From 841b683804c6f706554bf7607e52072575358445 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 6 Sep 2020 10:45:34 +0700 Subject: [PATCH 01/10] [twitch] Rework extractors (closes #12297, closes #20414, closes #20604, closes #21811, closes #21812, closes #22979, closes #24263, closes #25010, closes #25553, closes #25606) * Switch to GraphQL. + Add support for collections. + Add support for clips and collections playlists. --- youtube_dl/extractor/extractors.py | 11 +- youtube_dl/extractor/twitch.py | 674 ++++++++++++++++++----------- 2 files changed, 429 insertions(+), 256 deletions(-) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index 4b3092028..9564465a0 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -1229,14 +1229,11 @@ from .twentymin import TwentyMinutenIE from .twentythreevideo import TwentyThreeVideoIE from .twitcasting import TwitCastingIE from .twitch import ( - TwitchVideoIE, - TwitchChapterIE, TwitchVodIE, - TwitchProfileIE, - TwitchAllVideosIE, - TwitchUploadsIE, - TwitchPastBroadcastsIE, - TwitchHighlightsIE, + TwitchCollectionIE, + TwitchVideosIE, + TwitchVideosClipsIE, + TwitchVideosCollectionsIE, TwitchStreamIE, TwitchClipsIE, ) diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index e211cd4c8..eadc48c6d 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -1,24 +1,26 @@ # coding: utf-8 from __future__ import unicode_literals +import collections import itertools -import re -import random import json +import random +import re from .common import InfoExtractor from ..compat import ( compat_kwargs, compat_parse_qs, compat_str, + compat_urlparse, compat_urllib_parse_urlencode, compat_urllib_parse_urlparse, ) from ..utils import ( clean_html, ExtractorError, + float_or_none, int_or_none, - orderedSet, parse_duration, parse_iso8601, qualities, @@ -150,120 +152,16 @@ class TwitchBaseIE(InfoExtractor): }) self._sort_formats(formats) + def _download_access_token(self, channel_name): + return self._call_api( + 'api/channels/%s/access_token' % channel_name, channel_name, + 'Downloading access token JSON') -class TwitchItemBaseIE(TwitchBaseIE): - def _download_info(self, item, item_id): - return self._extract_info(self._call_api( - 'kraken/videos/%s%s' % (item, item_id), item_id, - 'Downloading %s info JSON' % self._ITEM_TYPE)) - - def _extract_media(self, item_id): - info = self._download_info(self._ITEM_SHORTCUT, item_id) - response = self._call_api( - 'api/videos/%s%s' % (self._ITEM_SHORTCUT, item_id), item_id, - 'Downloading %s playlist JSON' % self._ITEM_TYPE) - entries = [] - chunks = response['chunks'] - qualities = list(chunks.keys()) - for num, fragment in enumerate(zip(*chunks.values()), start=1): - formats = [] - for fmt_num, fragment_fmt in enumerate(fragment): - format_id = qualities[fmt_num] - fmt = { - 'url': fragment_fmt['url'], - 'format_id': format_id, - 'quality': 1 if format_id == 'live' else 0, - } - m = re.search(r'^(?P\d+)[Pp]', format_id) - if m: - fmt['height'] = int(m.group('height')) - formats.append(fmt) - self._sort_formats(formats) - entry = dict(info) - entry['id'] = '%s_%d' % (entry['id'], num) - entry['title'] = '%s part %d' % (entry['title'], num) - entry['formats'] = formats - entries.append(entry) - return self.playlist_result(entries, info['id'], info['title']) - - def _extract_info(self, info): - status = info.get('status') - if status == 'recording': - is_live = True - elif status == 'recorded': - is_live = False - else: - is_live = None - _QUALITIES = ('small', 'medium', 'large') - quality_key = qualities(_QUALITIES) - thumbnails = [] - preview = info.get('preview') - if isinstance(preview, dict): - for thumbnail_id, thumbnail_url in preview.items(): - thumbnail_url = url_or_none(thumbnail_url) - if not thumbnail_url: - continue - if thumbnail_id not in _QUALITIES: - continue - thumbnails.append({ - 'url': thumbnail_url, - 'preference': quality_key(thumbnail_id), - }) - return { - 'id': info['_id'], - 'title': info.get('title') or 'Untitled Broadcast', - 'description': info.get('description'), - 'duration': int_or_none(info.get('length')), - 'thumbnails': thumbnails, - 'uploader': info.get('channel', {}).get('display_name'), - 'uploader_id': info.get('channel', {}).get('name'), - 'timestamp': parse_iso8601(info.get('recorded_at')), - 'view_count': int_or_none(info.get('views')), - 'is_live': is_live, - } - - def _real_extract(self, url): - return self._extract_media(self._match_id(url)) + def _extract_channel_id(self, token, channel_name): + return compat_str(self._parse_json(token, channel_name)['channel_id']) -class TwitchVideoIE(TwitchItemBaseIE): - IE_NAME = 'twitch:video' - _VALID_URL = r'%s/[^/]+/b/(?P\d+)' % TwitchBaseIE._VALID_URL_BASE - _ITEM_TYPE = 'video' - _ITEM_SHORTCUT = 'a' - - _TEST = { - 'url': 'http://www.twitch.tv/riotgames/b/577357806', - 'info_dict': { - 'id': 'a577357806', - 'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG', - }, - 'playlist_mincount': 12, - 'skip': 'HTTP Error 404: Not Found', - } - - -class