7 from .common
import InfoExtractor
11 compat_urllib_parse_urlparse
,
36 class TwitchBaseIE(InfoExtractor
):
37 _VALID_URL_BASE
= r
'https?://(?:(?:www|go|m)\.)?twitch\.tv'
39 _API_BASE
= 'https://api.twitch.tv'
40 _USHER_BASE
= 'https://usher.ttvnw.net'
41 _LOGIN_FORM_URL
= 'https://www.twitch.tv/login'
42 _LOGIN_POST_URL
= 'https://passport.twitch.tv/login'
43 _NETRC_MACHINE
= 'twitch'
46 'CollectionSideBar': '27111f1b382effad0b6def325caef1909c733fe6a4fbabf54f8d491ef2cf2f14',
47 'FilterableVideoTower_Videos': 'a937f1d22e269e39a03b509f65a7490f9fc247d7f83d6ac1421523e3b68042cb',
48 'ClipsCards__User': 'b73ad2bfaecfd30a9e6c28fada15bd97032c83ec77a0440766a56fe0bd632777',
49 'ChannelCollectionsContent': '447aec6a0cc1e8d0a8d7732d47eb0762c336a2294fdb009e9c9d854e49d484b9',
50 'StreamMetadata': 'a647c2a13599e5991e175155f798ca7f1ecddde73f7f341f39009c14dbf59962',
51 'ComscoreStreamingQuery': 'e1edae8122517d013405f237ffcc124515dc6ded82480a88daef69c83b53ac01',
52 'VideoAccessToken_Clip': '36b89d2507fce29e5ca551df756d27c1cfe079e2609642b4390aa4c35796eb11',
53 'VideoPreviewOverlay': '3006e77e51b128d838fa4e835723ca4dc9a05c5efd4466c1085215c6e437e65c',
54 'VideoMetadata': '49b5b8f268cdeb259d75b58dcb0c1a748e3b575003448a2333dc5cdafd49adad',
55 'VideoPlayer_ChapterSelectButtonVideo': '8d2793384aac3773beab5e59bd5d6f585aedb923d292800119e03d40cd0f9b41',
56 'VideoPlayer_VODSeekbarPreviewVideo': '07e99e4d56c5a7c67117a154777b0baf85a5ffefa393b213f4bc712ccaf85dd6',
61 return self
._configuration
_arg
(
62 'client_id', ['ue6666qo983tsx6so1t0vnawi233wa'], ie_key
='Twitch', casesense
=True)[0]
64 def _perform_login(self
, username
, password
):
67 'Unable to login. Twitch said: %s' % message
, expected
=True)
69 def login_step(page
, urlh
, note
, data
):
70 form
= self
._hidden
_inputs
(page
)
74 post_url
= self
._search
_regex
(
75 r
'<form[^>]+action=(["\'])(?P
<url
>.+?
)\
1', page,
76 'post url
', default=self._LOGIN_POST_URL, group='url
')
77 post_url = urljoin(page_url, post_url)
81 'Origin
': 'https
://www
.twitch
.tv
',
82 'Content
-Type
': 'text
/plain
;charset
=UTF
-8',
85 response = self._download_json(
86 post_url, None, note, data=json.dumps(form).encode(),
87 headers=headers, expected_status=400)
88 error = dict_get(response, ('error
', 'error_description
', 'error_code
'))
92 if 'Authenticated successfully
' in response.get('message
', ''):
95 redirect_url = urljoin(
97 response.get('redirect
') or response['redirect_path
'])
98 return self._download_webpage_handle(
99 redirect_url, None, 'Downloading login redirect page
',
102 login_page, handle = self._download_webpage_handle(
103 self._LOGIN_FORM_URL, None, 'Downloading login page
')
105 # Some TOR nodes and public proxies are blocked completely
106 if 'blacklist_message
' in login_page:
107 fail(clean_html(login_page))
109 redirect_page, handle = login_step(
110 login_page, handle, 'Logging
in', {
111 'username
': username,
112 'password
': password,
113 'client_id
': self._CLIENT_ID,
117 if not redirect_page:
120 if re.search(r'(?i
)<form
[^
>]+id="two-factor-submit"', redirect_page) is not None:
121 # TODO: Add mechanism to request an SMS or phone call
122 tfa_token = self._get_tfa_info('two
-factor authentication token
')
123 login_step(redirect_page, handle, 'Submitting TFA token
', {
124 'authy_token
': tfa_token,
125 'remember_2fa
': 'true
',
128 def _prefer_source(self, formats):
130 source = next(f for f in formats if f['format_id
'] == 'Source
')
131 source['quality
'] = 10
132 except StopIteration:
134 if '/chunked
/' in f['url
