2 from __future__ import unicode_literals
10 from .common import InfoExtractor
11 from ..compat import (
16 compat_urllib_parse_urlencode,
17 compat_urllib_parse_urlparse,
35 class TwitchBaseIE(InfoExtractor):
36 _VALID_URL_BASE = r'https?://(?:(?:www|go|m)\.)?twitch\.tv'
38 _API_BASE = 'https://api.twitch.tv'
39 _USHER_BASE = 'https://usher.ttvnw.net'
40 _LOGIN_FORM_URL = 'https://www.twitch.tv/login'
41 _LOGIN_POST_URL = 'https://passport.twitch.tv/login'
42 _CLIENT_ID = 'kimne78kx3ncx6brgo4mv6wki5h1ko'
43 _NETRC_MACHINE = 'twitch'
45 def _handle_error(self, response):
46 if not isinstance(response, dict):
48 error = response.get('error')
51 '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
54 def _call_api(self, path, item_id, *args, **kwargs):
55 headers = kwargs.get('headers', {}).copy()
57 'Accept': 'application/vnd.twitchtv.v5+json; charset=UTF-8',
58 'Client-ID': self._CLIENT_ID,
62 'expected_status': (400, 410),
64 response = self._download_json(
65 '%s/%s' % (self._API_BASE, path), item_id,
66 *args, **compat_kwargs(kwargs))
67 self._handle_error(response)
70 def _real_initialize(self):
74 username, password = self._get_login_info()
80 'Unable to login. Twitch said: %s' % message, expected=True)
82 def login_step(page, urlh, note, data):
83 form = self._hidden_inputs(page)
86 page_url = urlh.geturl()
87 post_url = self._search_regex(
88 r'<form[^>]+action=(["\'])(?P<url>.+?)\1', page,
89 'post url', default=self._LOGIN_POST_URL, group='url')
90 post_url = urljoin(page_url, post_url)
95 'Content-Type': 'text/plain;charset=UTF-8',
98 response = self._download_json(
99 post_url, None, note, data=json.dumps(form).encode(),
100 headers=headers, expected_status=400)
101 error = response.get('error_description') or response.get('error_code')
105 if 'Authenticated successfully' in response.get('message', ''):
108 redirect_url = urljoin(
110 response.get('redirect') or response['redirect_path'])
111 return self._download_webpage_handle(
112 redirect_url, None, 'Downloading login redirect page',
115 login_page, handle = self._download_webpage_handle(
116 self._LOGIN_FORM_URL, None, 'Downloading login page')
118 # Some TOR nodes and public proxies are blocked completely
119 if 'blacklist_message' in login_page:
120 fail(clean_html(login_page))
122 redirect_page, handle = login_step(
123 login_page, handle, 'Logging in', {
124 'username': username,
125 'password': password,
126 'client_id': self._CLIENT_ID,
130 if not redirect_page:
133 if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None:
134 # TODO: Add mechanism to request an SMS or phone call
135 tfa_token = self._get_tfa_info('two-factor authentication token')
136 login_step(redirect_page, handle, 'Submitting TFA token', {
137 'authy_token': tfa_token,
138 'remember_2fa': 'true',
141 def _prefer_source(self, formats):
143 source = next(f for f in formats if f['format_id'] == 'Source')
144 source['quality'] = 10
145 except StopIteration:
147 if '/chunked/' in f['url']:
150 'format_note': 'Source',
152 self._sort_formats(formats)
154 def _download_access_token(self, channel_name):
155 return self._call_api(
156 'api/channels/%s/access_token' % channel_name, channel_name,
157 'Downloading access token JSON')
159 def _extract_channel_id(self, token, channel_name):
160 return compat_str(self._parse_json(token, channel_name)['channel_id'])
163 class TwitchVodIE(TwitchBaseIE):
164 IE_NAME = 'twitch:vod'
165 _VALID_URL = r'''(?x)
168 (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v(?:ideo)?|videos)/|
169 player\.twitch\.tv/\?.*?\bvideo=v?