TwitchChapterIE(TwitchItemBaseIE): - IE_NAME = 'twitch:chapter' - _VALID_URL = r'%s/[^/]+/c/(?P\d+)' % TwitchBaseIE._VALID_URL_BASE - _ITEM_TYPE = 'chapter' - _ITEM_SHORTCUT = 'c' - - _TESTS = [{ - 'url': 'http://www.twitch.tv/acracingleague/c/5285812', - 'info_dict': { - 'id': 'c5285812', - 'title': 'ACRL Off Season - Sports Cars @ Nordschleife', - }, - 'playlist_mincount': 3, - 'skip': 'HTTP Error 404: Not Found', - }, { - 'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361', - 'only_matching': True, - }] - - -class TwitchVodIE(TwitchItemBaseIE): +class TwitchVodIE(TwitchBaseIE): IE_NAME = 'twitch:vod' _VALID_URL = r'''(?x) https?:// @@ -332,17 +230,60 @@ class TwitchVodIE(TwitchItemBaseIE): 'only_matching': True, }] - def _real_extract(self, url): - item_id = self._match_id(url) + def _download_info(self, item_id): + return self._extract_info( + self._call_api( + 'kraken/videos/%s' % item_id, item_id, + 'Downloading video info JSON')) - info = self._download_info(self._ITEM_SHORTCUT, item_id) + @staticmethod + def _extract_info(info): + status = info.get('status') + if status == 'recording': + is_live = True + elif status == 'recorded': + is_live = False + else: + is_live = None + _QUALITIES = ('small', 'medium', 'large') + quality_key = qualities(_QUALITIES) + thumbnails = [] + preview = info.get('preview') + if isinstance(preview, dict): + for thumbnail_id, thumbnail_url in preview.items(): + thumbnail_url = url_or_none(thumbnail_url) + if not thumbnail_url: + continue + if thumbnail_id not in _QUALITIES: + continue + thumbnails.append({ + 'url': thumbnail_url, + 'preference': quality_key(thumbnail_id), + }) + return { + 'id': info['_id'], + 'title': info.get('title') or 'Untitled Broadcast', + 'description': info.get('description'), + 'duration': int_or_none(info.get('length')), + 'thumbnails': thumbnails, + 'uploader': info.get('channel', {}).get('display_name'), + 'uploader_id': info.get('channel', {}).get('name'), + 'timestamp': parse_iso8601(info.get('recorded_at')), + 'view_count': int_or_none(info.get('views')), + 'is_live': is_live, + } + + def _real_extract(self, url): + vod_id = self._match_id(url) + + info = self._download_info(vod_id) access_token = self._call_api( - 'api/vods/%s/access_token' % item_id, item_id, + 'api/vods/%s/access_token' % vod_id, vod_id, 'Downloading %s access token' % self._ITEM_TYPE) formats = self._extract_m3u8_formats( '%s/vod/%s.m3u8?%s' % ( - self._USHER_BASE, item_id, + self._USHER_BASE, vod_id, compat_urllib_parse_urlencode({ 'allow_source': 'true', 'allow_audio_only': 'true', @@ -352,7 +293,7 @@ class TwitchVodIE(TwitchItemBaseIE): 'nauth': access_token['token'], 'nauthsig': access_token['sig'], })), - item_id, 'mp4', entry_protocol='m3u8_native') + vod_id, 'mp4', entry_protocol='m3u8_native') self._prefer_source(formats) info['formats'] = formats @@ -366,7 +307,7 @@ class TwitchVodIE(TwitchItemBaseIE): info['subtitles'] = { 'rechat': [{ 'url': update_url_query( - 'https://api.twitch.tv/v5/videos/%s/comments' % item_id, { + 'https://api.twitch.tv/v5/videos/%s/comments' % vod_id, { 'client_id': self._CLIENT_ID, }), 'ext': 'json', @@ -376,164 +317,405 @@ class TwitchVodIE(TwitchItemBaseIE): return info -class TwitchPlaylistBaseIE(TwitchBaseIE): - _PLAYLIST_PATH = 'kraken/channels/%s/videos/?offset=%d&limit=%d' +def _make_video_result(node): + assert isinstance(node, dict) + video_id = node.get('id') + if not video_id: + return + return { + '_type': 'url_transparent', + 'ie_key': TwitchVodIE.ie_key(), + 'id': video_id, + 'url': 'https://www.twitch.tv/videos/%s' % video_id, + 'title': node.get('title'), + 'thumbnail': node.get('previewThumbnailURL'), + 'duration': float_or_none(node.get('lengthSeconds')), + 'view_count': int_or_none(node.get('viewCount')), + } + + +class TwitchGraphQLBaseIE(TwitchBaseIE): _PAGE_LIMIT = 100 - def _extract_playlist(self, channel_id): - info = self._call_api( - 'kraken/channels/%s' % channel_id, - channel_id, 'Downloading channel info JSON') - channel_name = info.get('display_name') or info.get('name') + def _download_gql(self, video_id, op, variables, sha256_hash, note, fatal=True): + return self._download_json( + 'https://gql.twitch.tv/gql', video_id, note, + data=json.dumps({ + 'operationName': op, + 'variables': variables, + 'extensions': { + 'persistedQuery': { + 'version': 1, + 'sha256Hash': sha256_hash, + } + } + }).encode(), + headers={ + 'Content-Type': 'text/plain;charset=UTF-8', + 'Client-ID': self._CLIENT_ID, + }, fatal=fatal) + + +class TwitchCollectionIE(TwitchGraphQLBaseIE): + _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/collections/(?P[^/]+)' + + _TESTS = [{ + 'url': 'https://www.twitch.tv/collections/wlDCoH0zEBZZbQ', + 'info_dict': { + 'id': 'wlDCoH0zEBZZbQ', + 'title': 'Overthrow Nook, capitalism for children', + }, + 'playlist_mincount': 13, + }] + + _OPERATION_NAME = 'CollectionSideBar' + _SHA256_HASH = '27111f1b382effad0b6def325caef1909c733fe6a4fbabf54f8d491ef2cf2f14' + + def _real_extract(self, url): + collection_id = self._match_id(url) + collection = self._download_gql( + collection_id, self._OPERATION_NAME, + {'collectionID': collection_id}, self._SHA256_HASH, + 'Downloading collection GraphQL')['data']['collection'] + title = collection.get('title') entries = [] + for edge in collection['items']['edges']: + if not isinstance(edge, dict): + continue + node = edge.get('node') + if not isinstance(node, dict): + continue + video = _make_video_result(node) + if video: + entries.append(video) + return self.playlist_result( + entries, playlist_id=collection_id, playlist_title=title) + + +class TwitchPlaylistBaseIE(TwitchGraphQLBaseIE): + def _entries(self, channel_name, *args): + cursor = None + variables_common = self._make_variables(channel_name, *args) + entries_key = '%ss' % self._ENTRY_KIND + for page_num in itertools.count(1): + variables = variables_common.copy() + variables['limit'] = self._PAGE_LIMIT + if cursor: + variables['cursor'] = cursor + page = self._download_gql( + channel_name, self._OPERATION_NAME, variables, + self._SHA256_HASH, + 'Downloading %ss GraphQL page %s' % (self._NODE_KIND, page_num), + fatal=False) + if not page: + break + edges = try_get( + page, lambda x: x['data']['user'][entries_key]['edges'], list) + if not edges: + break + for edge in edges: + if not isinstance(edge, dict): + continue + if edge.get('__typename') != self._EDGE_KIND: + continue + node = edge.get('node') + if not isinstance(node, dict): + continue + if node.get('__typename') != self._NODE_KIND: + continue + entry = self._extract_entry(node) + if entry: + cursor = edge.get('cursor') + yield entry + if not cursor or not isinstance(cursor, compat_str): + break + + # Deprecated kraken v5 API + def _entries_kraken(self, channel_name, broadcast_type, sort): + access_token = self._download_access_token(channel_name) + channel_id = self._extract_channel_id(access_token['token'], channel_name) offset = 0 - limit = self._PAGE_LIMIT - broken_paging_detected = False counter_override = None for counter in itertools.count(1): response = self._call_api( - self._PLAYLIST_PATH % (channel_id, offset, limit), + 'kraken/channels/%s/videos/' % channel_id, channel_id, - 'Downloading %s JSON page %s' - % (self._PLAYLIST_TYPE, counter_override or counter)) - page_entries = self._extract_playlist_page(response) - if not page_entries: + 'Downloading video JSON page %s' % (counter_override or counter), + query={ + 'offset': offset, + 'limit': self._PAGE_LIMIT, + 'broadcast_type': broadcast_type, + 'sort': sort, + }) + videos = response.get('videos') + if not isinstance(videos, list): break + for video in videos: + if not isinstance(video, dict): + continue + video_url = url_or_none(video.get('url')) + if not video_url: + continue + yield { + '_type': 'url_transparent', + 'ie_key': TwitchVodIE.ie_key(), + 'id': video.get('_id'), + 'url': video_url, + 'title': video.get('title'), + 'description': video.get('description'), + 'timestamp': unified_timestamp(video.get('published_at')), + 'duration': float_or_none(video.get('length')), + 'view_count': int_or_none(video.get('views')), + 'language': video.get('language'), + } + offset += self._PAGE_LIMIT total = int_or_none(response.get('_total')) - # Since the beginning of March 2016 twitch's paging mechanism - # is completely broken on the twitch side. It simply ignores - # a limit and returns the whole offset number of videos. - # Working around by just requesting all videos at once. - # Upd: pagination bug was fixed by twitch on 15.03.2016. - if not broken_paging_detected and total and len(page_entries) > limit: - self.report_warning( - 'Twitch pagination is broken on twitch side, requesting all videos at once', - channel_id) - broken_paging_detected = True - offset = total - counter_override = '(all at once)' - continue - entries.extend(page_entries) - if broken_paging_detected or total and len(page_entries) >= total: + if total and offset >= total: break - offset += limit - return self.playlist_result( - [self._make_url_result(entry) for entry in orderedSet(entries)], - channel_id, channel_name) - - def _make_url_result(self, url): - try: - video_id = 'v%s' % TwitchVodIE._match_id(url) - return self.url_result(url, TwitchVodIE.ie_key(), video_id=video_id) - except AssertionError: - return self.url_result(url) - - def _extract_playlist_page(self, response): - videos = response.get('videos') - return [video['url'] for video in videos] if videos else [] - - def _real_extract(self, url): - return self._extract_playlist(self._match_id(url)) -class TwitchProfileIE(TwitchPlaylistBaseIE): - IE_NAME = 'twitch:profile' - _VALID_URL = r'%s/(?P[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE - _PLAYLIST_TYPE = 'profile' +class TwitchVideosIE(TwitchPlaylistBaseIE): + _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P[^/]+)/(?:videos|profile)' _TESTS = [{ - 'url': 'http://www.twitch.tv/vanillatv/profile', - 'info_dict': { - 'id': 'vanillatv', - 'title': 'VanillaTV', - }, - 