7 from .common
import InfoExtractor
11 compat_urllib_parse_urlencode
,
12 compat_urllib_parse_urlparse
,
37 class TwitchBaseIE(InfoExtractor
):
38 _VALID_URL_BASE
= r
'https?://(?:(?:www|go|m)\.)?twitch\.tv'
40 _API_BASE
= 'https://api.twitch.tv'
41 _USHER_BASE
= 'https://usher.ttvnw.net'
42 _LOGIN_FORM_URL
= 'https://www.twitch.tv/login'
43 _LOGIN_POST_URL
= 'https://passport.twitch.tv/login'
44 _CLIENT_ID
= 'kimne78kx3ncx6brgo4mv6wki5h1ko'
45 _NETRC_MACHINE
= 'twitch'
48 'CollectionSideBar': '27111f1b382effad0b6def325caef1909c733fe6a4fbabf54f8d491ef2cf2f14',
49 'FilterableVideoTower_Videos': 'a937f1d22e269e39a03b509f65a7490f9fc247d7f83d6ac1421523e3b68042cb',
50 'ClipsCards__User': 'b73ad2bfaecfd30a9e6c28fada15bd97032c83ec77a0440766a56fe0bd632777',
51 'ChannelCollectionsContent': '07e3691a1bad77a36aba590c351180439a40baefc1c275356f40fc7082419a84',
52 'StreamMetadata': '1c719a40e481453e5c48d9bb585d971b8b372f8ebb105b17076722264dfa5b3e',
53 'ComscoreStreamingQuery': 'e1edae8122517d013405f237ffcc124515dc6ded82480a88daef69c83b53ac01',
54 'VideoAccessToken_Clip': '36b89d2507fce29e5ca551df756d27c1cfe079e2609642b4390aa4c35796eb11',
55 'VideoPreviewOverlay': '3006e77e51b128d838fa4e835723ca4dc9a05c5efd4466c1085215c6e437e65c',
56 'VideoMetadata': '226edb3e692509f727fd56821f5653c05740242c82b0388883e0c0e75dcbf687',
57 'VideoPlayer_ChapterSelectButtonVideo': '8d2793384aac3773beab5e59bd5d6f585aedb923d292800119e03d40cd0f9b41',
58 'VideoPlayer_VODSeekbarPreviewVideo': '07e99e4d56c5a7c67117a154777b0baf85a5ffefa393b213f4bc712ccaf85dd6',
61 def _perform_login(self
, username
, password
):
64 'Unable to login. Twitch said: %s' % message
, expected
=True)
66 def login_step(page
, urlh
, note
, data
):
67 form
= self
._hidden
_inputs
(page
)
70 page_url
= urlh
.geturl()
71 post_url
= self
._search
_regex
(
72 r
'<form[^>]+action=(["\'])(?P
<url
>.+?
)\
1', page,
73 'post url
', default=self._LOGIN_POST_URL, group='url
')
74 post_url = urljoin(page_url, post_url)
78 'Origin
': 'https
://www
.twitch
.tv
',
79 'Content
-Type
': 'text
/plain
;charset
=UTF
-8',
82 response = self._download_json(
83 post_url, None, note, data=json.dumps(form).encode(),
84 headers=headers, expected_status=400)
85 error = dict_get(response, ('error
', 'error_description
', 'error_code
'))
89 if 'Authenticated successfully
' in response.get('message
', ''):
92 redirect_url = urljoin(
94 response.get('redirect
') or response['redirect_path
'])
95 return self._download_webpage_handle(
96 redirect_url, None, 'Downloading login redirect page
',
99 login_page, handle = self._download_webpage_handle(
100 self._LOGIN_FORM_URL, None, 'Downloading login page
')
102 # Some TOR nodes and public proxies are blocked completely
103 if 'blacklist_message
' in login_page:
104 fail(clean_html(login_page))
106 redirect_page, handle = login_step(
107 login_page, handle, 'Logging
in', {
108 'username
': username,
109 'password
': password,
110 'client_id
': self._CLIENT_ID,
114 if not redirect_page:
117 if re.search(r'(?i
)<form
[^
>]+id="two-factor-submit"', redirect_page) is not None:
118 # TODO: Add mechanism to request an SMS or phone call
119 tfa_token = self._get_tfa_info('two
-factor authentication token
')
120 login_step(redirect_page, handle, 'Submitting TFA token
', {
121 'authy_token
': tfa_token,
122 'remember_2fa
': 'true
',
125 def _prefer_source(self, formats):
127 source = next(f for f in formats if f['format_id
'] == 'Source
')
128 source['quality
'] = 10
129 except StopIteration:
131 if '/chunked
/' in f['url
']:
134 'format_note
': 'Source
',
137 def _download_base_gql(self, video_id, ops, note, fatal=True):
