[yt-dlp.git] / yt_dlp / extractor / lbry.py

# coding: utf-8
from __future__ import unicode_literals

import functools
import json

from .common import InfoExtractor
from ..compat import (
    compat_str,
    compat_urllib_parse_unquote,
)
from ..utils import (
    determine_ext,
    ExtractorError,
    int_or_none,
    mimetype2ext,
    OnDemandPagedList,
    try_get,
    urljoin,
)


class LBRYBaseIE(InfoExtractor):
    _BASE_URL_REGEX = r'(?:https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/|lbry://)'
    _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
    _OPT_CLAIM_ID = '[^:/?#&]+(?:[:#]%s)?' % _CLAIM_ID_REGEX
    _SUPPORTED_STREAM_TYPES = ['video', 'audio']

    def _call_api_proxy(self, method, display_id, params, resource):
        return self._download_json(
            'https://api.lbry.tv/api/v1/proxy',
            display_id, 'Downloading %s JSON metadata' % resource,
            headers={'Content-Type': 'application/json-rpc'},
            data=json.dumps({
                'method': method,
                'params': params,
            }).encode())['result']

    def _resolve_url(self, url, display_id, resource):
        return self._call_api_proxy(
            'resolve', display_id, {'urls': url}, resource)[url]

    def _permanent_url(self, url, claim_name, claim_id):
        return urljoin(
            url.replace('lbry://', 'https://lbry.tv/'),
            '/%s:%s' % (claim_name, claim_id))

    def _parse_stream(self, stream, url):
        stream_value = stream.get('value') or {}
        stream_type = stream_value.get('stream_type')
        source = stream_value.get('source') or {}
        media = stream_value.get(stream_type) or {}
        signing_channel = stream.get('signing_channel') or {}
        channel_name = signing_channel.get('name')
        channel_claim_id = signing_channel.get('claim_id')
        channel_url = None
        if channel_name and channel_claim_id:
            channel_url = self._permanent_url(url, channel_name, channel_claim_id)

        info = {
            'thumbnail': try_get(stream_value, lambda x: x['thumbnail']['url'], compat_str),
            'description': stream_value.get('description'),
            'license': stream_value.get('license'),
            'timestamp': int_or_none(stream.get('timestamp')),
            'tags': stream_value.get('tags'),
            'duration': int_or_none(media.get('duration')),
            'channel': try_get(signing_channel, lambda x: x['value']['title']),
            'channel_id': channel_claim_id,
            'channel_url': channel_url,
            'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
            'filesize': int_or_none(source.get('size')),
        }
        if stream_type == 'audio':
            info['vcodec'] = 'none'
        else:
            info.update({
                'width': int_or_none(media.get('width')),
                'height': int_or_none(media.get('height')),
            })
        return info