'playlist_mincount': 412, - }, { - 'url': 'http://m.twitch.tv/vanillatv/profile', - 'only_matching': True, - }] - - -class TwitchVideosBaseIE(TwitchPlaylistBaseIE): - _VALID_URL_VIDEOS_BASE = r'%s/(?P[^/]+)/videos' % TwitchBaseIE._VALID_URL_BASE - _PLAYLIST_PATH = TwitchPlaylistBaseIE._PLAYLIST_PATH + '&broadcast_type=' - - -class TwitchAllVideosIE(TwitchVideosBaseIE): - IE_NAME = 'twitch:videos:all' - _VALID_URL = r'%s/all' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE - _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive,upload,highlight' - _PLAYLIST_TYPE = 'all videos' - - _TESTS = [{ - 'url': 'https://www.twitch.tv/spamfish/videos/all', + # All Videos sorted by Date + 'url': 'https://www.twitch.tv/spamfish/videos?filter=all', 'info_dict': { 'id': 'spamfish', - 'title': 'Spamfish', + 'title': 'spamfish - All Videos sorted by Date', }, - 'playlist_mincount': 869, + 'playlist_mincount': 924, + }, { + # All Videos sorted by Popular + 'url': 'https://www.twitch.tv/spamfish/videos?filter=all&sort=views', + 'info_dict': { + 'id': 'spamfish', + 'title': 'spamfish - All Videos sorted by Popular', + }, + 'playlist_mincount': 931, + }, { + # Past Broadcasts sorted by Date + 'url': 'https://www.twitch.tv/spamfish/videos?filter=archives', + 'info_dict': { + 'id': 'spamfish', + 'title': 'spamfish - Past Broadcasts sorted by Date', + }, + 'playlist_mincount': 27, + }, { + # Highlights sorted by Date + 'url': 'https://www.twitch.tv/spamfish/videos?filter=highlights', + 'info_dict': { + 'id': 'spamfish', + 'title': 'spamfish - Highlights sorted by Date', + }, + 'playlist_mincount': 901, + }, { + # Uploads sorted by Date + 'url': 'https://www.twitch.tv/esl_csgo/videos?filter=uploads&sort=time', + 'info_dict': { + 'id': 'esl_csgo', + 'title': 'esl_csgo - Uploads sorted by Date', + }, + 'playlist_mincount': 5, + }, { + # Past Premieres sorted by Date + 'url': 'https://www.twitch.tv/spamfish/videos?filter=past_premieres', + 'info_dict': { + 'id': 'spamfish', + 'title': 'spamfish - Past Premieres sorted by Date', + }, + 'playlist_mincount': 1, + }, { + 'url': 'https://www.twitch.tv/spamfish/videos/all', + 'only_matching': True, }, { 'url': 'https://m.twitch.tv/spamfish/videos/all', 'only_matching': True, - }] - - -class TwitchUploadsIE(TwitchVideosBaseIE): - IE_NAME = 'twitch:videos:uploads' - _VALID_URL = r'%s/uploads' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE - _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'upload' - _PLAYLIST_TYPE = 'uploads' - - _TESTS = [{ - 'url': 'https://www.twitch.tv/spamfish/videos/uploads', - 'info_dict': { - 'id': 'spamfish', - 'title': 'Spamfish', - }, - 'playlist_mincount': 0, }, { - 'url': 'https://m.twitch.tv/spamfish/videos/uploads', + 'url': 'https://www.twitch.tv/spamfish/videos', 'only_matching': True, }] + Broadcast = collections.namedtuple('Broadcast', ['type', 'label']) -class TwitchPastBroadcastsIE(TwitchVideosBaseIE): - IE_NAME = 'twitch:videos:past-broadcasts' - _VALID_URL = r'%s/past-broadcasts' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE - _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive' - _PLAYLIST_TYPE = 'past broadcasts' + _DEFAULT_BROADCAST = Broadcast(None, 'All Videos') + _BROADCASTS = { + 'archives': Broadcast('ARCHIVE', 'Past Broadcasts'), + 'highlights': Broadcast('HIGHLIGHT', 'Highlights'), + 'uploads': Broadcast('UPLOAD', 'Uploads'), + 'past_premieres': Broadcast('PAST_PREMIERE', 'Past Premieres'), + 'all': _DEFAULT_BROADCAST, + } + + _DEFAULT_SORTED_BY = 'Date' + _SORTED_BY = { + 'time': _DEFAULT_SORTED_BY, + 'views': 'Popular', + } + + _SHA256_HASH = 'a937f1d22e269e39a03b509f65a7490f9fc247d7f83d6ac1421523e3b68042cb' + _OPERATION_NAME = 'FilterableVideoTower_Videos' + _ENTRY_KIND = 'video' + _EDGE_KIND = 'VideoEdge' + _NODE_KIND = 'Video' + + @classmethod + def suitable(cls, url): + return (False + if any(ie.suitable(url) for ie in ( + TwitchVideosClipsIE, + TwitchVideosCollectionsIE)) + else super(TwitchVideosIE, cls).suitable(url)) + + @staticmethod + def _make_variables(channel_name, broadcast_type, sort): + return { + 'channelOwnerLogin': channel_name, + 'broadcastType': broadcast_type, + 'videoSort': sort.upper(), + } + + @staticmethod + def _extract_entry(node): + return _make_video_result(node) + + def _real_extract(self, url): + channel_name = self._match_id(url) + qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + filter = qs.get('filter', ['all'])[0] + sort = qs.get('sort', ['time'])[0] + broadcast = self._BROADCASTS.get(filter, self._DEFAULT_BROADCAST) + return self.playlist_result( + self._entries(channel_name, broadcast.type, sort), + playlist_id=channel_name, + playlist_title='%s - %s sorted by %s' + % (channel_name, broadcast.label, + self._SORTED_BY.get(sort, self._DEFAULT_SORTED_BY))) + + +class TwitchVideosClipsIE(TwitchPlaylistBaseIE): + _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P[^/]+)/(?:clips|videos/*?\?.*?