177 'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s',
181 'title': 'LCK Summer Split - Week 6 Day 1',
182 'thumbnail': r're:^https?://.*\.jpg$',
184 'timestamp': 1435131709,
185 'upload_date': '20150624',
186 'uploader': 'Riot Games',
187 'uploader_id': 'riotgames',
193 'skip_download': True,
196 # Untitled broadcast (title is None)
197 'url': 'http://www.twitch.tv/belkao_o/v/11230755',
201 'title': 'Untitled Broadcast',
202 'thumbnail': r're:^https?://.*\.jpg$',
204 'timestamp': 1439746708,
205 'upload_date': '20150816',
206 'uploader': 'BelkAO_o',
207 'uploader_id': 'belkao_o',
212 'skip_download': True,
214 'skip': 'HTTP Error 404: Not Found',
216 'url': 'http://player.twitch.tv/?t=5m10s&video=v6528877',
217 'only_matching': True,
219 'url': 'https://www.twitch.tv/videos/6528877',
220 'only_matching': True,
222 'url': 'https://m.twitch.tv/beagsandjam/v/247478721',
223 'only_matching': True,
225 'url': 'https://www.twitch.tv/northernlion/video/291940395',
226 'only_matching': True,
228 'url': 'https://player.twitch.tv/?video=480452374',
229 'only_matching': True,
232 def _download_info(self, item_id):
233 return self._extract_info(
235 'kraken/videos/%s' % item_id, item_id,
236 'Downloading video info JSON'))
239 def _extract_info(info):
240 status = info.get('status')
241 if status == 'recording':
243 elif status == 'recorded':
247 _QUALITIES = ('small', 'medium', 'large')
248 quality_key = qualities(_QUALITIES)
250 preview = info.get('preview')
251 if isinstance(preview, dict):
252 for thumbnail_id, thumbnail_url in preview.items():
253 thumbnail_url = url_or_none(thumbnail_url)
254 if not thumbnail_url:
256 if thumbnail_id not in _QUALITIES:
259 'url': thumbnail_url,
260 'preference': quality_key(thumbnail_id),
264 'title': info.get('title') or 'Untitled Broadcast',
265 'description': info.get('description'),
266 'duration': int_or_none(info.get('length')),
267 'thumbnails': thumbnails,
268 'uploader': info.get('channel', {}).get('display_name'),
269 'uploader_id': info.get('channel', {}).get('name'),
270 'timestamp': parse_iso8601(info.get('recorded_at')),
271 'view_count': int_or_none(info.get('views')),
275 def _real_extract(self, url):
276 vod_id = self._match_id(url)
278 info = self._download_info(vod_id)
279 access_token = self._call_api(
280 'api/vods/%s/access_token' % vod_id, vod_id,
281 'Downloading %s access token' % self._ITEM_TYPE)
283 formats = self._extract_m3u8_formats(
284 '%s/vod/%s.m3u8?%s' % (
285 self._USHER_BASE, vod_id,
286 compat_urllib_parse_urlencode({
287 'allow_source': 'true',
288 'allow_audio_only': 'true',
289 'allow_spectre': 'true',
290 'player': 'twitchweb',
291 'playlist_include_framerate': 'true',
292 'nauth': access_token['token'],
293 'nauthsig': access_token['sig'],
295 vod_id, 'mp4', entry_protocol='m3u8_native')
297 self._prefer_source(formats)
298 info['formats'] = formats
300 parsed_url = compat_urllib_parse_urlparse(url)
301 query = compat_parse_qs(parsed_url.query)
303 info['start_time'] = parse_duration(query['t'][0])
305 if info.get('timestamp') is not None:
306 info['subtitles'] = {
308 'url': update_url_query(
309 'https://api.twitch.tv/v5/videos/%s/comments' % vod_id, {
310 'client_id': self._CLIENT_ID,
319 def _make_video_result(node):
320 assert isinstance(node, dict)
321 video_id = node.get('id')
325 '_type': 'url_transparent',
326 'ie_key': TwitchVodIE.ie_key(),
328 'url': 'https://www.twitch.tv/videos/%s' % video_id,
329 'title': node.get('title'),
330 'thumbnail': node.get('previewThumbnailURL'),
331 'duration': float_or_none(node.get('lengthSeconds')),
332 'view_count': int_or_none(node.get('viewCount')),
336 class TwitchGraphQLBaseIE(TwitchBaseIE):
339 _OPERATION_HASHES = {
340 'CollectionSideBar': '27111f1b382effad0b6def325caef1909c733fe6a4fbabf54f8d491ef2cf2f14',
341 'FilterableVideoTower_Videos': 'a937f1d22e269e39a03b509f65a7490f9fc247d7f83d6ac1421523e3b68042cb',
