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