\bfilter=clips)' _TESTS = [{ - 'url': 'https://www.twitch.tv/spamfish/videos/past-broadcasts', + # Clips + 'url': 'https://www.twitch.tv/vanillatv/clips?filter=clips&range=all', 'info_dict': { - 'id': 'spamfish', - 'title': 'Spamfish', + 'id': 'vanillatv', + 'title': 'vanillatv - Clips Top All', }, - 'playlist_mincount': 0, + 'playlist_mincount': 1, }, { - 'url': 'https://m.twitch.tv/spamfish/videos/past-broadcasts', + 'url': 'https://www.twitch.tv/dota2ruhub/videos?filter=clips&range=7d', 'only_matching': True, }] + Clip = collections.namedtuple('Clip', ['filter', 'label']) -class TwitchHighlightsIE(TwitchVideosBaseIE): - IE_NAME = 'twitch:videos:highlights' - _VALID_URL = r'%s/highlights' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE - _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'highlight' - _PLAYLIST_TYPE = 'highlights' + _DEFAULT_CLIP = Clip('LAST_WEEK', 'Top 7D') + _RANGE = { + '24hr': Clip('LAST_DAY', 'Top 24H'), + '7d': _DEFAULT_CLIP, + '30d': Clip('LAST_MONTH', 'Top 30D'), + 'all': Clip('ALL_TIME', 'Top All'), + } + + # NB: values other than 20 result in skipped videos + _PAGE_LIMIT = 20 + + _SHA256_HASH = 'b73ad2bfaecfd30a9e6c28fada15bd97032c83ec77a0440766a56fe0bd632777' + _OPERATION_NAME = 'ClipsCards__User' + _ENTRY_KIND = 'clip' + _EDGE_KIND = 'ClipEdge' + _NODE_KIND = 'Clip' + + @staticmethod + def _make_variables(channel_name, filter): + return { + 'login': channel_name, + 'criteria': { + 'filter': filter, + }, + } + + @staticmethod + def _extract_entry(node): + assert isinstance(node, dict) + clip_url = url_or_none(node.get('url')) + if not clip_url: + return + return { + '_type': 'url_transparent', + 'ie_key': TwitchClipsIE.ie_key(), + 'id': node.get('id'), + 'url': clip_url, + 'title': node.get('title'), + 'thumbnail': node.get('thumbnailURL'), + 'duration': float_or_none(node.get('durationSeconds')), + 'timestamp': unified_timestamp(node.get('createdAt')), + 'view_count': int_or_none(node.get('viewCount')), + 'language': node.get('language'), + } + + def _real_extract(self, url): + channel_name = self._match_id(url) + qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + range = qs.get('range', ['7d'])[0] + clip = self._RANGE.get(range, self._DEFAULT_CLIP) + return self.playlist_result( + self._entries(channel_name, clip.filter), + playlist_id=channel_name, + playlist_title='%s - Clips %s' % (channel_name, clip.label)) + + +class TwitchVideosCollectionsIE(TwitchPlaylistBaseIE): + _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P[^/]+)/videos/*?\?.*?\bfilter=collections' _TESTS = [{ - 'url': 'https://www.twitch.tv/spamfish/videos/highlights', + # Collections + 'url': 'https://www.twitch.tv/spamfish/videos?filter=collections', 'info_dict': { 'id': 'spamfish', - 'title': 'Spamfish', + 'title': 'spamfish - Collections', }, - 'playlist_mincount': 805, - }, { - 'url': 'https://m.twitch.tv/spamfish/videos/highlights', - 'only_matching': True, + 'playlist_mincount': 3, }] + _SHA256_HASH = '07e3691a1bad77a36aba590c351180439a40baefc1c275356f40fc7082419a84' + _OPERATION_NAME = 'ChannelCollectionsContent' + _ENTRY_KIND = 'collection' + _EDGE_KIND = 'CollectionsItemEdge' + _NODE_KIND = 'Collection' + + @staticmethod + def _make_variables(channel_name): + return { + 'ownerLogin': channel_name, + } + + @staticmethod + def _extract_entry(node): + assert isinstance(node, dict) + collection_id = node.get('id') + if not collection_id: + return + return { + '_type': 'url_transparent', + 'ie_key': TwitchCollectionIE.ie_key(), + 'id': collection_id, + 'url': 'https://www.twitch.tv/collections/%s' % collection_id, + 'title': node.get('title'), + 'thumbnail': node.get('thumbnailURL'), + 'duration': float_or_none(node.get('lengthSeconds')), + 'timestamp': unified_timestamp(node.get('updatedAt')), + 'view_count': int_or_none(node.get('viewCount')), + } + + def _real_extract(self, url): + channel_name = self._match_id(url) + return self.playlist_result( + self._entries(channel_name), playlist_id=channel_name, + playlist_title='%s - Collections' % channel_name) + class TwitchStreamIE(TwitchBaseIE): IE_NAME = 'twitch:stream' @@ -583,27 +765,21 @@ class TwitchStreamIE(TwitchBaseIE): def suitable(cls, url): return (False if any(ie.suitable(url) for ie in ( - TwitchVideoIE, - TwitchChapterIE, TwitchVodIE, - TwitchProfileIE, - TwitchAllVideosIE, - TwitchUploadsIE, - TwitchPastBroadcastsIE, - TwitchHighlightsIE, + TwitchCollectionIE, + TwitchVideosIE, + TwitchVideosClipsIE, + TwitchVideosCollectionsIE, TwitchClipsIE)) else super(TwitchStreamIE, cls).suitable(url)) def _real_extract(self, url): channel_name = self._match_id(url) - access_token = self._call_api( - 'api/channels/%s/access_token' % channel_name, channel_name, - 'Downloading access token JSON') + access_token = self._download_access_token(channel_name) token = access_token['token'] - channel_id = compat_str(self._parse_json( - token, channel_name)['channel_id']) + channel_id = self._extract_channel_id(token, channel_name) stream = self._call_api( 'kraken/streams/%s?stream_type=all' % channel_id, From 