139 'Content
-Type
': 'text
/plain
;charset
=UTF
-8',
140 'Client
-ID
': self._CLIENT_ID,
142 gql_auth = self._get_cookies('https
://gql
.twitch
.tv
').get('auth
-token
')
144 headers['Authorization
'] = 'OAuth
' + gql_auth.value
145 return self._download_json(
146 'https
://gql
.twitch
.tv
/gql
', video_id, note,
147 data=json.dumps(ops).encode(),
148 headers=headers, fatal=fatal)
150 def _download_gql(self, video_id, ops, note, fatal=True):
155 'sha256Hash
': self._OPERATION_HASHES[op['operationName
']],
158 return self._download_base_gql(video_id, ops, note)
160 def _download_access_token(self, video_id, token_kind, param_name):
161 method = '%sPlaybackAccessToken
' % token_kind
168 playerBackend: "mediaplayer",
176 }''' % (method, param_name, video_id),
178 return self._download_base_gql(
180 'Downloading
%s access token GraphQL
' % token_kind)['data
'][method]
183 class TwitchVodIE(TwitchBaseIE):
184 IE_NAME = 'twitch
:vod
'
185 _VALID_URL = r'''(?x)
188 (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v(?:ideo)?|videos)/|
189 player\.twitch\.tv/\?.*?\bvideo=v?
195 'url
': 'http
://www
.twitch
.tv
/riotgames
/v
/6528877?t
=5m10s
',
199 'title
': 'LCK Summer Split
- Week
6 Day
1',
200 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
202 'timestamp
': 1435131734,
203 'upload_date
': '20150624',
204 'uploader
': 'Riot Games
',
205 'uploader_id
': 'riotgames
',
212 'title
': 'League of Legends
'
215 'live_status
': 'was_live
',
219 'skip_download
': True,
222 # Untitled broadcast (title is None)
223 'url
': 'http
://www
.twitch
.tv
/belkao_o
/v
/11230755',
227 'title
': 'Untitled Broadcast
',
228 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
230 'timestamp
': 1439746708,
231 'upload_date
': '20150816',
232 'uploader
': 'BelkAO_o
',
233 'uploader_id
': 'belkao_o
',
238 'skip_download
': True,
240 'skip
': 'HTTP Error
404: Not Found
',
242 'url
': 'http
://player
.twitch
.tv
/?t
=5m10s
&video
=v6528877
',
243 'only_matching
': True,
245 'url
': 'https
://www
.twitch
.tv
/videos
/6528877',
246 'only_matching
': True,
248 'url
': 'https
://m
.twitch
.tv
/beagsandjam
/v
/247478721',
249 'only_matching
': True,
251 'url
': 'https
://www
.twitch
.tv
/northernlion
/video
/291940395',
252 'only_matching
': True,
254 'url
': 'https
://player
.twitch
.tv
/?video
=480452374',
255 'only_matching
': True,
257 'url
': 'https
://www
.twitch
.tv
/videos
/635475444',
261 'title
': 'Riot Games
',
263 'uploader
': 'Riot Games
',
264 'uploader_id
': 'riotgames
',
265 'timestamp
': 1590770569,
266 'upload_date
': '20200529',
271 'title
': 'League of Legends
'
276 'title
': 'Legends of Runeterra
'
284 'live_status
': 'was_live
',
285 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
289 'skip_download
': True
292 'note
': 'Storyboards
',
293 'url
': 'https
://www
.twitch
.tv
/videos
/635475444',
298 'title
': 'Riot Games
',
300 'uploader
': 'Riot Games
',
301 'uploader_id
': 'riotgames
',
302 'timestamp
': 1590770569,
303 'upload_date
': '20200529',
308 'title
': 'League of Legends
'
313 'title
': 'Legends of Runeterra
'
321 'live_status
': 'was_live
',
322 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
329 'skip_download
': True
332 'note
': 'VOD
with single chapter
',
333 'url
': 'https
://www
.twitch
.tv
/videos
/1536751224',
337 'title
': 'Porter Robinson Star Guardian Stream Tour
with LilyPichu
',
339 'uploader
': 'Riot Games
',
340 'uploader_id
': 'riotgames
',
341 'timestamp
': 1658267731,
342 'upload_date
': '20220719',
347 'title
': 'League of Legends
'
350 'live_status
': 'was_live
',
351 'thumbnail
': r're
:^https?
