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_parse_qs,
  10     compat_str,
  11     compat_urllib_parse_unquote,
  12     compat_urllib_parse_urlparse,
  13 )
  14 from ..utils import (
  15     determine_ext,
  16     ExtractorError,
  17     int_or_none,
  18     mimetype2ext,
  19     OnDemandPagedList,
  20     try_get,
  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         return 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())['result']
  40
  41     def _resolve_url(self, url, display_id, resource):
  42         return self._call_api_proxy(
  43             'resolve', display_id, {'urls': url}, resource)[url]
  44
  45     def _permanent_url(self, url, claim_name, claim_id):
  46         return urljoin(
  47             url.replace('lbry://', 'https://lbry.tv/'),
  48             '/%s:%s' % (claim_name, claim_id))
  49
  50     def _parse_stream(self, stream, url):
  51         stream_value = stream.get('value') or {}
  52         stream_type = stream_value.get('stream_type')
  53         source = stream_value.get('source') or {}
  54         media = stream_value.get(stream_type) or {}
  55         signing_channel = stream.get('signing_channel') or {}
  56         channel_name = signing_channel.get('name')
  57         channel_claim_id = signing_channel.get('claim_id')
  58         channel_url = None
  59         if channel_name and channel_claim_id:
  60             channel_url = self._permanent_url(url, channel_name, channel_claim_id)
  61
  62         info = {
  63             'thumbnail': try_get(stream_value, lambda x: x['thumbnail']['url'], compat_str),
  64             'description': stream_value.get('description'),
  65             'license': stream_value.get('license'),
  66             'timestamp': int_or_none(stream.get('timestamp')),
  67             'release_timestamp': int_or_none(stream_value.get('release_time')),
  68             'tags': stream_value.get('tags'),
  69             'duration': int_or_none(media.get('duration')),
  70             'channel': try_get(signing_channel, lambda x: x['value']['title']),
  71             'channel_id': channel_claim_id,
  72             'channel_url': channel_url,
  73             'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
  74             'filesize': int_or_none(source.get('size')),
  75         }
  76         if stream_type == 'audio':
  77             info['vcodec'] = 'none'
  78         else:
  79             info.update({
  80                 'width': int_or_none(media.get('width')),
  81                 'height': int_or_none(media.get('height')),
  82             })
  83         return info
  84
  85
  86 class LBRYIE(LBRYBaseIE):
  87     IE_NAME = 'lbry'
  88     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>\$/[^/]+/[^/]+/{1}|@{0}/{0}|(?!@){0})'.format(LBRYBaseIE._OPT_CLAIM_ID, LBRYBaseIE._CLAIM_ID_REGEX)
  89     _TESTS = [{
  90         # Video
  91         'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
  92         'md5': '65bd7ec1f6744ada55da8e4c48a2edf9',
  93         'info_dict': {
  94             'id': '17f983b61f53091fb8ea58a9c56804e4ff8cff4d',
  95             'ext': 'mp4',
  96             'title': 'First day in LBRY? Start HERE!',
  97             'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
  98             'timestamp': 1595694354,
  99             'upload_date': '20200725',
 100             'release_timestamp': 1595340697,
 101             'release_date': '20200721',
 102             'width': 1280,
 103             'height': 720,
 104         }
 105     }, {
 106         # Audio
 107         'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
 108         'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
 109         'info_dict': {
 110             'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 111             'ext': 'mp3',
 112             'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
 113             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
 114             'timestamp': 1591312601,
 115             'upload_date': '20200604',
 116             'release_timestamp': 1591312421,
 117             'release_date': '20200604',
 118             'tags': list,
 119             'duration': 2570,
 120             'channel': 'The LBRY Foundation',
 121             'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 122             'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
 123             'vcodec': 'none',
 124         }
 125     }, {
 126         # HLS
 127         'url': 'https://odysee.com/@gardeningincanada:b/plants-i-will-never-grow-again.-the:e',
 128         'md5': 'fc82f45ea54915b1495dd7cb5cc1289f',
 129         'info_dict': {
 130             'id': 'e51671357333fe22ae88aad320bde2f6f96b1410',
 131             'ext': 'mp4',
 132             'title': 'PLANTS I WILL NEVER GROW AGAIN. THE BLACK LIST PLANTS FOR A CANADIAN GARDEN | Gardening in Canada 🍁',
 133             'description': 'md5:9c539c6a03fb843956de61a4d5288d5e',
 134             'timestamp': 1618254123,
 135             'upload_date': '20210412',
 136             'release_timestamp': 1618254002,
 137             'release_date': '20210412',
 138             'tags': list,
 139             'duration': 554,
 140             'channel': 'Gardening In Canada',
 141             'channel_id': 'b8be0e93b423dad221abe29545fbe8ec36e806bc',
 142             'channel_url': 'https://odysee.com/@gardeningincanada:b8be0e93b423dad221abe29545fbe8ec36e806bc',
 143             'formats': 'mincount:3',
 144         }
 145     }, {
 146         'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
 147         'only_matching': True,
 148     }, {
 149         'url': "https://odysee.com/@ScammerRevolts:b0/I-SYSKEY'D-THE-SAME-SCAMMERS-3-TIMES!:b",
 150         'only_matching': True,
 151     }, {