5ed05f26adea517aa715d5ec4b0dccfea30b2b8e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 6 Sep 2020 10:45:57 +0700 Subject: [PATCH 02/10] [svtplay] Fix svt id extraction (closes #26425, closes #26428, closes #26438) --- youtube_dl/extractor/svt.py | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py index e12389cad..8e9ec2ca3 100644 --- a/youtube_dl/extractor/svt.py +++ b/youtube_dl/extractor/svt.py @@ -224,9 +224,15 @@ class SVTPlayIE(SVTPlayBaseIE): self._adjust_title(info_dict) return info_dict - svt_id = self._search_regex( - r']+data-video-id=["\']([\da-zA-Z-]+)', - webpage, 'video id') + svt_id = try_get( + data, lambda x: x['statistics']['dataLake']['content']['id'], + compat_str) + + if not svt_id: + svt_id = self._search_regex( + (r']+data-video-id=["\']([\da-zA-Z-]+)', + r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"'), + webpage, 'video id') return self._extract_by_video_id(svt_id, webpage) From 62ae19ff760b9df6a2600430e42e2c32f1449e7a Mon Sep 17 00:00:00 2001 From: TheRealDude2 Date: Sun, 6 Sep 2020 06:10:27 +0200 Subject: [PATCH 03/10] [xhamster] Improve initials regex (#26526) (closes #26353) --- youtube_dl/extractor/xhamster.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index 902a3ed33..76aeaf9a4 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -138,7 +138,8 @@ class XHamsterIE(InfoExtractor): initials = self._parse_json( self._search_regex( - r'window\.initials\s*=\s*({.+?})\s*;', webpage, 'initials', + (r'window\.initials\s*=\s*({.+?})\s*;\s*', + r'window\.initials\s*=\s*({.+?})\s*;'), webpage, 'initials', default='{}'), video_id, fatal=False) if initials: From 1d9bf655e68c873dc1f04739640046c43a1443d5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 6 Sep 2020 11:19:53 +0700 Subject: [PATCH 04/10] [utils] Recognize wav mimetype (closes #26463) --- test/test_utils.py | 2 ++ youtube_dl/utils.py | 1 + 2 files changed, 3 insertions(+) diff --git a/test/test_utils.py b/test/test_utils.py index 0896f4150..962fd8d75 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -803,6 +803,8 @@ class TestUtil(unittest.TestCase): self.assertEqual(mimetype2ext('text/vtt'), 'vtt') self.assertEqual(mimetype2ext('text/vtt;charset=utf-8'), 'vtt') self.assertEqual(mimetype2ext('text/html; charset=utf-8'), 'html') + self.assertEqual(mimetype2ext('audio/x-wav'), 'wav') + self.assertEqual(mimetype2ext('audio/x-wav;codec=pcm'), 'wav') def test_month_by_name(self): self.assertEqual(month_by_name(None), None) diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index d1eca3760..01d9c0362 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -4198,6 +4198,7 @@ def mimetype2ext(mt): 'vnd.ms-sstr+xml': 'ism', 'quicktime': 'mov', 'mp2t': 'ts', + 'x-wav': 'wav', }.get(res, res) From 67171ed7e940d9022e3388ddd029978bc426ea1f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 6 Sep 2020 11:31:28 +0700 Subject: [PATCH 05/10] [youtube:user] Extend _VALID_URL (closes #26443) --- youtube_dl/extractor/youtube.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index b35bf03aa..6611caf06 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -3008,7 +3008,7 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor): class YoutubeUserIE(YoutubeChannelIE): IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' - _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?Puser|c)/)?(?!(?:attribution_link|watch|results|shared)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P[A-Za-z0-9_-]+)' + _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?Puser|c)/)?(?!(?:attribution_link|watch|results|shared)(?:$|[^a-z_A-Z0-9%-])))|ytuser:)(?!feed/)(?P[A-Za-z0-9_%-]+)' _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos' IE_NAME = 'youtube:user' @@ -3038,6 +3038,9 @@ class YoutubeUserIE(YoutubeChannelIE): }, { 'url': 'https://www.youtube.com/c/gametrailers', 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/c/Pawe%C5%82Zadro%C5%BCniak', + 'only_matching': True, }, { 'url': 'https://www.youtube.com/gametrailers', 'only_matching': True, From 16ee69c1b7d9877c852d50428d8f047ada45d539 Mon Sep 17 00:00:00 2001 From: random-nick Date: Sun, 6 Sep 2020 04:44:53 +0000 Subject: [PATCH 06/10] [youtube] Fix age gate content detection (#26100) (closes #26152, closes #26311, closes #26384) --- youtube_dl/extractor/youtube.