://.*\
.jpg$
',
355 'skip_download
': True
357 'expected_warnings
': ['Unable to download JSON metadata
: HTTP Error
403: Forbidden
']
360 def _download_info(self, item_id):
361 data = self._download_gql(
363 'operationName
': 'VideoMetadata
',
369 'operationName
': 'VideoPlayer_ChapterSelectButtonVideo
',
371 'includePrivate
': False,
375 'operationName
': 'VideoPlayer_VODSeekbarPreviewVideo
',
377 'includePrivate
': False,
381 'Downloading stream metadata GraphQL
')
383 video = traverse_obj(data, (0, 'data
', 'video
'))
384 video['moments
'] = traverse_obj(data, (1, 'data
', 'video
', 'moments
', 'edges
', ..., 'node
'))
385 video['storyboard
'] = traverse_obj(data, (2, 'data
', 'video
', 'seekPreviewsURL
'), expected_type=url_or_none)
388 raise ExtractorError(
389 'Video
%s does
not exist
' % item_id, expected=True)
392 def _extract_info(self, info):
393 status = info.get('status
')
394 if status == 'recording
':
396 elif status == 'recorded
':
400 _QUALITIES = ('small
', 'medium
', 'large
')
401 quality_key = qualities(_QUALITIES)
403 preview = info.get('preview
')
404 if isinstance(preview, dict):
405 for thumbnail_id, thumbnail_url in preview.items():
406 thumbnail_url = url_or_none(thumbnail_url)
407 if not thumbnail_url:
409 if thumbnail_id not in _QUALITIES:
412 'url
': thumbnail_url,
413 'preference
': quality_key(thumbnail_id),
417 'title
': info.get('title
') or 'Untitled Broadcast
',
418 'description
': info.get('description
'),
419 'duration
': int_or_none(info.get('length
')),
420 'thumbnails
': thumbnails,
421 'uploader
': info.get('channel
', {}).get('display_name
'),
422 'uploader_id
': info.get('channel
', {}).get('name
'),
423 'timestamp
': parse_iso8601(info.get('recorded_at
')),
424 'view_count
': int_or_none(info.get('views
')),
429 def _extract_chapters(self, info, item_id):
430 if not info.get('moments
'):
431 game = traverse_obj(info, ('game
', 'displayName
'))
433 yield {'title': game}
436 for moment in info['moments
']:
437 start_time = int_or_none(moment.get('positionMilliseconds
'), 1000)
438 duration = int_or_none(moment.get('durationMilliseconds
'), 1000)
439 name = str_or_none(moment.get('description
'))
441 if start_time is None or duration is None:
442 self.report_warning(f'Important chapter information missing
for chapter {name}
', item_id)
445 'start_time
': start_time,
446 'end_time
': start_time + duration,
450 def _extract_info_gql(self, info, item_id):
451 vod_id = info.get('id') or item_id
452 # id backward compatibility for download archives
454 vod_id = 'v
%s' % vod_id
455 thumbnail = url_or_none(info.get('previewThumbnailURL
'))
458 if thumbnail.endswith('/404_processing
_{width}x{height}
.png
'):
459 is_live, thumbnail = True, None
462 for p in ('width
', 'height
'):
463 thumbnail = thumbnail.replace('{%s}
' % p, '0')
467 'title
': info.get('title
') or 'Untitled Broadcast
',
468 'description
': info.get('description
'),
469 'duration
': int_or_none(info.get('lengthSeconds
')),
470 'thumbnail
': thumbnail,
471 'uploader
': try_get(info, lambda x: x['owner
']['displayName
'], compat_str),
