5 from .common
import InfoExtractor
6 from .periscope
import PeriscopeBaseIE
, PeriscopeIE
7 from ..compat
import functools
# isort: split
10 compat_urllib_parse_unquote
,
11 compat_urllib_parse_urlparse
,
32 class TwitterBaseIE(InfoExtractor
):
33 _API_BASE
= 'https://api.twitter.com/1.1/'
34 _GRAPHQL_API_BASE
= 'https://twitter.com/i/api/graphql/'
36 'AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA': None,
37 'AAAAAAAAAAAAAAAAAAAAAPYXBAAAAAAACLXUNDekMxqa8h%2F40K4moUkGsoc%3DTYfbDKbT3jJPCEVnMYqilB28NHfOPqkca3qaAxGfsyKCs0wRbw': None,
39 _BASE_REGEX
= r
'https?://(?:(?:www|m(?:obile)?)\.)?(?:twitter\.com|twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid\.onion)/'
41 def _extract_variant_formats(self
, variant
, video_id
):
42 variant_url
= variant
.get('url')
45 elif '.m3u8' in variant_url
:
46 return self
._extract
_m
3u8_formats
_and
_subtitles
(
47 variant_url
, video_id
, 'mp4', 'm3u8_native',
48 m3u8_id
='hls', fatal
=False)
50 tbr
= int_or_none(dict_get(variant
, ('bitrate', 'bit_rate')), 1000) or None
53 'format_id': 'http' + ('-%d' % tbr
if tbr
else ''),
56 self
._search
_dimensions
_in
_video
_url
(f
, variant_url
)
59 def _extract_formats_from_vmap_url(self
, vmap_url
, video_id
):
60 vmap_url
= url_or_none(vmap_url
)
63 vmap_data
= self
._download
_xml
(vmap_url
, video_id
)
67 for video_variant
in vmap_data
.findall('.//{http://twitter.com/schema/videoVMapV2.xsd}videoVariant'):
68 video_variant
.attrib
['url'] = compat_urllib_parse_unquote(
69 video_variant
.attrib
['url'])
70 urls
.append(video_variant
.attrib
['url'])
71 fmts
, subs
= self
._extract
_variant
_formats
(
72 video_variant
.attrib
, video_id
)
74 subtitles
= self
._merge
_subtitles
(subtitles
, subs
)
75 video_url
= strip_or_none(xpath_text(vmap_data
, './/MediaFile'))
76 if video_url
not in urls
:
77 fmts
, subs
= self
._extract
_variant
_formats
({'url': video_url}
, video_id
)
79 subtitles
= self
._merge
_subtitles
(subtitles
, subs
)
80 return formats
, subtitles
83 def _search_dimensions_in_video_url(a_format
, video_url
):
84 m
= re
.search(r
'/(?P<width>\d+)x(?P<height>\d+)/', video_url
)
87 'width': int(m
.group('width')),
88 'height': int(m
.group('height')),
91 @functools.cached_property
92 def is_logged_in(self
):
93 return bool(self
._get
_cookies
(self
._API
_BASE
).get('auth_token'))
95 def _call_api(self
, path
, video_id
, query
={}, graphql
=False):
96 cookies
= self
._get
_cookies
(self
._API
_BASE
)
99 csrf_cookie
= cookies
.get('ct0')
101 headers
['x-csrf-token'] = csrf_cookie
.value
103 if self
.is_logged_in
:
105 'x-twitter-auth-type': 'OAuth2Session',
106 'x-twitter-client-language': 'en',
107 'x-twitter-active-user': 'yes',
110 result
, last_error
= None, None
111 for bearer_token
in self
._TOKENS
:
112 headers
['Authorization'] = f
'Bearer {bearer_token}'
114 if not self
.is_logged_in
:
115 if not self
._TOKENS
[bearer_token
]:
116 headers
.pop('x-guest-token', None)
117 guest_token_response
= self
._download
_json
(
118 self
._API
_BASE
+ 'guest/activate.json', video_id
,
119 'Downloading guest token', data
=b
'', headers
=headers
)
121 self
._TOKENS
[bearer_token
] = guest_token_response
.get('guest_token')
122 if not self
._TOKENS
[bearer_token
]:
123 raise ExtractorError('Could not retrieve guest token')
124 headers
['x-guest-token'] = self
._TOKENS
[bearer_token
]
127 allowed_status
= {400, 403, 404}
if graphql
else {403}
128 result
= self
._download
_json
(
129 (self
._GRAPHQL
_API
_BASE
if graphql
else self
._API
_BASE
) + path
,
130 video_id
, headers
=headers
, query
=query