']:
137 'format_note
': 'Source
',
140 def _download_base_gql(self, video_id, ops, note, fatal=True):
142 'Content
-Type
': 'text
/plain
;charset
=UTF
-8',
143 'Client
-ID
': self._CLIENT_ID,
145 gql_auth = self._get_cookies('https
://gql
.twitch
.tv
').get('auth
-token
')
147 headers['Authorization
'] = 'OAuth
' + gql_auth.value
148 return self._download_json(
149 'https
://gql
.twitch
.tv
/gql
', video_id, note,
150 data=json.dumps(ops).encode(),
151 headers=headers, fatal=fatal)
153 def _download_gql(self, video_id, ops, note, fatal=True):
158 'sha256Hash
': self._OPERATION_HASHES[op['operationName
']],
161 return self._download_base_gql(video_id, ops, note)
163 def _download_access_token(self, video_id, token_kind, param_name):
164 method = '%sPlaybackAccessToken
' % token_kind
171 playerBackend: "mediaplayer",
179 }''' % (method, param_name, video_id),
181 return self._download_base_gql(
183 'Downloading
%s access token GraphQL
' % token_kind)['data
'][method]
185 def _get_thumbnails(self, thumbnail):
187 'url
': re.sub(r'\d
+x\d
+(\
.\w
+)($|
(?
=[?
#]))', r'0x0\g<1>', thumbnail),
191 }] if thumbnail
else None
193 def _extract_twitch_m3u8_formats(self
, video_id
, token
, signature
):
194 """Subclasses must define _M3U8_PATH"""
195 return self
._extract
_m
3u8_formats
(
196 f
'{self._USHER_BASE}/{self._M3U8_PATH}/{video_id}.m3u8', video_id
, 'mp4', query
={
197 'allow_source': 'true',
198 'allow_audio_only': 'true',
199 'allow_spectre': 'true',
200 'p': random
.randint(1000000, 10000000),
201 'player': 'twitchweb',
202 'playlist_include_framerate': 'true',
208 class TwitchVodIE(TwitchBaseIE
):
209 IE_NAME
= 'twitch:vod'
210 _VALID_URL
= r
'''(?x)
213 (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v(?:ideo)?|videos)/|
214 player\.twitch\.tv/\?.*?\bvideo=v?|
215 www\.twitch\.tv/[^/]+/schedule\?vodID=
222 'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s',
226 'title': 'LCK Summer Split - Week 6 Day 1',
227 'thumbnail': r
're:^https?://.*\.jpg$',
229 'timestamp': 1435131734,
230 'upload_date': '20150624',
231 'uploader': 'Riot Games',
232 'uploader_id': 'riotgames',
239 'title': 'League of Legends'
242 'live_status': 'was_live',
246 'skip_download': True,
249 # Untitled broadcast (title is None)
250 'url': 'http://www.twitch.tv/belkao_o/v/11230755',
254 'title': 'Untitled Broadcast',
255 'thumbnail': r
're:^https?://.*\.jpg$',
257 'timestamp': 1439746708,
258 'upload_date': '20150816',
259 'uploader': 'BelkAO_o',
260 'uploader_id': 'belkao_o',
265 'skip_download': True,
267 'skip': 'HTTP Error 404: Not Found',
269 'url': 'http://player.twitch.tv/?t=5m10s&video=v6528877',
270 'only_matching': True,
272 'url': 'https://www.twitch.tv/videos/6528877',
273 'only_matching': True,
275 'url': 'https://m.twitch.tv/beagsandjam/v/247478721',
276 'only_matching': True,
278 'url': 'https://www.twitch.tv/northernlion/video/291940395',
279 'only_matching': True,
281 'url': 'https://player.twitch.tv/?video=480452374',
282 'only_matching': True,
284 'url': 'https://www.twitch.tv/videos/635475444',
288 'title': 'Riot Games',
290 'uploader': 'Riot Games',
291 'uploader_id': 'riotgames',
292 'timestamp': 1590770569,
293 'upload_date': '20200529',
298 'title': 'League of Legends'
303 'title': 'Legends of Runeterra'
311 'live_status': 'was_live',
312 'thumbnail': r
're:^https?://.*\.jpg$',
316 'skip_download': True
319 'note': 'Storyboards',
320 'url': 'https://www.twitch.tv/videos/635475444',
325 'title': 'Riot Games',
327 'uploader': 'Riot Games',
328 'uploader_id': 'riotgames',