472 'uploader_id
': try_get(info, lambda x: x['owner
']['login
'], compat_str),
473 'timestamp
': unified_timestamp(info.get('publishedAt
')),
474 'view_count
': int_or_none(info.get('viewCount
')),
475 'chapters
': list(self._extract_chapters(info, item_id)),
480 def _extract_storyboard(self, item_id, storyboard_json_url, duration):
481 if not duration or not storyboard_json_url:
483 spec = self._download_json(storyboard_json_url, item_id, 'Downloading storyboard metadata JSON
', fatal=False) or []
484 # sort from highest quality to lowest
485 # This makes sb0 the highest-quality format, sb1 - lower, etc which is consistent with youtube sb ordering
486 spec.sort(key=lambda x: int_or_none(x.get('width
')) or 0, reverse=True)
487 base = base_url(storyboard_json_url)
488 for i, s in enumerate(spec):
489 count = int_or_none(s.get('count
'))
490 images = s.get('images
')
491 if not (images and count):
493 fragment_duration = duration / len(images)
495 'format_id
': f'sb{i}
',
496 'format_note
': 'storyboard
',
501 'url
': urljoin(base, images[0]),
502 'width
': int_or_none(s.get('width
')),
503 'height
': int_or_none(s.get('height
')),
504 'fps
': count / duration,
505 'rows
': int_or_none(s.get('rows
')),
506 'columns
': int_or_none(s.get('cols
')),
508 'url
': urljoin(base, path),
509 'duration
': fragment_duration,
510 } for path in images],
513 def _real_extract(self, url):
514 vod_id = self._match_id(url)
516 video = self._download_info(vod_id)
517 info = self._extract_info_gql(video, vod_id)
518 access_token = self._download_access_token(vod_id, 'video
', 'id')
520 formats = self._extract_m3u8_formats(
521 '%s/vod
/%s.m3u8?
%s' % (
522 self._USHER_BASE, vod_id,
523 compat_urllib_parse_urlencode({
524 'allow_source
': 'true
',
525 'allow_audio_only
': 'true
',
526 'allow_spectre
': 'true
',
527 'player
': 'twitchweb
',
528 'playlist_include_framerate
': 'true
',
529 'nauth
': access_token['value
'],
530 'nauthsig
': access_token['signature
'],
532 vod_id, 'mp4
', entry_protocol='m3u8_native
')
534 formats.extend(self._extract_storyboard(vod_id, video.get('storyboard
'), info.get('duration
')))
536 self._prefer_source(formats)
537 info['formats
'] = formats
539 parsed_url = compat_urllib_parse_urlparse(url)
540 query = compat_parse_qs(parsed_url.query)
542 info['start_time
'] = parse_duration(query['t
'][0])
544 if info.get('timestamp
') is not None:
545 info['subtitles
'] = {
547 'url
': update_url_query(
548 'https
://api
.twitch
.tv
/v5
/videos
/%s/comments
' % vod_id, {
549 'client_id
': self._CLIENT_ID,
558 def _make_video_result(node):
559 assert isinstance(node, dict)
560 video_id = node.get('id')
564 '_type
': 'url_transparent
',
565 'ie_key
': TwitchVodIE.ie_key(),
566 'id': 'v
' + video_id,
567 'url
': 'https
://www
.twitch
.tv
/videos
/%s' % video_id,
568 'title
': node.get('title
'),
569 'thumbnail
': node.get('previewThumbnailURL
'),
570 'duration
': float_or_none(node.get('lengthSeconds
')),
571 'view_count
': int_or_none(node.get('viewCount
')),
575 class TwitchCollectionIE(TwitchBaseIE):
576 _VALID_URL = r'https?
://(?
:(?
