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     UnsupportedError,
  21     urljoin,
  22 )
  23
  24
  25 class LBRYBaseIE(InfoExtractor):
  26     _BASE_URL_REGEX = r'(?:https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/|lbry://)'
  27     _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
  28     _OPT_CLAIM_ID = '[^:/?#&]+(?:[:#]%s)?' % _CLAIM_ID_REGEX
  29     _SUPPORTED_STREAM_TYPES = ['video', 'audio']
  30
  31     def _call_api_proxy(self, method, display_id, params, resource):
  32         response = self._download_json(
  33             'https://api.lbry.tv/api/v1/proxy',
  34             display_id, 'Downloading %s JSON metadata' % resource,
  35             headers={'Content-Type': 'application/json-rpc'},
  36             data=json.dumps({
  37                 'method': method,
  38                 'params': params,
  39             }).encode())
  40         err = response.get('error')
  41         if err:
  42             raise ExtractorError(
  43                 f'{self.IE_NAME} said: {err.get("code")} - {err.get("message")}', expected=True)
  44         return response['result']
  45
  46     def _resolve_url(self, url, display_id, resource):
  47         return self._call_api_proxy(
  48             'resolve', display_id, {'urls': url}, resource)[url]
  49
  50     def _permanent_url(self, url, claim_name, claim_id):
  51         return urljoin(
  52             url.replace('lbry://', 'https://lbry.tv/'),
  53             '/%s:%s' % (claim_name, claim_id))
  54
  55     def _parse_stream(self, stream, url):
  56         stream_value = stream.get('value') or {}
  57         stream_type = stream_value.get('stream_type')
  58         source = stream_value.get('source') or {}
  59         media = stream_value.get(stream_type) or {}
  60         signing_channel = stream.get('signing_channel') or {}
  61         channel_name = signing_channel.get('name')
  62         channel_claim_id = signing_channel.get('claim_id')
  63         channel_url = None
  64         if channel_name and channel_claim_id:
  65             channel_url = self._permanent_url(url, channel_name, channel_claim_id)
  66
  67         info = {
  68             'thumbnail': try_get(stream_value, lambda x: x['thumbnail']['url'], compat_str),
  69             'description': stream_value.get('description'),
  70             'license': stream_value.get('license'),
  71             'timestamp': int_or_none(stream.get('timestamp')),
  72             'release_timestamp': int_or_none(stream_value.get('release_time')),
  73             'tags': stream_value.get('tags'),
  74             'duration': int_or_none(media.get('duration')),
  75             'channel': try_get(signing_channel, lambda x: x['value']['title']),
  76             'channel_id': channel_claim_id,
  77             'channel_url': channel_url,
  78             'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
  79             'filesize': int_or_none(source.get('size')),
  80         }
  81         if stream_type == 'audio':
  82             info['vcodec'] = 'none'
  83         else:
  84             info.update({
  85                 'width': int_or_none(media.get('width')),
  86                 'height': int_or_none(media.get('height')),
  87             })
  88         return info
  89
  90
  91 class LBRYIE(LBRYBaseIE):
  92     IE_NAME = 'lbry'
  93     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>\$/[^/]+/[^/]+/{1}|@{0}/{0}|(?!@){0})'.format(LBRYBaseIE._OPT_CLAIM_ID, LBRYBaseIE._CLAIM_ID_REGEX)
  94     _TESTS = [{
  95         # Video
  96         'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
  97         'md5': '65bd7ec1f6744ada55da8e4c48a2edf9',
  98         'info_dict': {
  99             'id': '17f983b61f53091fb8ea58a9c56804e4ff8cff4d',
 100             'ext': 'mp4',
 101             'title': 'First day in LBRY? Start HERE!',
 102             'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
 103             'timestamp': 1595694354,
 104             'upload_date': '20200725',
 105             'release_timestamp': 1595340697,
 106             'release_date': '20200721',
 107             'width': 1280,
 108             'height': 720,
 109         }
 110     }, {
 111         # Audio
 112         'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
 113         'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
 114         'info_dict': {
 115             'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 116             'ext': 'mp3',
 117             'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
 118             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
 119             'timestamp': 1591312601,
 120             'upload_date': '20200604',
 121             'release_timestamp': 1591312421,
 122             'release_date': '20200604',
 123             'tags': list,
 124             'duration': 2570,
 125             'channel': 'The LBRY Foundation',
 126             'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 127             'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
 128             'vcodec': 'none',
 129         }
 130     }, {
 131         # HLS
 132         'url': 'https://odysee.com/@gardeningincanada:b/plants-i-will-never-grow-again.-the:e',
 133         'md5': 'fc82f45ea54915b1495dd7cb5cc1289f',
 134         'info_dict': {
 135             'id': 'e51671357333fe22ae88aad320bde2f6f96b1410',
 136             'ext': 'mp4',
 137             'title': 'PLANTS I WILL NEVER GROW AGAIN. THE BLACK LIST PLANTS FOR A CANADIAN GARDEN | Gardening in Canada 🍁',
 138             'description': 'md5:9c539c6a03fb843956de61a4d5288d5e',
 139             'timestamp': 1618254123,
 140             'upload_date': '20210412',
 141             'release_timestamp': 1618254002,
 142             'release_date': '20210412',
 143             'tags': list,
 144             'duration': 554,
 145             'channel': 'Gardening In Canada',
 146             'channel_id': 'b8be0e93b423dad221abe29545fbe8ec36e806bc',
 147             'channel_url': 'https://odysee.com/@gardeningincanada:b8be0e93b423dad221abe29545fbe8ec36e806bc',
 148             'formats': 'mincount:3',
 149         }
 150     }, {
 151         'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
 152         'only_matching': True,
 153     }, {
 154         'url': "https://odysee.com/@ScammerRevolts:b0/I-SYSKEY'D-THE-SAME-SCAMMERS-3-TIMES!:b",
 155         'only_matching': True,
 156     }, {
