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