:www|go|m
)\
.)?twitch\
.tv
/collections
/(?P
<id>[^
/]+)'
579 'url
': 'https
://www
.twitch
.tv
/collections
/wlDCoH0zEBZZbQ
',
581 'id': 'wlDCoH0zEBZZbQ
',
582 'title
': 'Overthrow Nook
, capitalism
for children
',
584 'playlist_mincount
': 13,
587 _OPERATION_NAME = 'CollectionSideBar
'
589 def _real_extract(self, url):
590 collection_id = self._match_id(url)
591 collection = self._download_gql(
593 'operationName
': self._OPERATION_NAME,
594 'variables
': {'collectionID': collection_id},
596 'Downloading collection GraphQL
')[0]['data
']['collection
']
597 title = collection.get('title
')
599 for edge in collection['items
']['edges
']:
600 if not isinstance(edge, dict):
602 node = edge.get('node
')
603 if not isinstance(node, dict):
605 video = _make_video_result(node)
607 entries.append(video)
608 return self.playlist_result(
609 entries, playlist_id=collection_id, playlist_title=title)
612 class TwitchPlaylistBaseIE(TwitchBaseIE):
615 def _entries(self, channel_name, *args):
617 variables_common = self._make_variables(channel_name, *args)
618 entries_key = '%ss' % self._ENTRY_KIND
619 for page_num in itertools.count(1):
620 variables = variables_common.copy()
621 variables['limit
'] = self._PAGE_LIMIT
623 variables['cursor
'] = cursor
624 page = self._download_gql(
626 'operationName
': self._OPERATION_NAME,
627 'variables
': variables,
629 'Downloading
%ss GraphQL page
%s' % (self._NODE_KIND, page_num),
634 page, lambda x: x[0]['data
']['user
'][entries_key]['edges
'], list)
638 if not isinstance(edge, dict):
640 if edge.get('__typename
') != self._EDGE_KIND:
642 node = edge.get('node
')
643 if not isinstance(node, dict):
645 if node.get('__typename
') != self._NODE_KIND:
647 entry = self._extract_entry(node)
649 cursor = edge.get('cursor
')
651 if not cursor or not isinstance(cursor, compat_str):
655 class TwitchVideosIE(TwitchPlaylistBaseIE):
656 _VALID_URL = r'https?
://(?
:(?
:www|go|m
)\
.)?twitch\
.tv
/(?P
<id>[^
/]+)/(?
:videos|profile
)'
659 # All Videos sorted by Date
660 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos?
filter=all
',
663 'title
': 'spamfish
- All Videos
sorted by Date
',
665 'playlist_mincount
': 924,
667 # All Videos sorted by Popular
668 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos?
filter=all
&sort
=views
',
671 'title
': 'spamfish
- All Videos
sorted by Popular
',
673 'playlist_mincount
': 931,
675 # Past Broadcasts sorted by Date
676 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos?
filter=archives
',
679 'title
': 'spamfish
- Past Broadcasts
sorted by Date
',
681 'playlist_mincount
': 27,
683 # Highlights sorted by Date
684 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos?
filter=highlights
',
687 'title
': 'spamfish
- Highlights
sorted by Date
',
689 'playlist_mincount
': 901,
691 # Uploads sorted by Date
692 'url
': 'https
://www
.twitch
.tv
/esl_csgo
/videos?
filter=uploads
&sort
=time
',
695 'title
': 'esl_csgo
- Uploads
sorted by Date
',
697 'playlist_mincount
': 5,
699 # Past Premieres sorted by Date
700 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos?
filter=past_premieres
',
703 'title
': 'spamfish
- Past Premieres
sorted by Date
',
705 'playlist_mincount
': 1,
707 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos
/all
',
708 'only_matching
': True,
710 'url
': 'https
://m
.twitch
.tv
/spamfish
/videos
/all
',
711 'only_matching
': True,
713 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos
',
714 'only_matching
': True,
717 Broadcast = collections.namedtuple('Broadcast
', ['type', 'label
'])
719 _DEFAULT_BROADCAST = Broadcast(None, 'All Videos
')
721 'archives
': Broadcast('ARCHIVE
', 'Past Broadcasts
'),
722 'highlights
': Broadcast('HIGHLIGHT
', 'Highlights
'),
723 'uploads
': Broadcast('UPLOAD
', 'Uploads
'),
724 'past_premieres
': Broadcast('PAST_PREMIERE
', 'Past Premieres
'),
725 'all
': _DEFAULT_BROADCAST,
728 _DEFAULT_SORTED_BY = 'Date
'
730 'time
': _DEFAULT_SORTED_BY,
734 _OPERATION_NAME = 'FilterableVideoTower_Videos
'
735 _ENTRY_KIND = 'video
'
736 _EDGE_KIND = 'VideoEdge
'
740 def suitable(cls, url):
742 if any(ie.suitable(url) for ie in (
744 TwitchVideosCollectionsIE))
745 else super(TwitchVideosIE, cls).suitable(url))
748 def _make_variables(channel_name, broadcast_type, sort):
750 'channelOwnerLogin
': channel_name,
751 'broadcastType
': broadcast_type,
752 'videoSort
': sort.upper(),
756 def _extract_entry(node):
757 return _make_video_result(node)
759 def _real_extract(self, url):
760 channel_name = self._match_id(url)
762 filter = qs.get('filter', ['all
'])[0]
763 sort = qs.get('sort
', ['time
'])[0]
764 broadcast = self._BROADCASTS.get(filter, self._DEFAULT_BROADCAST)
765 return self.playlist_result(
766 self._entries(channel_name, broadcast.type, sort),
767 playlist_id=channel_name,
768 playlist_title='%s - %s sorted by
%s'
769 % (channel_name, broadcast.label,
770 self._SORTED_BY.get(sort, self._DEFAULT_SORTED_BY)))
773 class TwitchVideosClipsIE(TwitchPlaylistBaseIE):
774 _VALID_URL = r'https?