class LBRYIE(LBRYBaseIE):
    IE_NAME = 'lbry'
    _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>\$/[^/]+/[^/]+/{1}|@{0}/{0}|(?!@){0})'.format(LBRYBaseIE._OPT_CLAIM_ID, LBRYBaseIE._CLAIM_ID_REGEX)
    _TESTS = [{
        # Video
        'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
        'md5': '65bd7ec1f6744ada55da8e4c48a2edf9',
        'info_dict': {
            'id': '17f983b61f53091fb8ea58a9c56804e4ff8cff4d',
            'ext': 'mp4',
            'title': 'First day in LBRY? Start HERE!',
            'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
            'timestamp': 1595694354,
            'upload_date': '20200725',
            'width': 1280,
            'height': 720,
        }
    }, {
        # Audio
        'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
        'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
        'info_dict': {
            'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
            'ext': 'mp3',
            'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
            'description': 'md5:661ac4f1db09f31728931d7b88807a61',
            'timestamp': 1591312601,
            'upload_date': '20200604',
            'tags': list,
            'duration': 2570,
            'channel': 'The LBRY Foundation',
            'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
            'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
            'vcodec': 'none',
        }
    }, {
        'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
        'only_matching': True,
    }, {
        'url': "https://odysee.com/@ScammerRevolts:b0/I-SYSKEY'D-THE-SAME-SCAMMERS-3-TIMES!:b",
        'only_matching': True,
    }, {
        'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
        'only_matching': True,
    }, {
        'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
        'only_matching': True,
    }, {
        'url': 'https://lbry.tv/Episode-1:e7',
        'only_matching': True,
    }, {
        'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
        'only_matching': True,
    }, {
        'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
        'only_matching': True,
    }, {
        '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',
        'only_matching': True,
    }, {
        'url': 'lbry://@lbry#3f/odysee#7',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        display_id = self._match_id(url)
        if display_id.startswith('$/'):
            display_id = display_id.split('/', 2)[-1].replace('/', ':')
        else:
            display_id = display_id.replace(':', '#')
        display_id = compat_urllib_parse_unquote(display_id)
        uri = 'lbry://' + display_id
        result = self._resolve_url(uri, display_id, 'stream')
        result_value = result['value']
        if result_value.get('stream_type') not in self._SUPPORTED_STREAM_TYPES:
            raise ExtractorError('Unsupported URL', expected=True)
        claim_id = result['claim_id']
        title = result_value['title']
        streaming_url = self._call_api_proxy(
            'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
        info = self._parse_stream(result, url)
        info.update({
            'id': claim_id,
            'title': title,
            'url': streaming_url,
        })
        return info


class LBRYChannelIE(LBRYBaseIE):
    IE_NAME = 'lbry:channel'
    _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?&]|$)' % LBRYBaseIE._OPT_CLAIM_ID
    _TESTS = [{
        'url': 'https://lbry.tv/@LBRYFoundation:0',
        'info_dict': {
            'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
            'title': 'The LBRY Foundation',
            'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
        },
        'playlist_count': 29,
    }, {
        'url': 'https://lbry.tv/@LBRYFoundation',
        'only_matching': True,
    }, {
        'url': 'lbry://@lbry#3f',
        'only_matching': True,
    }]
    _PAGE_SIZE = 50

    def _fetch_page(self, claim_id, url, page):
        page += 1
        result = self._call_api_proxy(
            'claim_search', claim_id, {
                'channel_ids': [claim_id],
                'claim_type': 'stream',
                'no_totals': True,
                'page': page,
                'page_size': self._PAGE_SIZE,
                'stream_types': self._SUPPORTED_STREAM_TYPES,
            }, 'page %d' % page)
        for item in (result.get('items') or []):
            stream_claim_name = item.get('name')
            stream_claim_id = item.get('claim_id')
            if not (stream_claim_name and stream_claim_id):
                continue

            info = self._parse_stream(item, url)
            info.update({
                '_type': 'url',
                'id': stream_claim_id,
                'title': try_get(item, lambda x: x['value']['title']),
                'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
            })
            yield info