329 'timestamp': 1590770569,
330 'upload_date': '20200529',
335 'title': 'League of Legends'
340 'title': 'Legends of Runeterra'
348 'live_status': 'was_live',
349 'thumbnail': r
're:^https?://.*\.jpg$',
356 'skip_download': True
359 'note': 'VOD with single chapter',
360 'url': 'https://www.twitch.tv/videos/1536751224',
364 'title': 'Porter Robinson Star Guardian Stream Tour with LilyPichu',
366 'uploader': 'Riot Games',
367 'uploader_id': 'riotgames',
368 'timestamp': 1658267731,
369 'upload_date': '20220719',
374 'title': 'League of Legends'
377 'live_status': 'was_live',
378 'thumbnail': r
're:^https?://.*\.jpg$',
382 'skip_download': True
384 'expected_warnings': ['Unable to download JSON metadata: HTTP Error 403: Forbidden']
386 'url': 'https://www.twitch.tv/tangotek/schedule?vodID=1822395420',
387 'only_matching': True,
390 def _download_info(self
, item_id
):
391 data
= self
._download
_gql
(
393 'operationName': 'VideoMetadata',
399 'operationName': 'VideoPlayer_ChapterSelectButtonVideo',
401 'includePrivate': False,
405 'operationName': 'VideoPlayer_VODSeekbarPreviewVideo',
407 'includePrivate': False,
411 'Downloading stream metadata GraphQL')
413 video
= traverse_obj(data
, (..., 'data', 'video'), get_all
=False)
415 raise ExtractorError(f
'Video {item_id} does not exist', expected
=True)
417 video
['moments'] = traverse_obj(data
, (..., 'data', 'video', 'moments', 'edges', ..., 'node'))
418 video
['storyboard'] = traverse_obj(
419 data
, (..., 'data', 'video', 'seekPreviewsURL', {url_or_none}
), get_all
=False)
423 def _extract_info(self
, info
):
424 status
= info
.get('status')
425 if status
== 'recording':
427 elif status
== 'recorded':
431 _QUALITIES
= ('small', 'medium', 'large')
432 quality_key
= qualities(_QUALITIES
)
434 preview
= info
.get('preview')
435 if isinstance(preview
, dict):
436 for thumbnail_id
, thumbnail_url
in preview
.items():
437 thumbnail_url
= url_or_none(thumbnail_url
)
438 if not thumbnail_url
:
440 if thumbnail_id
not in _QUALITIES
:
443 'url': thumbnail_url
,
444 'preference': quality_key(thumbnail_id
),
448 'title': info
.get('title') or 'Untitled Broadcast',
449 'description': info
.get('description'),
450 'duration': int_or_none(info
.get('length')),
451 'thumbnails': thumbnails
,
452 'uploader': info
.get('channel', {}).get('display_name'),
453 'uploader_id': info
.get('channel', {}).get('name'),
454 'timestamp': parse_iso8601(info
.get('recorded_at')),
455 'view_count': int_or_none(info
.get('views')),
460 def _extract_chapters(self
, info
, item_id
):
461 if not info
.get('moments'):
462 game
= traverse_obj(info
, ('game', 'displayName'))
464 yield {'title': game}
467 for moment
in info
['moments']:
468 start_time
= int_or_none(moment
.get('positionMilliseconds'), 1000)
469 duration
= int_or_none(moment
.get('durationMilliseconds'), 1000)
470 name
= str_or_none(moment
.get('description'))
472 if start_time
is None or duration
is None:
473 self
.report_warning(f
'Important chapter information missing for chapter {name}', item_id
)
476 'start_time': start_time
,
477 'end_time': start_time
+ duration
,
481 def _extract_info_gql(self
, info
, item_id
):
482 vod_id
= info
.get('id') or item_id
483 # id backward compatibility for download archives
485 vod_id
= 'v%s' % vod_id
486 thumbnail
= url_or_none(info
.get('previewThumbnailURL'))
489 if re
.findall(r
'/404_processing_[^.?#]+\.png', thumbnail
):
490 is_live
, thumbnail
= True, None
496 'title': info
.get('title') or 'Untitled Broadcast',
497 'description': info