://(?
:(?
:www|go|m
)\
.)?twitch\
.tv
/(?P
<id>[^
/]+)/(?
:clips|videos
/*?
\?.*?
\bfilter
=clips
)'
778 'url
': 'https
://www
.twitch
.tv
/vanillatv
/clips?
filter=clips
&range=all
',
781 'title
': 'vanillatv
- Clips Top All
',
783 'playlist_mincount
': 1,
785 'url
': 'https
://www
.twitch
.tv
/dota2ruhub
/videos?
filter=clips
&range=7d
',
786 'only_matching
': True,
789 Clip = collections.namedtuple('Clip
', ['filter', 'label
'])
791 _DEFAULT_CLIP = Clip('LAST_WEEK
', 'Top
7D
')
793 '24hr
': Clip('LAST_DAY
', 'Top
24H
'),
795 '30d
': Clip('LAST_MONTH
', 'Top
30D
'),
796 'all
': Clip('ALL_TIME
', 'Top All
'),
799 # NB: values other than 20 result in skipped videos
802 _OPERATION_NAME = 'ClipsCards__User
'
804 _EDGE_KIND = 'ClipEdge
'
808 def _make_variables(channel_name, filter):
810 'login
': channel_name,
817 def _extract_entry(node):
818 assert isinstance(node, dict)
819 clip_url = url_or_none(node.get('url
'))
823 '_type
': 'url_transparent
',
824 'ie_key
': TwitchClipsIE.ie_key(),
825 'id': node.get('id'),
827 'title
': node.get('title
'),
828 'thumbnail
': node.get('thumbnailURL
'),
829 'duration
': float_or_none(node.get('durationSeconds
')),
830 'timestamp
': unified_timestamp(node.get('createdAt
')),
831 'view_count
': int_or_none(node.get('viewCount
')),
832 'language
': node.get('language
'),
835 def _real_extract(self, url):
836 channel_name = self._match_id(url)
838 range = qs.get('range', ['7d
'])[0]
839 clip = self._RANGE.get(range, self._DEFAULT_CLIP)
840 return self.playlist_result(
841 self._entries(channel_name, clip.filter),
842 playlist_id=channel_name,
843 playlist_title='%s - Clips
%s' % (channel_name, clip.label))
846 class TwitchVideosCollectionsIE(TwitchPlaylistBaseIE):
847 _VALID_URL = r'https?
://(?
:(?
:www|go|m
)\
.)?twitch\
.tv
/(?P
<id>[^
/]+)/videos
/*?
\?.*?
\bfilter
=collections
'
851 'url
': 'https
://www
.twitch
.tv
/spamfish
/videos?
