not live_data['live']:
 200                 raise ExtractorError('This stream is not live', expected=True)
 201             streaming_url = final_url = live_data['url']
 202         else:
 203             raise ExtractorError('Unsupported URL', expected=True)
 204
 205         info = self._parse_stream(result, url)
 206         if determine_ext(final_url) == 'm3u8':
 207             info['formats'] = self._extract_m3u8_formats(
 208                 final_url, display_id, 'mp4', 'm3u8_native', m3u8_id='hls', live=is_live, headers=headers)
 209             self._sort_formats(info['formats'])
 210         else:
 211             info['url'] = streaming_url
 212         return {
 213             **info,
 214             'id': claim_id,
 215             'title': result['value']['title'],
 216             'is_live': is_live,
 217             'http_headers': headers,
 218         }
 219
 220
 221 class LBRYChannelIE(LBRYBaseIE):
 222     IE_NAME = 'lbry:channel'
 223     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
 224     _TESTS = [{
 225         'url': 'https://lbry.tv/@LBRYFoundation:0',
 226         'info_dict': {
 227             'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 228             'title': 'The LBRY Foundation',
 229             'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
 230         },
 231         'playlist_count': 29,
 232     }, {
 233         'url': 'https://lbry.tv/@LBRYFoundation',
 234         'only_matching': True,
 235     }, {
 236         'url': 'lbry://@lbry#3f',
 237         'only_matching': True,
 238     }]
 239     _PAGE_SIZE = 50
 240
 241     def _fetch_page(self, claim_id, url, params, page):
 242         page += 1
 243         page_params = {
 244             'channel_ids': [claim_id],
 245             'claim_type': 'stream',
 246             'no_totals': True,
 247             'page': page,
 248             'page_size': self._PAGE_SIZE,
 249         }
 250         page_params.update(params)
 251         result = self._call_api_proxy(
 252             'claim_search', claim_id, page_params, 'page %d' % page)
 253         for item in (result.get('items') or []):
 254             stream_claim_name = item.get('name')
 255             stream_claim_id = item.get('claim_id')
 256             if not (stream_claim_name and stream_claim_id):
 257                 continue
 258
 259             info = self._parse_stream(item, url)
 260             info.update({
 261                 '_type': 'url',
 262                 'id': stream_claim_id,
 263                 'title': try_get(item, lambda x: x['value']['title']),
 264                 'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
 265             })
 266             yield info
 267
 268     def _real_extract(self, url):
 269         display_id = self._match_id(url).replace(':', '#')
 270         result = self._resolve_url(
 271             'lbry://' + display_id, display_id, 'channel')
 272         claim_id = result['claim_id']
 273         qs = parse_qs(url)
 274         content = qs.get('content', [None])[0]
 275         params = {
 276             'fee_amount': qs.get('fee_amount', ['>=0'])[0],
 277             'order_by': {
 278                 'new': ['release_time'],
 279                 'top': ['effective_amount'],
 280                 'trending': ['trending_group', 'trending_mixed'],
 281             }[qs.get('order', ['new'])[0]],
 282             'stream_types': [content] if content in ['audio', 'video'] else self._SUPPORTED_STREAM_TYPES,
 283         }
 284         duration = qs.get('duration', [None])[0]
 285         if duration:
 286             params['duration'] = {
 287                 'long': '>=1200',
 288                 'short': '<=240',
 289             }[duration]
 290         language = qs.get('language', ['all'])[0]
 291         if language != 'all':
 292             languages = [language]
 293             if language == 'en':
 294                 languages.append('none')
 295             params['any_languages'] = languages
 296         entries = OnDemandPagedList(
 297             functools.partial(self._fetch_page, claim_id, url, params),
 298             self._PAGE_SIZE)
 299         result_value = result.get('value') or {}
 300         return self.playlist_result(
 301             entries, claim_id, result_value.get('title'),
 302             result_value.get('description'))