, expected_status
=allowed_status
)
133 except ExtractorError
as e
:
136 elif not isinstance(e
.cause
, urllib
.error
.HTTPError
) or e
.cause
.code
!= 404:
140 'Twitter API gave 404 response, retrying with deprecated token. '
141 'Only one media item can be extracted')
143 if result
.get('errors'):
144 error_message
= ', '.join(set(traverse_obj(
145 result
, ('errors', ..., 'message'), expected_type
=str))) or 'Unknown error'
146 raise ExtractorError(f
'Error(s) while querying api: {error_message}', expected
=True)
148 assert result
is not None
151 def _build_graphql_query(self
, media_id
):
152 raise NotImplementedError('Method must be implemented to support GraphQL')
154 def _call_graphql_api(self
, endpoint
, media_id
):
155 data
= self
._build
_graphql
_query
(media_id
)
156 query
= {key: json.dumps(value, separators=(',', ':')) for key, value in data.items()}
157 return traverse_obj(self
._call
_api
(endpoint
, media_id
, query
=query
, graphql
=True), 'data')
160 class TwitterCardIE(InfoExtractor
):
161 IE_NAME
= 'twitter:card'
162 _VALID_URL
= TwitterBaseIE
._BASE
_REGEX
+ r
'i/(?:cards/tfw/v1|videos(?:/tweet)?)/(?P<id>\d+)'
165 'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
166 # MD5 checksums are different in different places
168 'id': '560070131976392705',
170 'title': "Twitter - You can now shoot, edit and share video on Twitter. Capture life's most moving moments from your perspective.",
171 'description': 'md5:18d3e24bb4f6e5007487dd546e53bd96',
172 'uploader': 'Twitter',
173 'uploader_id': 'Twitter',
174 'thumbnail': r
're:^https?://.*\.jpg',
176 'timestamp': 1422366112,
177 'upload_date': '20150127',
179 'comment_count': int,
183 'display_id': '560070183650213889',
184 'uploader_url': 'https://twitter.com/Twitter',
188 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
189 'md5': '7137eca597f72b9abbe61e5ae0161399',
191 'id': '623160978427936768',
193 'title': "NASA - Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video.",
194 'description': "Fly over Pluto's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video. https://t.co/BJYgOjSeGA",
196 'uploader_id': 'NASA',
197 'timestamp': 1437408129,
198 'upload_date': '20150720',
199 'uploader_url': 'https://twitter.com/NASA',
201 'comment_count': int,
204 'tags': ['PlutoFlyby'],
206 'params': {'format': '[protocol=https]'}
209 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
210 'md5': 'b6d9683dd3f48e340ded81c0e917ad46',
214 'title': 'Ubuntu 11.10 Overview',
215 'description': 'md5:a831e97fa384863d6e26ce48d1c43376',
216 'upload_date': '20111013',
217 'uploader': 'OMG! UBUNTU!',
218 'uploader_id': 'omgubuntu',
219 'channel_url': 'https://www.youtube.com/channel/UCIiSwcm9xiFb3Y4wjzR41eQ',
220 'channel_id': 'UCIiSwcm9xiFb3Y4wjzR41eQ',
221 'channel_follower_count': int,
222 'chapters': 'count:8',
223 'uploader_url': 'http://www.youtube.com/user/omgubuntu',
225 'categories': ['Film & Animation'],
227 'comment_count': int,
228 'availability': 'public',
230 'thumbnail': 'https://i.ytimg.com/vi/dq4Oj5quskI/maxresdefault.jpg',
233 'channel': 'OMG! UBUNTU!',
234 'playable_in_embed': True,
236 'add_ie': ['Youtube'],
239 'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568',
243 'upload_date': '20151113',
244 'uploader_id': '1189339351084113920',
245 'uploader': 'ArsenalTerje',
246 'title': 'Vine by ArsenalTerje',
247 'timestamp': 1447451307,
248 'alt_title': 'Vine by ArsenalTerje',
249 'comment_count': int,
251 'thumbnail': r
're:^https?://[^?#]+\.jpg',
256 'params': {'skip_download': 'm3u8'}
,