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 6611caf06..6ae2e58c1 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1825,7 +1825,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Get video info video_info = {} embed_webpage = None - if re.search(r'player-age-gate-content">', video_webpage) is not None: + if (self._og_search_property('restrictions:age', video_webpage, default=None) == '18+' + or re.search(r'player-age-gate-content">', video_webpage) is not None): age_gate = True # We simulate the access to the video from www.youtube.com/v/{video_id} # this can be viewed without login into Youtube From 50e9fcc1fd5aa8e3c8bbeb33f2434fbf9c1200d7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 6 Sep 2020 12:42:57 +0700 Subject: [PATCH 07/10] [nrktv:episode] Improve video id extraction (closes #25594, closes #26369, closes #26409) --- youtube_dl/extractor/nrk.py | 20 +++++++++++++------- 1 file changed, 13 insertions(+), 7 deletions(-) diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index 94115534b..84aacbcda 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -11,7 +11,6 @@ from ..compat import ( from ..utils import ( ExtractorError, int_or_none, - JSON_LD_RE, js_to_json, NO_DEFAULT, parse_age_limit, @@ -425,13 +424,20 @@ class NRKTVEpisodeIE(InfoExtractor): webpage = self._download_webpage(url, display_id) - nrk_id = self._parse_json( - self._search_regex(JSON_LD_RE, webpage, 'JSON-LD', group='json_ld'), - display_id)['@id'] - + info = self._search_json_ld(webpage, display_id, default={}) + nrk_id = info.get('@id') or self._html_search_meta( + 'nrk:program-id', webpage, default=None) or self._search_regex( + r'data-program-id=["\'](%s)' % NRKTVIE._EPISODE_RE, webpage, + 'nrk id') assert re.match(NRKTVIE._EPISODE_RE, nrk_id) - return self.url_result( - 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id) + + info.update({ + '_type': 'url_transparent', + 'id': nrk_id, + 'url': 'nrk:%s' % nrk_id, + 'ie_key': NRKIE.ie_key(), + }) + return info class NRKTVSerieBaseIE(InfoExtractor): From 6cd452acffe8d79c895a2ebd0346e2ba7f9e112f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 6 Sep 2020 12:57:56 +0700 Subject: [PATCH 08/10] [ChangeLog] Actualize [ci skip] --- ChangeLog | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/ChangeLog b/ChangeLog index bf515f784..8d410f600 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,24 @@ +version + +Core ++ [utils] Recognize wav mimetype (#26463) + +Extractors +* [nrktv:episode] Improve video id extraction (#25594, #26369, #26409) +* [youtube] Fix age gate content detection (#26100, #26152, #26311, #26384) +* [youtube:user] Extend URL regular expression (#26443) +* [xhamster] Improve initials regular expression (#26526, #26353) +* [svtplay] Fix video id extraction (#26425, #26428, #26438) +* [twitch] Rework extractors (#12297, #20414, #20604, #21811, #21812, #22979, + #24263, #25010, #25553, #25606) + * Switch to GraphQL + + Add support for collections + + Add support for clips and collections playlists +* [biqle] Improve video ext extraction +* [xhamster] Fix extraction (#26157, #26254) +* [xhamster] Extend URL regular expression (#25789, #25804, #25927)) + + version 2020.07.28 Extractors From d51e23d9fc709a80ca037199a46e59c729a1192e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sun, 6 Sep 2020 13:00:41 +0700 Subject: [PATCH 09/10] release 2020.09.06 --- .github/ISSUE_TEMPLATE/1_broken_site.md | 6 +++--- .github/ISSUE_TEMPLATE/2_site_support_request.md | 4 ++-- .github/ISSUE_TEMPLATE/3_site_feature_request.md | 4 ++-- .github/ISSUE_TEMPLATE/4_bug_report.md | 6 +++--- .github/ISSUE_TEMPLATE/5_feature_request.md | 4 ++-- ChangeLog | 2 +- docs/supportedsites.md | 11 ++++------- youtube_dl/version.py | 2 +- 8 files changed, 18 insertions(+), 21 deletions(-) diff --git a/.github/ISSUE_TEMPLATE/1_broken_site.md b/.github/ISSUE_TEMPLATE/1_broken_site.md index f2260db46..f05aa66e6 100644 --- a/.github/ISSUE_TEMPLATE/1_broken_site.md +++ b/.github/ISSUE_TEMPLATE/1_broken_site.md @@ -18,7 +18,7 @@ title: '' - [ ] I'm reporting a broken site support -- [ ] I've verified that I'm running youtube-dl version **2020.07.28** +- [ ] I've verified that I'm running youtube-dl version **2020.09.06** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar issues including closed ones @@ -41,7 +41,7 @@ Add the `-v` flag to your command line you run youtube-dl with (`youtube-dl -v < [debug] User config: [] [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj'] [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251 - [debug] youtube-dl version 2020.07.28 + [debug] youtube-dl version 2020.09.06 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4 [debug] Proxy map: {} diff --git a/.github/ISSUE_TEMPLATE/2_site_support_request.md b/.github/ISSUE_TEMPLATE/2_site_support_request.md index 8bc05c4ba..29beaf437 100644 --- a/.github/ISSUE_TEMPLATE/2_site_support_request.md +++ b/.github/ISSUE_TEMPLATE/2_site_support_request.md @@ -19,7 +19,7 @@ labels: 'site-support-request' - [ ] I'm reporting a new site support request -- [ ] I've verified that I'm running youtube-dl version **2020.07.28** +- [ ] I've verified that I'm running youtube-dl version **2020.09.06** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that none of provided URLs violate any copyrights - [ ] I've searched the bugtracker for similar site support requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/3_site_feature_request.md b/.github/ISSUE_TEMPLATE/3_site_feature_request.md index 98348e0cd..f96b8d2bb 100644 --- a/.github/ISSUE_TEMPLATE/3_site_feature_request.md +++ b/.github/ISSUE_TEMPLATE/3_site_feature_request.md @@ -18,13 +18,13 @@ title: '' - [ ] I'm