.get('description'),
498 'duration': int_or_none(info
.get('lengthSeconds')),
499 'thumbnails': self
._get
_thumbnails
(thumbnail
),
500 'uploader': try_get(info
, lambda x
: x
['owner']['displayName'], compat_str
),
501 'uploader_id': try_get(info
, lambda x
: x
['owner']['login'], compat_str
),
502 'timestamp': unified_timestamp(info
.get('publishedAt')),
503 'view_count': int_or_none(info
.get('viewCount')),
504 'chapters': list(self
._extract
_chapters
(info
, item_id
)),
509 def _extract_storyboard(self
, item_id
, storyboard_json_url
, duration
):
510 if not duration
or not storyboard_json_url
:
512 spec
= self
._download
_json
(storyboard_json_url
, item_id
, 'Downloading storyboard metadata JSON', fatal
=False) or []
513 # sort from highest quality to lowest
514 # This makes sb0 the highest-quality format, sb1 - lower, etc which is consistent with youtube sb ordering
515 spec
.sort(key
=lambda x
: int_or_none(x
.get('width')) or 0, reverse
=True)
516 base
= base_url(storyboard_json_url
)
517 for i
, s
in enumerate(spec
):
518 count
= int_or_none(s
.get('count'))
519 images
= s
.get('images')
520 if not (images
and count
):
522 fragment_duration
= duration
/ len(images
)
524 'format_id': f
'sb{i}',
525 'format_note': 'storyboard',
530 'url': urljoin(base
, images
[0]),
531 'width': int_or_none(s
.get('width')),
532 'height': int_or_none(s
.get('height')),
533 'fps': count
/ duration
,
534 'rows': int_or_none(s
.get('rows')),
535 'columns': int_or_none(s
.get('cols')),
537 'url': urljoin(base
, path
),
538 'duration': fragment_duration
,
539 } for path
in images
],
542 def _real_extract(self
, url
):
543 vod_id
= self
._match
_id
(url
)
545 video
= self
._download
_info
(vod_id
)
546 info
= self
._extract
_info
_gql
(video
, vod_id
)
547 access_token
= self
._download
_access
_token
(vod_id
, 'video', 'id')
549 formats
= self
._extract
_twitch
_m
3u8_formats
(
550 vod_id
, access_token
['value'], access_token
['signature'])
551 formats
.extend(self
._extract
_storyboard
(vod_id
, video
.get('storyboard'), info
.get('duration')))
553 self
._prefer
_source
(formats
)
554 info
['formats'] = formats
556 parsed_url
= compat_urllib_parse_urlparse(url
)
557 query
= compat_parse_qs(parsed_url
.query
)
559 info
['start_time'] = parse_duration(query
['t'][0])
561 if info
.get('timestamp') is not None:
562 info
['subtitles'] = {
564 'url': update_url_query(
565 'https://api.twitch.tv/v5/videos/%s/comments' % vod_id
, {
566 'client_id': self
._CLIENT
_ID
,
575 def _make_video_result(node
):
576 assert isinstance(node
, dict)
577 video_id
= node
.get('id')
581 '_type': 'url_transparent',
582 'ie_key': TwitchVodIE
.ie_key(),
583 'id': 'v' + video_id
,
584 'url': 'https://www.twitch.tv/videos/%s' % video_id
,
585 'title': node
.get('title'),
586 'thumbnail': node
.get('previewThumbnailURL'),
587 'duration': float_or_none(node
.get('lengthSeconds')),
588 'view_count': int_or_none(node
.get('viewCount')),
592 class TwitchCollectionIE(TwitchBaseIE
):
593 _VALID_URL
= r
'https?://(?:(?:www|go|m)\.)?twitch\.tv/collections/(?P<id>[^/]+)'
596 'url': 'https://www.twitch.tv/collections/wlDCoH0zEBZZbQ',
598 'id': 'wlDCoH0zEBZZbQ',
599 'title': 'Overthrow Nook, capitalism for children',
601 'playlist_mincount': 13,
604 _OPERATION_NAME
= 'CollectionSideBar'
606 def _real_extract(self
, url
):
607 collection_id
= self
._match
_id
(url
)
608 collection
= self
._download
_gql
(
610 'operationName': self
._OPERATION
_NAME
,
611 'variables': {'collectionID': collection_id}
,