filter=collections
',
854 'title
': 'spamfish
- Collections
',
856 'playlist_mincount
': 3,
859 _OPERATION_NAME = 'ChannelCollectionsContent
'
860 _ENTRY_KIND = 'collection
'
861 _EDGE_KIND = 'CollectionsItemEdge
'
862 _NODE_KIND = 'Collection
'
865 def _make_variables(channel_name):
867 'ownerLogin
': channel_name,
871 def _extract_entry(node):
872 assert isinstance(node, dict)
873 collection_id = node.get('id')
874 if not collection_id:
877 '_type
': 'url_transparent
',
878 'ie_key
': TwitchCollectionIE.ie_key(),
880 'url
': 'https
://www
.twitch
.tv
/collections
/%s' % collection_id,
881 'title
': node.get('title
'),
882 'thumbnail
': node.get('thumbnailURL
'),
883 'duration
': float_or_none(node.get('lengthSeconds
')),
884 'timestamp
': unified_timestamp(node.get('updatedAt
')),
885 'view_count
': int_or_none(node.get('viewCount
')),
888 def _real_extract(self, url):
889 channel_name = self._match_id(url)
890 return self.playlist_result(
891 self._entries(channel_name), playlist_id=channel_name,
892 playlist_title='%s - Collections
' % channel_name)
895 class TwitchStreamIE(TwitchBaseIE):
896 IE_NAME = 'twitch
:stream
'
897 _VALID_URL = r'''(?x)
900 (?:(?:www|go|m)\.)?twitch\.tv/|
901 player\.twitch\.tv/\?.*?\bchannel=
907 'url
': 'http
://www
.twitch
.tv
/shroomztv
',
910 'display_id
': 'shroomztv
',
912 'title
': 're
:^ShroomzTV
[0-9]{4}
-[0-9]{2}
-[0-9]{2}
[0-9]{2}
:[0-9]{2}$
',
913 'description
': 'H1Z1
- lonewolfing
with ShroomzTV | A3 Battle Royale later
- @ShroomzTV',
915 'timestamp
': 1421928037,
916 'upload_date
': '20150122',
917 'uploader
': 'ShroomzTV
',
918 'uploader_id
': 'shroomztv
',
923 'skip_download
': True,
926 'url
': 'http
://www
.twitch
.tv
/miracle_doto
#profile-0',
927 'only_matching': True,
929 'url': 'https://player.twitch.tv/?channel=lotsofs',
930 'only_matching': True,
932 'url': 'https://go.twitch.tv/food',
933 'only_matching': True,
935 'url': 'https://m.twitch.tv/food',
936 'only_matching': True,
940 def suitable(cls
, url
):
942 if any(ie
.suitable(url
) for ie
in (
947 TwitchVideosCollectionsIE
,
949 else super(TwitchStreamIE
, cls
).suitable(url
))
951 def _real_extract(self
, url
):
952 channel_name
= self
._match
_id
(url
).lower()
954 gql
= self
._download
_gql
(
956 'operationName': 'StreamMetadata',
957 'variables': {'channelLogin': channel_name}
,
959 'operationName': 'ComscoreStreamingQuery',
961 'channel': channel_name
,
965 'isVodOrCollection': False,
969 'operationName': 'VideoPreviewOverlay',
970 'variables': {'login': channel_name}
,
972 'Downloading stream GraphQL')
974 user
= gql
[0]['data']['user']
977 raise ExtractorError(
978 '%s does not exist' % channel_name
, expected
=True)
980 stream
= user
['stream']
983 raise UserNotLive(video_id
=channel_name
)
985 access_token
= self
._download
_access
_token
(
986 channel_name
, 'stream', 'channelName')
987 token
= access_token
['value']
989 stream_id
= stream
.get('id') or channel_name
991 'allow_source': 'true',
992 'allow_audio_only': 'true',
993 'allow_spectre': 'true',
994 'p': random
.randint(1000000, 10000000),
995 'player': 'twitchweb',
996 'playlist_include_framerate': 'true',
997 'segment_preference': '4',
998 'sig': access_token
['signature'].encode('utf-8'),
999 'token': token
.encode('utf-8'),
1001 formats
= self
._extract
_m
3u8_formats
(
1002 '%s/api/channel/hls/%s.m3u8' % (self
._USHER
_BASE
, channel_name
),
1003 stream_id
, 'mp4', query
=query
)
1004 self
._prefer
_source
(formats
)
1006 view_count
= stream
.get('viewers')
1007 timestamp
= unified_timestamp(stream
.get('createdAt'))
1009 sq_user
= try_get(gql
, lambda x
: x
[1]['data']['user'], dict) or {}
1010 uploader
= sq_user
.get('displayName')
1011 description
= try_get(
1012 sq_user
, lambda x
: x
['broadcastSettings']['title'], compat_str
)
