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