reporting a site feature request -- [ ] I've verified that I'm running youtube-dl version **2020.07.28** +- [ ] I've verified that I'm running youtube-dl version **2020.09.06** - [ ] I've searched the bugtracker for similar site feature requests including closed ones diff --git a/.github/ISSUE_TEMPLATE/4_bug_report.md b/.github/ISSUE_TEMPLATE/4_bug_report.md index 86706f528..3a175aa4d 100644 --- a/.github/ISSUE_TEMPLATE/4_bug_report.md +++ b/.github/ISSUE_TEMPLATE/4_bug_report.md @@ -18,7 +18,7 @@ title: '' - [ ] I'm reporting a broken site support issue -- [ ] I've verified that I'm running youtube-dl version **2020.07.28** +- [ ] I've verified that I'm running youtube-dl version **2020.09.06** - [ ] I've checked that all provided URLs are alive and playable in a browser - [ ] I've checked that all URLs and arguments with special characters are properly quoted or escaped - [ ] I've searched the bugtracker for similar bug reports including closed ones @@ -43,7 +43,7 @@ Add the `-v` flag to your command line you run youtube-dl with (`youtube-dl -v < [debug] User config: [] [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj'] [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251 - [debug] youtube-dl version 2020.07.28 + [debug] youtube-dl version 2020.09.06 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4 [debug] Proxy map: {} diff --git a/.github/ISSUE_TEMPLATE/5_feature_request.md b/.github/ISSUE_TEMPLATE/5_feature_request.md index 52c2709f9..4977079de 100644 --- a/.github/ISSUE_TEMPLATE/5_feature_request.md +++ b/.github/ISSUE_TEMPLATE/5_feature_request.md @@ -19,13 +19,13 @@ labels: 'request' - [ ] I'm reporting a feature request -- [ ] I've verified that I'm running youtube-dl version **2020.07.28** +- [ ] I've verified that I'm running youtube-dl version **2020.09.06** - [ ] I've searched the bugtracker for similar feature requests including closed ones diff --git a/ChangeLog b/ChangeLog index 8d410f600..86b0e8ccb 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,4 +1,4 @@ -version +version 2020.09.06 Core + [utils] Recognize wav mimetype (#26463) diff --git a/docs/supportedsites.md b/docs/supportedsites.md index 35c1050e5..5c4e1d58c 100644 --- a/docs/supportedsites.md +++ b/docs/supportedsites.md @@ -950,16 +950,13 @@ - **TVPlayHome** - **Tweakers** - **TwitCasting** - - **twitch:chapter** - **twitch:clips** - - **twitch:profile** - **twitch:stream** - - **twitch:video** - - **twitch:videos:all** - - **twitch:videos:highlights** - - **twitch:videos:past-broadcasts** - - **twitch:videos:uploads** - **twitch:vod** + - **TwitchCollection** + - **TwitchVideos** + - **TwitchVideosClips** + - **TwitchVideosCollections** - **twitter** - **twitter:amplify** - **twitter:broadcast** diff --git a/youtube_dl/version.py b/youtube_dl/version.py index 17101fa47..45b4d3291 100644 --- a/youtube_dl/version.py +++ b/youtube_dl/version.py @@ -1,3 +1,3 @@ from __future__ import unicode_literals -__version__ = '2020.07.28' +__version__ = '2020.09.06' From d53fec697ea4afad261a8a1832bd9be205f9aecd Mon Sep 17 00:00:00 2001 From: nixxo Date: Mon, 7 Sep 2020 14:21:21 +0200 Subject: [PATCH 10/10] [mailru] fixed extraction improved valid_url regex to match more urls like >https://my.mail.ru/mail/7bw7/video/embed/_myvideo/6778 >https://videoapi.my.mail.ru/videos/embed/mail/7bw7/_myvideo/6778.html previously not matched. Fixed json extraction to download metadata and video url. --- youtube_dl/extractor/mailru.py | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/youtube_dl/extractor/mailru.py b/youtube_dl/extractor/mailru.py index 65cc474db..6fdf70aa6 100644 --- a/youtube_dl/extractor/mailru.py +++ b/youtube_dl/extractor/mailru.py @@ -20,10 +20,10 @@ class MailRuIE(InfoExtractor): IE_DESC = 'Видео@Mail.Ru' _VALID_URL = r'''(?x) https?:// - (?:(?:www|m)\.)?my\.mail\.ru/+ + (?:(?:www|m|videoapi)\.)?my\.mail\.ru/+ (?: video/.*\#video=/?(?P(?:[^/]+/){3}\d+)| - (?:(?P(?:[^/]+/+){2})video/(?P[^/]+/\d+))\.html| + (?:videos/embed/)?(?:(?P(?:[^/]+/+){2})(?:video/(?:embed/)?)?(?P[^/]+/\d+))(?:\.html)?| (?:video/embed|\+/video/meta)/(?P\d+) ) ''' @@ -108,15 +108,21 @@ class MailRuIE(InfoExtractor): if not video_id: video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix') webpage = self._download_webpage(url, video_id) - page_config = self._parse_json(self._search_regex( + page_config = self._parse_json(self._search_regex([ r'(?s)]+class="sp-video__page-config"[^>]*>(.+?)', + r'(?s)"video":\s*(\{.+?\}),'], webpage, 'page config', default='{}'), video_id, fatal=False) if page_config: - meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl') + meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl') or page_config.get('metadataUrl') else: meta_url = None video_data = None + + # fix meta_url if missing the host address + if re.match(r'^\/\+\/', meta_url): + meta_url = 'https://my.mail.ru' + meta_url + if meta_url: video_data = self._download_json( meta_url, video_id or meta_id, 'Downloading video meta JSON',