259 'url': 'https://twitter.com/i/videos/tweet/705235433198714880',
260 'md5': '884812a2adc8aaf6fe52b15ccbfa3b88',
262 'id': '705235433198714880',
264 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
265 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
266 'uploader': 'Brent Yarina',
267 'uploader_id': 'BTNBrentYarina',
268 'timestamp': 1456976204,
269 'upload_date': '20160303',
271 'skip': 'This content is no longer available.',
274 'url': 'https://twitter.com/i/videos/752274308186120192',
275 'only_matching': True,
279 def _real_extract(self
, url
):
280 status_id
= self
._match
_id
(url
)
281 return self
.url_result(
282 'https://twitter.com/statuses/' + status_id
,
283 TwitterIE
.ie_key(), status_id
)
286 class TwitterIE(TwitterBaseIE
):
288 _VALID_URL
= TwitterBaseIE
._BASE
_REGEX
+ r
'(?:(?:i/web|[^/]+)/status|statuses)/(?P<id>\d+)'
291 'url': 'https://twitter.com/freethenipple/status/643211948184596480',
293 'id': '643211870443208704',
294 'display_id': '643211948184596480',
296 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
297 'thumbnail': r
're:^https?://.*\.jpg',
298 'description': 'FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ',
299 'uploader': 'FREE THE NIPPLE',
300 'uploader_id': 'freethenipple',
302 'timestamp': 1442188653,
303 'upload_date': '20150913',
304 'uploader_url': 'https://twitter.com/freethenipple',
305 'comment_count': int,
312 'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
313 'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
315 'id': '657991469417025536',
317 'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
318 'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
319 'thumbnail': r
're:^https?://.*\.png',
321 'uploader_id': 'giphz',
323 'expected_warnings': ['height', 'width'],
324 'skip': 'Account suspended',
326 'url': 'https://twitter.com/starwars/status/665052190608723968',
328 'id': '665052190608723968',
329 'display_id': '665052190608723968',
331 'title': 'md5:3f57ab5d35116537a2ae7345cd0060d8',
332 'description': 'A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens. https://t.co/OkSqT2fjWJ',
333 'uploader_id': 'starwars',
334 'uploader': r
're:Star Wars.*',
335 'timestamp': 1447395772,
336 'upload_date': '20151113',
337 'uploader_url': 'https://twitter.com/starwars',
338 'comment_count': int,
341 'tags': ['TV', 'StarWars', 'TheForceAwakens'],
345 'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880',
347 'id': '705235433198714880',
349 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
350 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
351 'uploader_id': 'BTNBrentYarina',
352 'uploader': 'Brent Yarina',
353 'timestamp': 1456976204,
354 'upload_date': '20160303',
355 'uploader_url': 'https://twitter.com/BTNBrentYarina',
356 'comment_count': int,
363 # The same video as https://twitter.com/i/videos/tweet/705235433198714880
364 # Test case of TwitterCardIE
365 'skip_download': True,
368 'url': 'https://twitter.com/jaydingeer/status/700207533655363584',
370 'id': '700207414000242688',
371 'display_id': '700207533655363584',
373 'title': 'jaydin donte geer - BEAT PROD: @suhmeduh #Damndaniel',
374 'description': 'BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ',
375 'thumbnail': r
're:^https?://.*\.jpg',
376 'uploader': 'jaydin donte geer',
377 'uploader_id': 'jaydingeer',
379 'timestamp': 1455777459,
380 'upload_date': '20160218',
381 'uploader_url': 'https://twitter.com/jaydingeer',
382 'comment_count': int,
385 'tags': ['Damndaniel'],
389 'url': 'https://twitter.com/Filmdrunk/status/713801302971588609',