342 'ClipsCards__User': 'b73ad2bfaecfd30a9e6c28fada15bd97032c83ec77a0440766a56fe0bd632777',
343 'ChannelCollectionsContent': '07e3691a1bad77a36aba590c351180439a40baefc1c275356f40fc7082419a84',
344 'StreamMetadata': '1c719a40e481453e5c48d9bb585d971b8b372f8ebb105b17076722264dfa5b3e',
345 'ComscoreStreamingQuery': 'e1edae8122517d013405f237ffcc124515dc6ded82480a88daef69c83b53ac01',
346 'VideoPreviewOverlay': '3006e77e51b128d838fa4e835723ca4dc9a05c5efd4466c1085215c6e437e65c',
349 def _download_gql(self, video_id, ops, note, fatal=True):
354 'sha256Hash': self._OPERATION_HASHES[op['operationName']],
357 return self._download_json(
358 'https://gql.twitch.tv/gql', video_id, note,
359 data=json.dumps(ops).encode(),
361 'Content-Type': 'text/plain;charset=UTF-8',
362 'Client-ID': self._CLIENT_ID,
366 class TwitchCollectionIE(TwitchGraphQLBaseIE):
367 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/collections/(?P<id>[^/]+)'
370 'url': 'https://www.twitch.tv/collections/wlDCoH0zEBZZbQ',
372 'id': 'wlDCoH0zEBZZbQ',
373 'title': 'Overthrow Nook, capitalism for children',
375 'playlist_mincount': 13,
378 _OPERATION_NAME = 'CollectionSideBar'
380 def _real_extract(self, url):
381 collection_id = self._match_id(url)
382 collection = self._download_gql(
384 'operationName': self._OPERATION_NAME,
385 'variables': {'collectionID': collection_id},
387 'Downloading collection GraphQL')[0]['data']['collection']
388 title = collection.get('title')
390 for edge in collection['items']['edges']:
391 if not isinstance(edge, dict):
393 node = edge.get('node')
394 if not isinstance(node, dict):
396 video = _make_video_result(node)
398 entries.append(video)
399 return self.playlist_result(
400 entries, playlist_id=collection_id, playlist_title=title)
403 class TwitchPlaylistBaseIE(TwitchGraphQLBaseIE):
404 def _entries(self, channel_name, *args):
406 variables_common = self._make_variables(channel_name, *args)
407 entries_key = '%ss' % self._ENTRY_KIND
408 for page_num in itertools.count(1):
409 variables = variables_common.copy()
410 variables['limit'] = self._PAGE_LIMIT
412 variables['cursor'] = cursor
413 page = self._download_gql(
415 'operationName': self._OPERATION_NAME,
416 'variables': variables,
418 'Downloading %ss GraphQL page %s' % (self._NODE_KIND, page_num),
423 page, lambda x: x[0]['data']['user'][entries_key]['edges'], list)
427 if not isinstance(edge, dict):
429 if edge.get('__typename') != self._EDGE_KIND:
431 node = edge.get('node')
432 if not isinstance(node, dict):
434 if node.get('__typename') != self._NODE_KIND:
436 entry = self._extract_entry(node)
438 cursor = edge.get('cursor')
440 if not cursor or not isinstance(cursor, compat_str):
443 # Deprecated kraken v5 API
444 def _entries_kraken(self, channel_name, broadcast_type, sort):
445 access_token = self._download_access_token(channel_name)
446 channel_id = self._extract_channel_id(access_token['token'], channel_name)
448 counter_override = None
449 for counter in itertools.count(1):
450 response = self._call_api(
451 'kraken/channels/%s/videos/' % channel_id,
453 'Downloading video JSON page %s' % (counter_override or counter),
456 'limit': self._PAGE_LIMIT,
457 'broadcast_type': broadcast_type,
460 videos = response.get('videos')
461 if not isinstance(videos, list):
464 if not isinstance(video, dict):
466 video_url = url_or_none(video.get('url'))
470 '_type': 'url_transparent',
471 'ie_key': TwitchVodIE.ie_key(),
472 'id': video.get('_id'),
474 'title': video.get('title'),
475 'description': video.get('description'),
476 'timestamp': unified_timestamp(video.get('published_at')),
477 'duration': float_or_none(video.get('length')),
478 'view_count': int_or_none(video.get('views')),
479 'language': video.get('language'),
481 offset += self._PAGE_LIMIT
482 total = int_or_none(response.get('_total'))
483 if total and offset >= total:
487 class TwitchVideosIE(TwitchPlaylistBaseIE):
488 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:videos|profile)'
491 # All Videos sorted by Date
492 'url': 'https://www.twitch.tv/spamfish/videos?filter=all',
495 'title': 'spamfish - All Videos sorted by Date',
497 'playlist_mincount': 924,
499 # All Videos sorted by Popular
500 'url': 'https://www.twitch.tv/spamfish/videos?filter=all&sort=views',
503 'title': 'spamfish - All Videos sorted by Popular',
505 'playlist_mincount': 931,
507 # Past Broadcasts sorted by Date
508 'url': 'https://www.twitch.tv/spamfish/videos?filter=archives',
511 'title': 'spamfish - Past Broadcasts sorted by Date',
513 'playlist_mincount': 27,
515 # Highlights sorted by Date
516 'url': 'https://www.twitch.tv/spamfish/videos?filter=highlights',
519 'title': 'spamfish - Highlights sorted by Date',
521 'playlist_mincount': 901,
523 # Uploads sorted by Date
524 'url': 'https://www.twitch.tv/esl_csgo/videos?filter=uploads&sort=time',
527 'title': 'esl_csgo - Uploads sorted by Date',
529 'playlist_mincount': 5,
531 # Past Premieres sorted by Date
532 'url': 'https://www.twitch.tv/spamfish/videos?filter=past_premieres',
535 'title': 'spamfish - Past Premieres sorted by Date',
537 'playlist_mincount': 1,
539 'url': 'https://www.twitch.tv/spamfish/videos/all',
540 'only_matching': True,
542 'url': 'https://m.twitch.tv/spamfish/videos/all',
543 'only_matching': True,
545 'url': 'https://www.twitch.tv/spamfish/videos',
546 'only_matching': True,
549 Broadcast = collections.namedtuple('Broadcast', ['type', 'label'])
551 _DEFAULT_BROADCAST = Broadcast(None, 'All Videos')
553 'archives': Broadcast('ARCHIVE', 'Past Broadcasts'),
554 'highlights': Broadcast('HIGHLIGHT', 'Highlights'),
555 'uploads': Broadcast('UPLOAD', 'Uploads'),
556 'past_premieres': Broadcast('PAST_PREMIERE', 'Past Premieres'),
557 'all': _DEFAULT_BROADCAST,
560 _DEFAULT_SORTED_BY = 'Date'
562 'time': _DEFAULT_SORTED_BY,
566 _OPERATION_NAME = 'FilterableVideoTower_Videos'
567 _ENTRY_KIND = 'video'
568 _EDGE_KIND = 'VideoEdge'
572 def suitable(cls, url):
574 if any(ie.suitable(url) for ie in (
576 TwitchVideosCollectionsIE))
577 else super(TwitchVideosIE, cls).suitable(url))
580 def _make_variables(channel_name, broadcast_type, sort):
582 'channelOwnerLogin': channel_name,
583 'broadcastType': broadcast_type,
584 'videoSort': sort.upper(),
588 def _extract_entry(node):
589 return _make_video_result(node)
591 def _real_extract(self, url):
592 channel_name = self._match_id(url)
593 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
594 filter = qs.get('filter', ['all'])[0]
595 sort = qs.get('sort', ['time'])[0]
596 broadcast = self._BROADCASTS.get(filter, self._DEFAULT_BROADCAST)
597 return self.playlist_result(
598 self._entries(channel_name, broadcast.type, sort),
599 playlist_id=channel_name,
600 playlist_title='%s - %s sorted by %s'
601 % (channel_name, broadcast.label,
602 self._SORTED_BY.get(sort, self._DEFAULT_SORTED_BY)))
605 class TwitchVideosClipsIE(TwitchPlaylistBaseIE):
606 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:clips|videos/*?\?.*?\bfilter=clips)'
610 'url': 'https://www.twitch.tv/vanillatv/clips?filter=clips&range=all',
613 'title': 'vanillatv - Clips Top All',
615 'playlist_mincount': 1,
617 'url': 'https://www.twitch.tv/dota2ruhub/videos?filter=clips&range=7d',
618 'only_matching': True,
621 Clip = collections.namedtuple('Clip', ['filter', 'label'])
623 _DEFAULT_CLIP = Clip('LAST_WEEK', 'Top 7D')
625 '24hr': Clip('LAST_DAY', 'Top 24H'),
627 '30d': Clip('LAST_MONTH', 'Top 30D'),
628 'all': Clip('ALL_TIME', 'Top All'),
631 # NB: values other than 20 result in skipped videos
634 _OPERATION_NAME = 'ClipsCards__User'
636 _EDGE_KIND = 'ClipEdge'
640 def _make_variables(channel_name, filter):
642 'login': channel_name,
649 def _extract_entry(node):
650 assert isinstance(node, dict)