1014 thumbnail
= url_or_none(try_get(
1015 gql
, lambda x
: x
[2]['data']['user']['stream']['previewImageURL'],
1018 title
= uploader
or channel_name
1019 stream_type
= stream
.get('type')
1020 if stream_type
in ['rerun', 'live']:
1021 title
+= ' (%s)' % stream_type
1025 'display_id': channel_name
,
1027 'description': description
,
1028 'thumbnail': thumbnail
,
1029 'uploader': uploader
,
1030 'uploader_id': channel_name
,
1031 'timestamp': timestamp
,
1032 'view_count': view_count
,
1034 'is_live': stream_type
== 'live',
1038 class TwitchClipsIE(TwitchBaseIE
):
1039 IE_NAME
= 'twitch:clips'
1040 _VALID_URL
= r
'''(?x)
1043 clips\.twitch\.tv/(?:embed\?.*?\bclip=|(?:[^/]+/)*)|
1044 (?:(?:www|go|m)\.)?twitch\.tv/[^/]+/clip/
1050 'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',
1051 'md5': '761769e1eafce0ffebfb4089cb3847cd',
1054 'display_id': 'FaintLightGullWholeWheat',
1056 'title': 'EA Play 2016 Live from the Novo Theatre',
1057 'thumbnail': r
're:^https?://.*\.jpg',
1058 'timestamp': 1465767393,
1059 'upload_date': '20160612',
1061 'uploader': 'stereotype_',
1062 'uploader_id': '43566419',
1066 'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy',
1067 'only_matching': True,
1069 'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan',
1070 'only_matching': True,
1072 'url': 'https://clips.twitch.tv/embed?clip=InquisitiveBreakableYogurtJebaited',
1073 'only_matching': True,
1075 'url': 'https://m.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
1076 'only_matching': True,
1078 'url': 'https://go.twitch.tv/rossbroadcast/clip/ConfidentBraveHumanChefFrank',
1079 'only_matching': True,
1082 def _real_extract(self
, url
):
1083 video_id
= self
._match
_id
(url
)
1085 clip
= self
._download
_gql
(
1087 'operationName': 'VideoAccessToken_Clip',
1092 'Downloading clip access token GraphQL')[0]['data']['clip']
1095 raise ExtractorError(
1096 'This clip is no longer available', expected
=True)
1099 'sig': clip
['playbackAccessToken']['signature'],
1100 'token': clip
['playbackAccessToken']['value'],
1103 data
= self
._download
_base
_gql
(
1117 tiny: thumbnailURL(width: 86, height: 45)
1118 small: thumbnailURL(width: 260, height: 147)
1119 medium: thumbnailURL(width: 480, height: 272)
1128 }''' % video_id
}, 'Downloading clip GraphQL', fatal
=False)
1131 clip
= try_get(data
, lambda x
: x
['data']['clip'], dict) or clip
1134 for option
in clip
.get('videoQualities', []):
1135 if not isinstance(option
, dict):
1137 source
= url_or_none(option
.get('sourceURL'))
1141 'url': update_url_query(source
, access_query
),
1142 'format_id': option
.get('quality'),
1143 'height': int_or_none(option
.get('quality')),
1144 'fps': int_or_none(option
.get('frameRate')),
1148 for thumbnail_id
in ('tiny', 'small', 'medium'):
1149 thumbnail_url
= clip
.get(thumbnail_id
)
1150 if not thumbnail_url
:
1154 'url': thumbnail_url
,
1156 mobj
= re
.search(r
'-(\d+)x(\d+)\.', thumbnail_url
)
1159 'height': int(mobj
.group(2)),
1160 'width': int(mobj
.group(1)),
1162 thumbnails
.append(thumb
)
1164 old_id
= self
._search
_regex
(r
'%7C(\d+)(?:-\d+)?.mp4', formats
[-1]['url'], 'old id', default
=None)
1167 'id': clip
.get('id') or video_id
,
1168 '_old_archive_ids': [make_archive_id(self
, old_id
)] if old_id
else None,
1169 'display_id': video_id
,
1170 'title': clip
.get('title'),
1172 'duration': int_or_none(clip
.get('durationSeconds')),
1173 'view_count': int_or_none(clip
.get('viewCount')),
1174 'timestamp': unified_timestamp(clip
.get('createdAt')),
1175 'thumbnails': thumbnails
,
1176 'creator': try_get(clip
, lambda x
: x
['broadcaster']['displayName'], compat_str
),
1177 'uploader': try_get(clip
, lambda x
: x
['curator']['displayName'], compat_str
),
1178 'uploader_id': try_get(clip
, lambda x
: x
['curator']['id'], compat_str
),