    def _real_extract(self, url):
        display_id = self._match_id(url).replace(':', '#')
        result = self._resolve_url(
            'lbry://' + display_id, display_id, 'channel')
        claim_id = result['claim_id']
        entries = OnDemandPagedList(
            functools.partial(self._fetch_page, claim_id, url),
            self._PAGE_SIZE)
        result_value = result.get('value') or {}
        return self.playlist_result(
            entries, claim_id, result_value.get('title'),
            result_value.get('description'))
Commit	Line	Data
8bdd16b4	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
29f7c58a	4	import functools
8bdd16b4	5	import json
	6
	7	from .common import InfoExtractor
a820dc72 RA	8	from ..compat import (
	9	compat_str,
	10	compat_urllib_parse_unquote,
	11	)
8bdd16b4	12	from ..utils import (
	13	determine_ext,
	14	ExtractorError,
	15	int_or_none,
	16	mimetype2ext,
29f7c58a	17	OnDemandPagedList,
8bdd16b4	18	try_get,
29f7c58a	19	urljoin,
8bdd16b4	20	)
	21
	22
29f7c58a	23	class LBRYBaseIE(InfoExtractor):
57db6a87	24	_BASE_URL_REGEX = r'(?:https?://(?:www\.)?(?:lbry\.tv\|odysee\.com)/\|lbry://)'
29f7c58a	25	_CLAIM_ID_REGEX = r'[0-9a-f]{1,40}'
57db6a87	26	_OPT_CLAIM_ID = '[^:/?#&]+(?:[:#]%s)?' % _CLAIM_ID_REGEX
29f7c58a	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):
57db6a87	44	return urljoin(
	45	url.replace('lbry://', 'https://lbry.tv/'),
	46	'/%s:%s' % (claim_name, claim_id))
29f7c58a	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	'tags': stream_value.get('tags'),
	66	'duration': int_or_none(media.get('duration')),
	67	'channel': try_get(signing_channel, lambda x: x['value']['title']),
	68	'channel_id': channel_claim_id,
	69	'channel_url': channel_url,
	70	'ext': determine_ext(source.get('name')) or mimetype2ext(source.get('media_type')),
	71	'filesize': int_or_none(source.get('size')),
	72	}
	73	if stream_type == 'audio':
	74	info['vcodec'] = 'none'
	75	else:
	76	info.update({
	77	'width': int_or_none(media.get('width')),
	78	'height': int_or_none(media.get('height')),
	79	})
	80	return info
	81
	82
	83	class LBRYIE(LBRYBaseIE):
	84	IE_NAME = 'lbry'
	85	_VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>\$/[^/]+/[^/]+/{1}\|@{0}/{0}\|(?!@){0})'.format(LBRYBaseIE._OPT_CLAIM_ID, LBRYBaseIE._CLAIM_ID_REGEX)
8bdd16b4	86	_TESTS = [{
	87	# Video
	88	'url': 'https://lbry.tv/@Mantega:1/First-day-LBRY:1',
	89	'md5': '65bd7ec1f6744ada55da8e4c48a2edf9',
	90	'info_dict': {
	91	'id': '17f983b61f53091fb8ea58a9c56804e4ff8cff4d',
	92	'ext': 'mp4',
	93	'title': 'First day in LBRY? Start HERE!',
	94	'description': 'md5:f6cb5c704b332d37f5119313c2c98f51',
	95	'timestamp': 1595694354,
	96	'upload_date': '20200725',
29f7c58a	97	'width': 1280,
29f7c58a	98	'height': 720,
8bdd16b4	99	}
	100	}, {
	101	# Audio
	102	'url': 'https://lbry.tv/@LBRYFoundation:0/Episode-1:e',
	103	'md5': 'c94017d3eba9b49ce085a8fad6b98d00',
	104	'info_dict': {
	105	'id': 'e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
	106	'ext': 'mp3',
	107	'title': 'The LBRY Foundation Community Podcast Episode 1 - Introduction, Streaming on LBRY, Transcoding',
	108	'description': 'md5:661ac4f1db09f31728931d7b88807a61',
	109	'timestamp': 1591312601,
	110	'upload_date': '20200604',
29f7c58a	111	'tags': list,
	112	'duration': 2570,
	113	'channel': 'The LBRY Foundation',
	114	'channel_id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
	115	'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212',
	116	'vcodec': 'none',
8bdd16b4	117	}
	118	}, {
	119	'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e',
	120	'only_matching': True,
70c5802b	121	}, {
	122	'url': "https://odysee.com/@ScammerRevolts:b0/I-SYSKEY'D-THE-SAME-SCAMMERS-3-TIMES!:b",
	123	'only_matching': True,
29f7c58a	124	}, {