613 'Downloading collection GraphQL')[0]['data']['collection']
614 title
= collection
.get('title')
616 for edge
in collection
['items']['edges']:
617 if not isinstance(edge
, dict):
619 node
= edge
.get('node')
620 if not isinstance(node
, dict):
622 video
= _make_video_result(node
)
624 entries
.append(video
)
625 return self
.playlist_result(
626 entries
, playlist_id
=collection_id
, playlist_title
=title
)
629 class TwitchPlaylistBaseIE(TwitchBaseIE
):
632 def _entries(self
, channel_name
, *args
):
634 variables_common
= self
._make
_variables
(channel_name
, *args
)
635 entries_key
= '%ss' % self
._ENTRY
_KIND
636 for page_num
in itertools
.count(1):
637 variables
= variables_common
.copy()
638 variables
['limit'] = self
._PAGE
_LIMIT
640 variables
['cursor'] = cursor
641 page
= self
._download
_gql
(
643 'operationName': self
._OPERATION
_NAME
,
644 'variables': variables
,
646 'Downloading %ss GraphQL page %s' % (self
._NODE
_KIND
, page_num
),
651 page
, lambda x
: x
[0]['data']['user'][entries_key
]['edges'], list)
655 if not isinstance(edge
, dict):
657 if edge
.get('__typename') != self
._EDGE
_KIND
:
659 node
= edge
.get('node')
660 if not isinstance(node
, dict):
662 if node
.get('__typename') != self
._NODE
_KIND
:
664 entry
= self
._extract
_entry
(node
)
666 cursor
= edge
.get('cursor')
668 if not cursor
or not isinstance(cursor
, compat_str
):
672 class TwitchVideosIE(TwitchPlaylistBaseIE
):
673 _VALID_URL
= r
'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:videos|profile)'
676 # All Videos sorted by Date
677 'url': 'https://www.twitch.tv/spamfish/videos?filter=all',
680 'title': 'spamfish - All Videos sorted by Date',
682 'playlist_mincount': 924,
684 # All Videos sorted by Popular
685 'url': 'https://www.twitch.tv/spamfish/videos?filter=all&sort=views',
688 'title': 'spamfish - All Videos sorted by Popular',
690 'playlist_mincount': 931,
692 # Past Broadcasts sorted by Date
693 'url': 'https://www.twitch.tv/spamfish/videos?filter=archives',
696 'title': 'spamfish - Past Broadcasts sorted by Date',
698 'playlist_mincount': 27,
700 # Highlights sorted by Date
701 'url': 'https://www.twitch.tv/spamfish/videos?filter=highlights',
704 'title': 'spamfish - Highlights sorted by Date',
706 'playlist_mincount': 901,
708 # Uploads sorted by Date
709 'url': 'https://www.twitch.tv/esl_csgo/videos?filter=uploads&sort=time',
712 'title': 'esl_csgo - Uploads sorted by Date',
714 'playlist_mincount': 5,
716 # Past Premieres sorted by Date
717 'url': 'https://www.twitch.tv/spamfish/videos?filter=past_premieres',
720 'title': 'spamfish - Past Premieres sorted by Date',
722 'playlist_mincount': 1,
724 'url': 'https://www.twitch.tv/spamfish/videos/all',
725 'only_matching': True,
727 'url': 'https://m.twitch.tv/spamfish/videos/all',
728 'only_matching': True,
730 'url': 'https://www.twitch.tv/spamfish/videos',
731 'only_matching': True,
734 Broadcast
= collections
.namedtuple('Broadcast', ['type', 'label'])
736 _DEFAULT_BROADCAST
= Broadcast(None, 'All Videos')
738 'archives': Broadcast('ARCHIVE', 'Past Broadcasts'),
739 'highlights': Broadcast('HIGHLIGHT', 'Highlights'),
740 'uploads': Broadcast('UPLOAD', 'Uploads'),
741 'past_premieres': Broadcast('PAST_PREMIERE', 'Past Premieres'),
742 'all': _DEFAULT_BROADCAST
,
745 _DEFAULT_SORTED_BY
= 'Date'
747 'time': _DEFAULT_SORTED_BY
,
751 _OPERATION_NAME
= 'FilterableVideoTower_Videos'
752 _ENTRY_KIND
= 'video'
753 _EDGE_KIND
= 'VideoEdge'
757 def suitable(cls
, url
):
759 if any(ie
.suitable(url
) for ie
in (
761 TwitchVideosCollectionsIE