651 clip_url = url_or_none(node.get('url'))
655 '_type': 'url_transparent',
656 'ie_key': TwitchClipsIE.ie_key(),
657 'id': node.get('id'),
659 'title': node.get('title'),
660 'thumbnail': node.get('thumbnailURL'),
661 'duration': float_or_none(node.get('durationSeconds')),
662 'timestamp': unified_timestamp(node.get('createdAt')),
663 'view_count': int_or_none(node.get('viewCount')),
664 'language': node.get('language'),
667 def _real_extract(self, url):
668 channel_name = self._match_id(url)
669 qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
670 range = qs.get('range', ['7d'])[0]
671 clip = self._RANGE.get(range, self._DEFAULT_CLIP)
672 return self.playlist_result(
673 self._entries(channel_name, clip.filter),
674 playlist_id=channel_name,
675 playlist_title='%s - Clips %s' % (channel_name, clip.label))
678 class TwitchVideosCollectionsIE(TwitchPlaylistBaseIE):
679 _VALID_URL = r'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/videos/*?\?.*?\bfilter=collections'
683 'url': 'https://www.twitch.tv/spamfish/videos?filter=collections',
686 'title': 'spamfish - Collections',
688 'playlist_mincount': 3,
691 _OPERATION_NAME = 'ChannelCollectionsContent'
692 _ENTRY_KIND = 'collection'
693 _EDGE_KIND = 'CollectionsItemEdge'
694 _NODE_KIND = 'Collection'
697 def _make_variables(channel_name):
699 'ownerLogin': channel_name,
703 def _extract_entry(node):
704 assert isinstance(node, dict)
705 collection_id = node.get('id')
706 if not collection_id:
709 '_type': 'url_transparent',
710 'ie_key': TwitchCollectionIE.ie_key(),
712 'url': 'https://www.twitch.tv/collections/%s' % collection_id,
713 'title': node.get('title'),
714 'thumbnail': node.get('thumbnailURL'),
715 'duration': float_or_none(node.get('lengthSeconds')),
716 'timestamp': unified_timestamp(node.get('updatedAt')),
717 'view_count': int_or_none(node.get('viewCount')),
720 def _real_extract(self, url):
721 channel_name = self._match_id(url)
722 return self.playlist_result(
723 self._entries(channel_name), playlist_id=channel_name,
724 playlist_title='%s - Collections' % channel_name)
727 class TwitchStreamIE(TwitchGraphQLBaseIE):
728 IE_NAME = 'twitch:stream'
729 _VALID_URL = r'''(?x)
732 (?:(?:www|go|m)\.)?twitch\.tv/|
733 player\.twitch\.tv/\?.*?\bchannel=
739 'url': 'http://www.twitch.tv/shroomztv',
742 'display_id': 'shroomztv',
744 'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
745 'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV',
747 'timestamp': 1421928037,
748 'upload_date': '20150122',
749 'uploader': 'ShroomzTV',
750 'uploader_id': 'shroomztv',
755 'skip_download': True,
758 'url': 'http://www.twitch.tv/miracle_doto#profile-0',
759 'only_matching': True,
761 'url': 'https://player.twitch.tv/?channel=lotsofs',
762 'only_matching': True,
764 'url': 'https://go.twitch.tv/food',
765 'only_matching': True,
767 'url': 'https://m.twitch.tv/food',
768 'only_matching': True,
772 def suitable(cls, url):
774 if any(ie.suitable(url) for ie in (
779 TwitchVideosCollectionsIE,
781 else super(TwitchStreamIE, cls).suitable(url))
783 def _real_extract(self, url):
784 channel_name = self._match_id(url).lower()
786 gql = self._download_gql(
788 'operationName': 'StreamMetadata',
789 'variables': {'channelLogin': channel_name},
791 'operationName': 'ComscoreStreamingQuery',
793 'channel': channel_name,
797 'isVodOrCollection': False,
801 'operationName': 'VideoPreviewOverlay',
802 'variables': {'login': channel_name},
804 'Downloading stream GraphQL')
806 user = gql[0]['data']['user']
809 raise ExtractorError(
810 '%s does not exist' % channel_name, expected=True)
812 stream = user['stream']
815 raise ExtractorError('%s is offline' % channel_name, expected=True)
817 access_token = self._download_access_token(channel_name)
818 token = access_token['token']
820 stream_id = stream.get('id') or channel_name
822 'allow_source': 'true',
823 'allow_audio_only': 'true',
