From c368dc98e06961a84b0751541ee2869c352253a8 Mon Sep 17 00:00:00 2001 From: Remita Amine Date: Tue, 8 Dec 2020 14:53:22 +0100 Subject: [PATCH] [lbry] add support for channel extraction(closes #25584) --- youtube_dl/extractor/extractors.py | 5 +- youtube_dl/extractor/lbry.py | 179 ++++++++++++++++++++++------- 2 files changed, 141 insertions(+), 43 deletions(-) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 86589a059..a94502eb3 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -534,7 +534,10 @@ from .laola1tv import ( EHFTVIE, ITTFIE, ) -from .lbry import LBRYIE +from .lbry import ( + LBRYIE, + LBRYChannelIE, +) from .lci import LCIIE from .lcp import ( LcpPlayIE, diff --git a/youtube_dl/extractor/lbry.py b/youtube_dl/extractor/lbry.py index 4fd50a60d..41cc245eb 100644 --- a/youtube_dl/extractor/lbry.py +++ b/youtube_dl/extractor/lbry.py @@ -1,6 +1,7 @@ # coding: utf-8 from __future__ import unicode_literals +import functools import json from .common import InfoExtractor @@ -10,15 +11,73 @@ from ..utils import ( ExtractorError, int_or_none, mimetype2ext, + OnDemandPagedList, try_get, urljoin, ) -class LBRYIE(InfoExtractor): - IE_NAME = 'lbry.tv' +class LBRYBaseIE(InfoExtractor): + _BASE_URL_REGEX = r'https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/' _CLAIM_ID_REGEX = r'[0-9a-f]{1,40}' - _VALID_URL = r'https?://(?:www\.)?(?:lbry\.tv|odysee\.com)/(?P@[^:]+:{0}/[^:]+:{0}|[^:]+:{0}|\$/embed/[^/]+/{0})'.format(_CLAIM_ID_REGEX) + _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, '/%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\$/[^/]+/[^/]+/{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', @@ -30,6 +89,8 @@ class LBRYIE(InfoExtractor): 'description': 'md5:f6cb5c704b332d37f5119313c2c98f51', 'timestamp': 1595694354, 'upload_date': '20200725', + 'width': 1280, + 'height': 720, } }, { # Audio @@ -47,6 +108,7 @@ class LBRYIE(InfoExtractor): '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', @@ -63,57 +125,90 @@ class LBRYIE(InfoExtractor): }, { 'url': 'https://lbry.tv/Episode-1:e7', 'only_matching': True, + }, { + 'url': 'https://lbry.tv/@LBRYFoundation/Episode-1', + 'only_matching': True, + }, { + 'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396', + 'only_matching': True, }] - def _call_api_proxy(self, method, display_id, params): - return self._download_json( - 'https://api.lbry.tv/api/v1/proxy', display_id, - headers={'Content-Type': 'application/json-rpc'}, - data=json.dumps({ - 'method': method, - 'params': params, - }).encode())['result'] - def _real_extract(self, url): display_id = self._match_id(url) - if display_id.startswith('$/embed/'): - display_id = display_id[8:].replace('/', ':') + if display_id.startswith('$/'): + display_id = display_id.split('/', 2)[-1].replace('/', ':') else: display_id = display_id.replace(':', '#') uri = 'lbry://' + display_id - result = self._call_api_proxy( - 'resolve', display_id, {'urls': [uri]})[uri] + result = self._resolve_url(uri, display_id, 'stream') result_value = result['value'] - if result_value.get('stream_type') not in ('video', 'audio'): + 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'] - source = result_value.get('source') or {} - media = result_value.get('video') or result_value.get('audio') or {} - signing_channel = result.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 = urljoin(url, '/%s:%s' % (channel_name, channel_claim_id)) - - return { + 'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url'] + info = self._parse_stream(result, url) + info.update({ 'id': claim_id, 'title': title, - 'thumbnail': try_get(result_value, lambda x: x['thumbnail']['url'], compat_str), - 'description': result_value.get('description'), - 'license': result_value.get('license'), - 'timestamp': int_or_none(result.get('timestamp')), - 'tags': result_value.get('tags'), - 'width': int_or_none(media.get('width')), - 'height': int_or_none(media.get('height')), - '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')), 'url': streaming_url, - } + }) + return info + + +class LBRYChannelIE(LBRYBaseIE): + IE_NAME = 'lbry:channel' + _VALID_URL = LBRYBaseIE._BASE_URL_REGEX + r'(?P@%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, + }] + _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'))