))
762 else super(TwitchVideosIE
, cls
).suitable(url
))
765 def _make_variables(channel_name
, broadcast_type
, sort
):
767 'channelOwnerLogin': channel_name
,
768 'broadcastType': broadcast_type
,
769 'videoSort': sort
.upper(),
773 def _extract_entry(node
):
774 return _make_video_result(node
)
776 def _real_extract(self
, url
):
777 channel_name
= self
._match
_id
(url
)
779 filter = qs
.get('filter', ['all'])[0]
780 sort
= qs
.get('sort', ['time'])[0]
781 broadcast
= self
._BROADCASTS
.get(filter, self
._DEFAULT
_BROADCAST
)
782 return self
.playlist_result(
783 self
._entries
(channel_name
, broadcast
.type, sort
),
784 playlist_id
=channel_name
,
785 playlist_title
='%s - %s sorted by %s'
786 % (channel_name
, broadcast
.label
,
787 self
._SORTED
_BY
.get(sort
, self
._DEFAULT
_SORTED
_BY
)))
790 class TwitchVideosClipsIE(TwitchPlaylistBaseIE
):
791 _VALID_URL
= r
'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/(?:clips|videos/*?\?.*?\bfilter=clips)'
795 'url': 'https://www.twitch.tv/vanillatv/clips?filter=clips&range=all',
798 'title': 'vanillatv - Clips Top All',
800 'playlist_mincount': 1,
802 'url': 'https://www.twitch.tv/dota2ruhub/videos?filter=clips&range=7d',
803 'only_matching': True,
806 Clip
= collections
.namedtuple('Clip', ['filter', 'label'])
808 _DEFAULT_CLIP
= Clip('LAST_WEEK', 'Top 7D')
810 '24hr': Clip('LAST_DAY', 'Top 24H'),
812 '30d': Clip('LAST_MONTH', 'Top 30D'),
813 'all': Clip('ALL_TIME', 'Top All'),
816 # NB: values other than 20 result in skipped videos
819 _OPERATION_NAME
= 'ClipsCards__User'
821 _EDGE_KIND
= 'ClipEdge'
825 def _make_variables(channel_name
, filter):
827 'login': channel_name
,
834 def _extract_entry(node
):
835 assert isinstance(node
, dict)
836 clip_url
= url_or_none(node
.get('url'))
840 '_type': 'url_transparent',
841 'ie_key': TwitchClipsIE
.ie_key(),
842 'id': node
.get('id'),
844 'title': node
.get('title'),
845 'thumbnail': node
.get('thumbnailURL'),
846 'duration': float_or_none(node
.get('durationSeconds')),
847 'timestamp': unified_timestamp(node
.get('createdAt')),
848 'view_count': int_or_none(node
.get('viewCount')),
849 'language': node
.get('language'),
852 def _real_extract(self
, url
):
853 channel_name
= self
._match
_id
(url
)
855 range = qs
.get('range', ['7d'])[0]
856 clip
= self
._RANGE
.get(range, self
._DEFAULT
_CLIP
)
857 return self
.playlist_result(
858 self
._entries
(channel_name
, clip
.filter),
859 playlist_id
=channel_name
,
860 playlist_title
='%s - Clips %s' % (channel_name
, clip
.label
))
863 class TwitchVideosCollectionsIE(TwitchPlaylistBaseIE
):
864 _VALID_URL
= r
'https?://(?:(?:www|go|m)\.)?twitch\.tv/(?P<id>[^/]+)/videos/*?\?.*?\bfilter=collections'
868 'url': 'https://www.twitch.tv/spamfish/videos?filter=collections',
871 'title': 'spamfish - Collections',
873 'playlist_mincount': 3,
875 'url': 'https://www.twitch.tv/monstercat/videos?filter=collections',
878 'title': 'monstercat - Collections',
880 'playlist_mincount': 13,
883 _OPERATION_NAME
= 'ChannelCollectionsContent'
884 _ENTRY_KIND
= 'collection'
885 _EDGE_KIND
= 'CollectionsItemEdge'
886 _NODE_KIND
= 'Collection'
889 def _make_variables(channel_name
):
891 'ownerLogin': channel_name
,
895 def _extract_entry(node
):
896 assert isinstance(node
, dict)
897 collection_id
= node
.get('id')
898 if not collection_id
:
901 '_type': 'url_transparent',
902 'ie_key': TwitchCollectionIE
.ie_key(),
904 'url': 'https://www.twitch.tv/collections/%s' % collection_id
,
905 'title': node
.get('title'),
906 'thumbnail': node