824 'allow_spectre': 'true',
825 'p': random.randint(1000000, 10000000),
826 'player': 'twitchweb',
827 'playlist_include_framerate': 'true',
828 'segment_preference': '4',
829 'sig': access_token['sig'].encode('utf-8'),
830 'token': token.encode('utf-8'),
832 formats = self._extract_m3u8_formats(
833 '%s/api/channel/hls/%s.m3u8' % (self._USHER_BASE, channel_name),
834 stream_id, 'mp4', query=query)
835 self._prefer_source(formats)
837 view_count = stream.get('viewers')
838 timestamp = unified_timestamp(stream.get('createdAt'))
840 sq_user = try_get(gql, lambda x: x[1]['data']['user'], dict) or {}
841 uploader = sq_user.get('displayName')
842 description = try_get(
843 sq_user, lambda x: x['broadcastSettings']['title'], compat_str)
845 thumbnail = url_or_none(try_get(
846 gql, lambda x: x[2]['data']['user']['stream']['previewImageURL'],
849 title = uploader or channel_name
850 stream_type = stream.get('type')
851 if stream_type in ['rerun', 'live']:
852 title += ' (%s)' % stream_type
856 'display_id': channel_name,
857 'title': self._live_title(title),
858 'description': description,
859 'thumbnail': thumbnail,
860 'uploader': uploader,
861 'uploader_id': channel_name,
862 'timestamp': timestamp,
863 'view_count': view_count,
865 'is_live': stream_type == 'live',
869 class TwitchClipsIE(TwitchBaseIE):
870 IE_NAME = 'twitch:clips'
871 _VALID_URL = r'''(?x)
874 clips\.twitch\.tv/(?:embed\?.*?\bclip=|(?:[^/]+/)*)|
875 (?:(?:www|go|m)\.)?twitch\.tv/[^/]+/clip/
881 'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',
882 'md5': '761769e1eafce0ffebfb4089cb3847cd',
886 'title': 'EA Play 2016 Live from the Novo Theatre',
887 'thumbnail': r're:^https?://.*\.jpg',
888 'timestamp': 1465767393,
889 'upload_date': '20160612',
891 'uploader': 'stereotype_',
892 'uploader_id': '43566419',
896 'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy',
897 'only_matching': True,
899 'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan',
900 'only_matching': True,
902 'url': 'https://clips.twitch.tv/embed?clip=InquisitiveBreakableYogurtJebaited',
903 'only_matching': True,
905 'url': 'https://m.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
906 'only_matching': True,
908 'url': 'https://go.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
909 'only_matching': True,
912 def _real_extract(self, url):
913 video_id = self._match_id(url)
915 clip = self._download_json(
916 'https://gql.twitch.tv/gql', video_id, data=json.dumps({
929 tiny: thumbnailURL(width: 86, height: 45)
930 small: thumbnailURL(width: 260, height: 147)
931 medium: thumbnailURL(width: 480, height: 272)
941 }).encode(), headers={
942 'Client-ID': self._CLIENT_ID,
946 raise ExtractorError(
947 'This clip is no longer available', expected=True)
950 for option in clip.get('videoQualities', []):
951 if not isinstance(option, dict):
953 source = url_or_none(option.get('sourceURL'))
958 'format_id': option.get('quality'),
959 'height': int_or_none(option.get('quality')),
960 'fps': int_or_none(option.get('frameRate')),
962 self._sort_formats(formats)
965 for thumbnail_id in ('tiny', 'small', 'medium'):
966 thumbnail_url = clip.get(thumbnail_id)
967 if not thumbnail_url:
971 'url': thumbnail_url,
973 mobj = re.search(r'-(\d+)x(\d+)\.', thumbnail_url)
976 'height': int(mobj.group(2)),
977 'width': int(mobj.group(1)),
979 thumbnails.append(thumb)
982 'id': clip.get('id') or video_id,
983 'title': clip.get('title') or video_id,
985 'duration': int_or_none(clip.get('durationSeconds')),
986 'views': int_or_none(clip.get('viewCount')),
987 'timestamp': unified_timestamp(clip.get('createdAt')),
988 'thumbnails': thumbnails,
989 'creator': try_get(clip, lambda x: x['broadcaster']['displayName'], compat_str),
990 'uploader': try_get(clip, lambda x: x['curator']['displayName'], compat_str),
991 'uploader_id': try_get(clip, lambda x: x['curator']['id'], compat_str),