 157         'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 158         'only_matching': True,
 159     }, {
 160         'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 161         'only_matching': True,
 162     }, {
 163         'url': 'https://lbry.tv/Episode-1:e7',
 164         'only_matching': True,
 165     }, {
 166         'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
 167         'only_matching': True,
 168     }, {
 169         'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 170         'only_matching': True,
 171     }, {
 172         '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',
 173         'only_matching': True,
 174     }, {
 175         'url': 'lbry://@lbry#3f/odysee#7',
 176         'only_matching': True,
 177     }]
 178
 179     def _real_extract(self, url):
 180         display_id = self._match_id(url)
 181         if display_id.startswith('$/'):
 182             display_id = display_id.split('/', 2)[-1].replace('/', ':')
 183         else:
 184             display_id = display_id.replace(':', '#')
 185         display_id = compat_urllib_parse_unquote(display_id)
 186         uri = 'lbry://' + display_id
 187         result = self._resolve_url(uri, display_id, 'stream')
 188         if result['value'].get('stream_type') in self._SUPPORTED_STREAM_TYPES:
 189             claim_id, is_live, headers = result['claim_id'], False, None
 190             streaming_url = self._call_api_proxy(
 191                 'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
 192             final_url = self._request_webpage(
 193                 streaming_url, display_id, note='Downloading streaming redirect url info').geturl()
 194         elif result.get('value_type') == 'stream':
 195             claim_id, is_live = result['signing_channel']['claim_id'], True
 196             headers = {'referer': 'https://player.odysee.live/'}
 197             live_data = self._download_json(
 198                 f'https://api.live.odysee.com/v1/odysee/live/{claim_id}', claim_id,
 199                 note='Downloading livestream JSON metadata')['data']
 200             streaming_url = final_url = live_data.get('url')
 201             if not final_url and not live_data.get('live'):
 202                 self.raise_no_formats('This stream is not live', True, claim_id)
 203         else:
 204             raise UnsupportedError(url)
 205
 206         info = self._parse_stream(result, url)
 207         if determine_ext(final_url) == 'm3u8':
 208             info['formats'] = self._extract_m3u8_formats(
 209                 final_url, display_id, 'mp4', 'm3u8_native', m3u8_id='hls', live=is_live, headers=headers)
 210             self._sort_formats(info['formats'])
 211         else:
 212             info['url'] = streaming_url
 213         return {
 214             **info,
 215             'id': claim_id,
 216             'title': result['value']['title'],
 217             'is_live': is_live,
 218             'http_headers': headers,
 219         }
 220
 221
 222 class LBRYChannelIE(LBRYBaseIE):
 223     IE_NAME = 'lbry:channel'
 224     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
 225     _TESTS = [{
 226         'url': 'https://lbry.tv/@LBRYFoundation:0',
 227         'info_dict': {
 228             'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 229             'title': 'The LBRY Foundation',
 230             'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
 231         },
 232         'playlist_count': 29,
 233     }, {
 234         'url': 'https://lbry.tv/@LBRYFoundation',
 235         'only_matching': True,
 236     }, {
 237         'url': 'lbry://@lbry#3f',
 238         'only_matching': True,
 239     }]
 240     _PAGE_SIZE = 50
 241
 242     def _fetch_page(self, claim_id, url, params, page):
 243         page += 1
 244         page_params = {
 245             'channel_ids': [claim_id],
 246             'claim_type': 'stream',
 247             'no_totals': True,
 248             'page': page,
 249             'page_size': self._PAGE_SIZE,
 250         }
 251         page_params.update(params)
 252         result = self._call_api_proxy(
 253             'claim_search', claim_id, page_params, 'page %d' % page)
 254         for item in (result.get('items') or []):
 255             stream_claim_name = item.get('name')
 256             stream_claim_id = item.get('claim_id')
 257             if not (stream_claim_name and stream_claim_id):
 258                 continue
 259
 260             info = self._parse_stream(item, url)
 261             info.update({
 262                 '_type': 'url',
 263                 'id': stream_claim_id,
 264                 'title': try_get(item, lambda x: x['value']['title']),
 265                 'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
 266             })
 267             yield info
 268
 269     def _real_extract(self, url):
 270         display_id = self._match_id(url).replace(':', '#')
 271         result = self._resolve_url(
 272             'lbry://' + display_id, display_id, 'channel')
 273         claim_id = result['claim_id']
 274         qs = parse_qs(url)
 275         content = qs.get('content', [None])[0]
 276         params = {
 277             'fee_amount': qs.get('fee_amount', ['>=0'])[0],
 278             'order_by': {
 279                 'new': ['release_time'],
 280                 'top': ['effective_amount'],
 281                 'trending': ['trending_group', 'trending_mixed'],
 282             }[qs.get('order', ['new'])[0]],
 283             'stream_types': [content] if content in ['audio', 'video'] else self._SUPPORTED_STREAM_TYPES,
 284         }
 285         duration = qs.get('duration', [None])[0]
 286         if duration:
 287             params['duration'] = {
 288                 'long': '>=1200',
 289                 'short': '<=240',
 290             }[duration]
 291         language = qs.get('language', ['all'])[0]
 292         if language != 'all':
 293             languages = [language]
 294             if language == 'en':
 295                 languages.append('none')
 296             params['any_languages'] = languages
 297         entries = OnDemandPagedList(
 298             functools.partial(self._fetch_page, claim_id, url, params),
 299             self._PAGE_SIZE)
 300         result_value = result.get('value') or {}
 301         return self.playlist_result(
 302             entries, claim_id, result_value.get('title'),
 303             result_value.get('description'))