.get('thumbnailURL'),
907 'duration': float_or_none(node
.get('lengthSeconds')),
908 'timestamp': unified_timestamp(node
.get('updatedAt')),
909 'view_count': int_or_none(node
.get('viewCount')),
912 def _real_extract(self
, url
):
913 channel_name
= self
._match
_id
(url
)
914 return self
.playlist_result(
915 self
._entries
(channel_name
), playlist_id
=channel_name
,
916 playlist_title
='%s - Collections' % channel_name
)
919 class TwitchStreamIE(TwitchBaseIE
):
920 IE_NAME
= 'twitch:stream'
921 _VALID_URL
= r
'''(?x)
924 (?:(?:www|go|m)\.)?twitch\.tv/|
925 player\.twitch\.tv/\?.*?\bchannel=
929 _M3U8_PATH
= 'api/channel/hls'
932 'url': 'http://www.twitch.tv/shroomztv',
935 'display_id': 'shroomztv',
937 'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
938 'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV',
940 'timestamp': 1421928037,
941 'upload_date': '20150122',
942 'uploader': 'ShroomzTV',
943 'uploader_id': 'shroomztv',
948 'skip_download': True,
950 'skip': 'User does not exist',
952 'url': 'http://www.twitch.tv/miracle_doto#profile-0',
953 'only_matching': True,
955 'url': 'https://player.twitch.tv/?channel=lotsofs',
956 'only_matching': True,
958 'url': 'https://go.twitch.tv/food',
959 'only_matching': True,
961 'url': 'https://m.twitch.tv/food',
962 'only_matching': True,
964 'url': 'https://www.twitch.tv/monstercat',
967 'display_id': 'monstercat',
968 'title': 're:Monstercat',
969 'description': 'md5:0945ad625e615bc8f0469396537d87d9',
971 'timestamp': 1677107190,
972 'upload_date': '20230222',
973 'uploader': 'Monstercat',
974 'uploader_id': 'monstercat',
975 'live_status': 'is_live',
976 'thumbnail': 're:https://.*.jpg',
980 'skip_download': 'Livestream',
985 def suitable(cls
, url
):
987 if any(ie
.suitable(url
) for ie
in (
992 TwitchVideosCollectionsIE
,
994 else super(TwitchStreamIE
, cls
).suitable(url
))
996 def _real_extract(self
, url
):
997 channel_name
= self
._match
_id
(url
).lower()
999 gql
= self
._download
_gql
(
1001 'operationName': 'StreamMetadata',
1002 'variables': {'channelLogin': channel_name}
,
1004 'operationName': 'ComscoreStreamingQuery',
1006 'channel': channel_name
,
1010 'isVodOrCollection': False,
1014 'operationName': 'VideoPreviewOverlay',
1015 'variables': {'login': channel_name}
,
1017 'Downloading stream GraphQL')
1019 user
= gql
[0]['data']['user']
1022 raise ExtractorError(
1023 '%s does not exist' % channel_name
, expected
=True)
1025 stream
= user
['stream']
1028 raise UserNotLive(video_id
=channel_name
)
1030 access_token
= self
._download
_access
_token
(
1031 channel_name
, 'stream', 'channelName')
1033 stream_id
= stream
.get('id') or channel_name
1034 formats
= self
._extract
_twitch
_m
3u8_formats
(
1035 channel_name
, access_token
['value'], access_token
['signature'])
1036 self
._prefer
_source
(formats
)
1038 view_count
= stream
.get('viewers')
1039 timestamp
= unified_timestamp(stream
.get('createdAt'))
1041 sq_user
= try_get(gql
, lambda x
: x
[1]['data']['user'], dict) or {}
1042 uploader
= sq_user
.get('displayName')
1043 description
= try_get(
1044 sq_user
, lambda x
: x
['broadcastSettings']['title'], compat_str
)
1046 thumbnail
= url_or_none(try_get(
1047 gql
, lambda x
: x
[2]['data']['user']['stream']['previewImageURL'],
1050 title
= uploader
or channel_name
1051 stream_type
= stream
.get('type')
1052 if stream_type
in ['rerun', 'live']:
1053 title
+= ' (%s)' % stream_type
1057 'display_id': channel_name
,
1059 'description': description
,
1060 'thumbnails': self
._get
_thumbnails
(thumbnail
),
