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     OnDemandPagedList,
  18     try_get,
  19     urljoin,
  20 )
  21
  22
  23 class LBRYBaseIE(InfoExtractor):
  24     _BASE_URL_REGEX = r'https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/'
  25     _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
  26     _OPT_CLAIM_ID = '[^:/?#&]+(?::%s)?' % _CLAIM_ID_REGEX
  27     _SUPPORTED_STREAM_TYPES = ['video', 'audio']
  28
  29     def _call_api_proxy(self, method, display_id, params, resource):
  30         return self._download_json(
  31             'https://api.lbry.tv/api/v1/proxy',
  32             display_id, 'Downloading %s JSON metadata' % resource,
  33             headers={'Content-Type': 'application/json-rpc'},
  34             data=json.dumps({
  35                 'method': method,
  36                 'params': params,
  37             }).encode())['result']
  38
  39     def _resolve_url(self, url, display_id, resource):
  40         return self._call_api_proxy(
  41             'resolve', display_id, {'urls': url}, resource)[url]
  42
  43     def _permanent_url(self, url, claim_name, claim_id):
  44         return urljoin(url, '/%s:%s' % (claim_name, claim_id))
  45
  46     def _parse_stream(self, stream, url):
  47         stream_value = stream.get('value') or {}
  48         stream_type = stream_value.get('stream_type')
  49         source = stream_value.get('source') or {}
  50         media = stream_value.get(stream_type) or {}
  51         signing_channel = stream.get('signing_channel') or {}
  52         channel_name = signing_channel.get('name')
  53         channel_claim_id = signing_channel.get('claim_id')
  54         channel_url = None
  55         if channel_name and channel_claim_id:
  56             channel_url = self._permanent_url(url, channel_name, channel_claim_id)
  57
  58         info = {
  59             'thumbnail': try_get(stream_value, lambda x: x['thumbnail']['url'], compat_str),
  60             'description': stream_value.get('description'),
  61             'license': stream_value.get('license'),
  62             'timestamp': int_or_none(stream.get('timestamp')),
  63             'tags': stream_value.get('tags'),
  64             'duration': int_or_none(media.get('duration')),
  65             'channel': try_get(signing_channel, lambda x: x['value']['title']),
  66             'channel_id': channel_claim_id,
  67             'channel_url': channel_url,
  68             'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
  69             'filesize': int_or_none(source.get('size')),
  70         }
  71         if stream_type == 'audio':
  72             info['vcodec'] = 'none'
  73         else:
  74             info.update({
  75                 'width': int_or_none(media.get('width')),
  76                 'height': int_or_none(media.get('height')),
  77             })
  78         return info
  79
  80
  81 class LBRYIE(LBRYBaseIE):
  82     IE_NAME = 'lbry'
  83     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>\$/[^/]+/[^/]+/{1}|@{0}/{0}|(?!@){0})'.format(LBRYBaseIE._OPT_CLAIM_ID, LBRYBaseIE._CLAIM_ID_REGEX)
  84     _TESTS = [{
  85         # Video
  86         'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
  87         'md5': '65bd7ec1f6744ada55da8e4c48a2edf9',
  88         'info_dict': {
  89             'id': '17f983b61f53091fb8ea58a9c56804e4ff8cff4d',
  90             'ext': 'mp4',
  91             'title': 'First day in LBRY? Start HERE!',
  92             'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
  93             'timestamp': 1595694354,
  94             'upload_date': '20200725',
  95             'width': 1280,
  96             'height': 720,
  97         }
  98     }, {
  99         # Audio
 100         'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
 101         'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
 102         'info_dict': {
 103             'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 104             'ext': 'mp3',
 105             'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
 106             'description': 'md5:661ac4f1db09f31728931d7b88807a61',
 107             'timestamp': 1591312601,
 108             'upload_date': '20200604',
 109             'tags': list,
 110             'duration': 2570,
 111             'channel': 'The LBRY Foundation',
 112             'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 113             'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