 152         'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 153         'only_matching': True,
 154     }, {
 155         'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 156         'only_matching': True,
 157     }, {
 158         'url': 'https://lbry.tv/Episode-1:e7',
 159         'only_matching': True,
 160     }, {
 161         'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
 162         'only_matching': True,
 163     }, {
 164         'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 165         'only_matching': True,
 166     }, {
 167         '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',
 168         'only_matching': True,
 169     }, {
 170         'url': 'lbry://@lbry#3f/odysee#7',
 171         'only_matching': True,
 172     }]
 173
 174     def _real_extract(self, url):
 175         display_id = self._match_id(url)
 176         if display_id.startswith('$/'):
 177             display_id = display_id.split('/', 2)[-1].replace('/', ':')
 178         else:
 179             display_id = display_id.replace(':', '#')
 180         display_id = compat_urllib_parse_unquote(display_id)
 181         uri = 'lbry://' + display_id
 182         result = self._resolve_url(uri, display_id, 'stream')
 183         result_value = result['value']
 184         if result_value.get('stream_type') not in self._SUPPORTED_STREAM_TYPES:
 185             raise ExtractorError('Unsupported URL', expected=True)
 186         claim_id = result['claim_id']
 187         title = result_value['title']
 188         streaming_url = self._call_api_proxy(
 189             'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
 190         info = self._parse_stream(result, url)
 191         urlh = self._request_webpage(
 192             streaming_url, display_id, note='Downloading streaming redirect url info')
 193         if determine_ext(urlh.geturl()) == 'm3u8':
 194             info['formats'] = self._extract_m3u8_formats(
 195                 urlh.geturl(), display_id, 'mp4', entry_protocol='m3u8_native',
 196                 m3u8_id='hls')
 197             self._sort_formats(info['formats'])
 198         else:
 199             info['url'] = streaming_url
 200         info.update({
 201             'id': claim_id,
 202             'title': title,
 203         })
 204         return info
 205
 206
 207 class LBRYChannelIE(LBRYBaseIE):
 208     IE_NAME = 'lbry:channel'
 209     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
 210     _TESTS = [{
 211         'url': 'https://lbry.tv/@LBRYFoundation:0',
 212         'info_dict': {
 213             'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 214             'title': 'The LBRY Foundation',
 215             'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
 216         },
 217         'playlist_count': 29,
 218     }, {
 219         'url': 'https://lbry.tv/@LBRYFoundation',
 220         'only_matching': True,
 221     }, {
 222         'url': 'lbry://@lbry#3f',
 223         'only_matching': True,
 224     }]
 225     _PAGE_SIZE = 50
 226
 227     def _fetch_page(self, claim_id, url, params, page):
 228         page += 1
 229         page_params = {
 230             'channel_ids': [claim_id],
 231             'claim_type': 'stream',
 232             'no_totals': True,
 233             'page': page,
 234             'page_size': self._PAGE_SIZE,
 235         }
 236         page_params.update(params)
 237         result = self._call_api_proxy(
 238             'claim_search', claim_id, page_params, 'page %d' % page)
 239         for item in (result.get('items') or []):
 240             stream_claim_name = item.get('name')
 241             stream_claim_id = item.get('claim_id')
 242             if not (stream_claim_name and stream_claim_id):
 243                 continue
 244
 245             info = self._parse_stream(item, url)
 246             info.update({
 247                 '_type': 'url',
 248                 'id': stream_claim_id,
 249                 'title': try_get(item, lambda x: x['value']['title']),
 250                 'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
 251             })
 252             yield info
 253
 254     def _real_extract(self, url):
 255         display_id = self._match_id(url).replace(':', '#')
 256         result = self._resolve_url(
 257             'lbry://' + display_id, display_id, 'channel')
 258         claim_id = result['claim_id']
 259         qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query)
 260         content = qs.get('content', [None])[0]
 261         params = {
 262             'fee_amount': qs.get('fee_amount', ['>=0'])[0],
 263             'order_by': {
 264                 'new': ['release_time'],
 265                 'top': ['effective_amount'],
 266                 'trending': ['trending_group', 'trending_mixed'],
 267             }[qs.get('order', ['new'])[0]],
 268             'stream_types': [content] if content in ['audio', 'video'] else self._SUPPORTED_STREAM_TYPES,
 269         }
 270         duration = qs.get('duration', [None])[0]
 271         if duration:
 272             params['duration'] = {
 273                 'long': '>=1200',
 274                 'short': '<=240',
 275             }[duration]
 276         language = qs.get('language', ['all'])[0]
 277         if language != 'all':
 278             languages = [language]
 279             if language == 'en':
 280                 languages.append('none')
 281             params['any_languages'] = languages
 282         entries = OnDemandPagedList(
 283             functools.partial(self._fetch_page, claim_id, url, params),
 284             self._PAGE_SIZE)
 285         result_value = result.get('value') or {}
 286         return self.playlist_result(
 287             entries, claim_id, result_value.get('title'),
 288             result_value.get('description'))