1061 'uploader': uploader
,
1062 'uploader_id': channel_name
,
1063 'timestamp': timestamp
,
1064 'view_count': view_count
,
1066 'is_live': stream_type
== 'live',
1070 class TwitchClipsIE(TwitchBaseIE
):
1071 IE_NAME
= 'twitch:clips'
1072 _VALID_URL
= r
'''(?x)
1075 clips\.twitch\.tv/(?:embed\?.*?\bclip=|(?:[^/]+/)*)|
1076 (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/)?clip/
1082 'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',
1083 'md5': '761769e1eafce0ffebfb4089cb3847cd',
1086 'display_id': 'FaintLightGullWholeWheat',
1088 'title': 'EA Play 2016 Live from the Novo Theatre',
1089 'thumbnail': r
're:^https?://.*\.jpg',
1090 'timestamp': 1465767393,
1091 'upload_date': '20160612',
1093 'uploader': 'stereotype_',
1094 'uploader_id': '43566419',
1098 'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy',
1099 'only_matching': True,
1101 'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan',
1102 'only_matching': True,
1104 'url': 'https://clips.twitch.tv/embed?clip=InquisitiveBreakableYogurtJebaited',
1105 'only_matching': True,
1107 'url': 'https://m.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
1108 'only_matching': True,
1110 'url': 'https://go.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
1111 'only_matching': True,
1113 'url': 'https://m.twitch.tv/clip/FaintLightGullWholeWheat',
1114 'only_matching': True,
1117 def _real_extract(self
, url
):
1118 video_id
= self
._match
_id
(url
)
1120 clip
= self
._download
_gql
(
1122 'operationName': 'VideoAccessToken_Clip',
1127 'Downloading clip access token GraphQL')[0]['data']['clip']
1130 raise ExtractorError(
1131 'This clip is no longer available', expected
=True)
1134 'sig': clip
['playbackAccessToken']['signature'],
1135 'token': clip
['playbackAccessToken']['value'],
1138 data
= self
._download
_base
_gql
(
1152 tiny: thumbnailURL(width: 86, height: 45)
1153 small: thumbnailURL(width: 260, height: 147)
1154 medium: thumbnailURL(width: 480, height: 272)
1163 }''' % video_id
}, 'Downloading clip GraphQL', fatal
=False)
1166 clip
= try_get(data
, lambda x
: x
['data']['clip'], dict) or clip
1169 for option
in clip
.get('videoQualities', []):
1170 if not isinstance(option
, dict):
1172 source
= url_or_none(option
.get('sourceURL'))
1176 'url': update_url_query(source
, access_query
),
1177 'format_id': option
.get('quality'),
1178 'height': int_or_none(option
.get('quality')),
1179 'fps': int_or_none(option
.get('frameRate')),
1183 for thumbnail_id
in ('tiny', 'small', 'medium'):
1184 thumbnail_url
= clip
.get(thumbnail_id
)
1185 if not thumbnail_url
:
1189 'url': thumbnail_url
,
1191 mobj
= re
.search(r
'-(\d+)x(\d+)\.', thumbnail_url
)
1194 'height': int(mobj
.group(2)),
1195 'width': int(mobj
.group(1)),
1197 thumbnails
.append(thumb
)
1199 old_id
= self
._search
_regex
(r
'%7C(\d+)(?:-\d+)?.mp4', formats
[-1]['url'], 'old id', default
=None)
1202 'id': clip
.get('id') or video_id
,
1203 '_old_archive_ids': [make_archive_id(self
, old_id
)] if old_id
else None,
1204 'display_id': video_id
,
1205 'title': clip
.get('title'),
1207 'duration': int_or_none(clip
.get('durationSeconds')),
1208 'view_count': int_or_none(clip
.get('viewCount')),
1209 'timestamp': unified_timestamp(clip
.get('createdAt')),
1210 'thumbnails': thumbnails
,
1211 'creator': try_get(clip
, lambda x
: x
['broadcaster']['displayName'], compat_str
),
1212 'uploader': try_get(clip
, lambda x
: x
['curator']['displayName'], compat_str
),
1213 'uploader_id': try_get(clip
, lambda x
: x
['curator']['id'], compat_str
),