 114             'vcodec': 'none',
 115         }
 116     }, {
 117         'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
 118         'only_matching': True,
 119     }, {
 120         'url': "https://odysee.com/@ScammerRevolts:b0/I-SYSKEY'D-THE-SAME-SCAMMERS-3-TIMES!:b",
 121         'only_matching': True,
 122     }, {
 123         'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 124         'only_matching': True,
 125     }, {
 126         'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 127         'only_matching': True,
 128     }, {
 129         'url': 'https://lbry.tv/Episode-1:e7',
 130         'only_matching': True,
 131     }, {
 132         'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
 133         'only_matching': True,
 134     }, {
 135         'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
 136         'only_matching': True,
 137     }, {
 138         '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',
 139         'only_matching': True,
 140     }]
 141
 142     def _real_extract(self, url):
 143         display_id = self._match_id(url)
 144         if display_id.startswith('$/'):
 145             display_id = display_id.split('/', 2)[-1].replace('/', ':')
 146         else:
 147             display_id = display_id.replace(':', '#')
 148         display_id = compat_urllib_parse_unquote(display_id)
 149         uri = 'lbry://' + display_id
 150         result = self._resolve_url(uri, display_id, 'stream')
 151         result_value = result['value']
 152         if result_value.get('stream_type') not in self._SUPPORTED_STREAM_TYPES:
 153             raise ExtractorError('Unsupported URL', expected=True)
 154         claim_id = result['claim_id']
 155         title = result_value['title']
 156         streaming_url = self._call_api_proxy(
 157             'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
 158         info = self._parse_stream(result, url)
 159         info.update({
 160             'id': claim_id,
 161             'title': title,
 162             'url': streaming_url,
 163         })
 164         return info
 165
 166
 167 class LBRYChannelIE(LBRYBaseIE):
 168     IE_NAME = 'lbry:channel'
 169     _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?#&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
 170     _TESTS = [{
 171         'url': 'https://lbry.tv/@LBRYFoundation:0',
 172         'info_dict': {
 173             'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
 174             'title': 'The LBRY Foundation',
 175             'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
 176         },
 177         'playlist_count': 29,
 178     }, {
 179         'url': 'https://lbry.tv/@LBRYFoundation',
 180         'only_matching': True,
 181     }]
 182     _PAGE_SIZE = 50
 183
 184     def _fetch_page(self, claim_id, url, page):
 185         page += 1
 186         result = self._call_api_proxy(
 187             'claim_search', claim_id, {
 188                 'channel_ids': [claim_id],
 189                 'claim_type': 'stream',
 190                 'no_totals': True,
 191                 'page': page,
 192                 'page_size': self._PAGE_SIZE,
 193                 'stream_types': self._SUPPORTED_STREAM_TYPES,
 194             }, 'page %d' % page)
 195         for item in (result.get('items') or []):
 196             stream_claim_name = item.get('name')
 197             stream_claim_id = item.get('claim_id')
 198             if not (stream_claim_name and stream_claim_id):
 199                 continue
 200
 201             info = self._parse_stream(item, url)
 202             info.update({
 203                 '_type': 'url',
 204                 'id': stream_claim_id,
 205                 'title': try_get(item, lambda x: x['value']['title']),
 206                 'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
 207             })
 208             yield info
 209
 210     def _real_extract(self, url):
 211         display_id = self._match_id(url).replace(':', '#')
 212         result = self._resolve_url(
 213             'lbry://' + display_id, display_id, 'channel')
 214         claim_id = result['claim_id']
 215         entries = OnDemandPagedList(
 216             functools.partial(self._fetch_page, claim_id, url),
 217             self._PAGE_SIZE)
 218         result_value = result.get('value') or {}
 219         return self.playlist_result(
 220             entries, claim_id, result_value.get('title'),
 221             result_value.get('description'))