	125	'url': 'https://lbry.tv/Episode-1:e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
	126	'only_matching': True,
	127	}, {
	128	'url': 'https://lbry.tv/$/embed/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
	129	'only_matching': True,
	130	}, {
	131	'url': 'https://lbry.tv/Episode-1:e7',
	132	'only_matching': True,
	133	}, {
	134	'url': 'https://lbry.tv/@LBRYFoundation/Episode-1',
	135	'only_matching': True,
	136	}, {
	137	'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396',
	138	'only_matching': True,
a820dc72 RA	139	}, {
	140	'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',
	141	'only_matching': True,
57db6a87	142	}, {
	143	'url': 'lbry://@lbry#3f/odysee#7',
	144	'only_matching': True,
8bdd16b4	145	}]
8bdd16b4	146
8bdd16b4	147	def _real_extract(self, url):
29f7c58a	148	display_id = self._match_id(url)
	149	if display_id.startswith('$/'):
	150	display_id = display_id.split('/', 2)[-1].replace('/', ':')
	151	else:
	152	display_id = display_id.replace(':', '#')
a820dc72	153	display_id = compat_urllib_parse_unquote(display_id)
8bdd16b4	154	uri = 'lbry://' + display_id
29f7c58a	155	result = self._resolve_url(uri, display_id, 'stream')
8bdd16b4	156	result_value = result['value']
29f7c58a	157	if result_value.get('stream_type') not in self._SUPPORTED_STREAM_TYPES:
8bdd16b4	158	raise ExtractorError('Unsupported URL', expected=True)
29f7c58a	159	claim_id = result['claim_id']
29f7c58a	160	title = result_value['title']
8bdd16b4	161	streaming_url = self._call_api_proxy(
29f7c58a	162	'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url']
	163	info = self._parse_stream(result, url)
	164	info.update({
	165	'id': claim_id,
	166	'title': title,
8bdd16b4	167	'url': streaming_url,
29f7c58a	168	})
	169	return info
	170
	171
	172	class LBRYChannelIE(LBRYBaseIE):
	173	IE_NAME = 'lbry:channel'
57db6a87	174	_VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P<id>@%s)/?(?:[?&]\|$)' % LBRYBaseIE._OPT_CLAIM_ID
29f7c58a	175	_TESTS = [{
	176	'url': 'https://lbry.tv/@LBRYFoundation:0',
	177	'info_dict': {
	178	'id': '0ed629d2b9c601300cacf7eabe9da0be79010212',
	179	'title': 'The LBRY Foundation',
	180	'description': 'Channel for the LBRY Foundation. Follow for updates and news.',
	181	},
	182	'playlist_count': 29,
	183	}, {
	184	'url': 'https://lbry.tv/@LBRYFoundation',
	185	'only_matching': True,
57db6a87	186	}, {
	187	'url': 'lbry://@lbry#3f',
	188	'only_matching': True,
29f7c58a	189	}]
	190	_PAGE_SIZE = 50
	191
	192	def _fetch_page(self, claim_id, url, page):
	193	page += 1
	194	result = self._call_api_proxy(
	195	'claim_search', claim_id, {
	196	'channel_ids': [claim_id],
	197	'claim_type': 'stream',
	198	'no_totals': True,
	199	'page': page,
	200	'page_size': self._PAGE_SIZE,
	201	'stream_types': self._SUPPORTED_STREAM_TYPES,
	202	}, 'page %d' % page)
	203	for item in (result.get('items') or []):
	204	stream_claim_name = item.get('name')
	205	stream_claim_id = item.get('claim_id')
	206	if not (stream_claim_name and stream_claim_id):
	207	continue
	208
	209	info = self._parse_stream(item, url)
	210	info.update({
	211	'_type': 'url',
	212	'id': stream_claim_id,
	213	'title': try_get(item, lambda x: x['value']['title']),
	214	'url': self._permanent_url(url, stream_claim_name, stream_claim_id),
	215	})
	216	yield info
	217
	218	def _real_extract(self, url):
	219	display_id = self._match_id(url).replace(':', '#')
	220	result = self._resolve_url(
	221	'lbry://' + display_id, display_id, 'channel')
	222	claim_id = result['claim_id']
	223	entries = OnDemandPagedList(
	224	functools.partial(self._fetch_page, claim_id, url),
	225	self._PAGE_SIZE)
	226	result_value = result.get('value') or {}
	227	return self.playlist_result(
	228	entries, claim_id, result_value.get('title'),
	229	result_value.get('description'))