390 'md5': '89a15ed345d13b86e9a5a5e051fa308a',
394 'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン',
395 'uploader': 'TAKUMA',
396 'uploader_id': '1004126642786242560',
397 'timestamp': 1402826626,
398 'upload_date': '20140615',
399 'thumbnail': r
're:^https?://.*\.jpg',
400 'alt_title': 'Vine by TAKUMA',
401 'comment_count': int,
408 'url': 'https://twitter.com/captainamerica/status/719944021058060289',
410 'id': '717462543795523584',
411 'display_id': '719944021058060289',
413 'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.',
414 'description': '@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI',
415 'uploader_id': 'CaptainAmerica',
416 'uploader': 'Captain America',
418 'timestamp': 1460483005,
419 'upload_date': '20160412',
420 'uploader_url': 'https://twitter.com/CaptainAmerica',
421 'thumbnail': r
're:^https?://.*\.jpg',
422 'comment_count': int,
429 'url': 'https://twitter.com/OPP_HSD/status/779210622571536384',
431 'id': '1zqKVVlkqLaKB',
433 'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence',
434 'upload_date': '20160923',
435 'uploader_id': '1PmKqpJdOJQoY',
436 'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police',
437 'timestamp': 1474613214,
438 'thumbnail': r
're:^https?://.*\.jpg',
440 'add_ie': ['Periscope'],
442 # has mp4 formats via mobile API
443 'url': 'https://twitter.com/news_al3alm/status/852138619213144067',
445 'id': '852138619213144067',
447 'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة',
448 'description': 'كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN',
449 'uploader': 'عالم الأخبار',
450 'uploader_id': 'news_al3alm',
452 'timestamp': 1492000653,
453 'upload_date': '20170412',
455 'skip': 'Account suspended',
457 'url': 'https://twitter.com/i/web/status/910031516746514432',
459 'id': '910030238373089285',
460 'display_id': '910031516746514432',
462 'title': 'Préfet de Guadeloupe - [Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre.',
463 'thumbnail': r
're:^https?://.*\.jpg',
464 'description': '[Direct] #Maria Le centre se trouve actuellement au sud de Basse-Terre. Restez confinés. Réfugiez-vous dans la pièce la + sûre. https://t.co/mwx01Rs4lo',
465 'uploader': 'Préfet de Guadeloupe',
466 'uploader_id': 'Prefet971',
468 'timestamp': 1505803395,
469 'upload_date': '20170919',
470 'uploader_url': 'https://twitter.com/Prefet971',
471 'comment_count': int,
478 'skip_download': True, # requires ffmpeg
481 # card via api.twitter.com/1.1/videos/tweet/config
482 'url': 'https://twitter.com/LisPower1/status/1001551623938805763',
484 'id': '1001551417340022785',
485 'display_id': '1001551623938805763',
487 'title': 're:.*?Shep is on a roll today.*?',
488 'thumbnail': r
're:^https?://.*\.jpg',
489 'description': 'md5:37b9f2ff31720cef23b2bd42ee8a0f09',
490 'uploader': 'Lis Power',
491 'uploader_id': 'LisPower1',
493 'timestamp': 1527623489,
494 'upload_date': '20180529',
495 'uploader_url': 'https://twitter.com/LisPower1',
496 'comment_count': int,
503 'skip_download': True, # requires ffmpeg
506 'url': 'https://twitter.com/foobar/status/1087791357756956680',
508 'id': '1087791272830607360',
509 'display_id': '1087791357756956680',
511 'title': 'Twitter - A new is coming. Some of you got an opt-in to try it now. Check out the emoji button, quick keyboard shortcuts, upgraded trends, advanced search, and more. Let us know your thoughts!',
512 'thumbnail': r
're:^https?://.*\.jpg',
513 'description': 'md5:6dfd341a3310fb97d80d2bf7145df976',
514 'uploader': 'Twitter',
515 'uploader_id': 'Twitter',
517 'timestamp': 1548184644,
518 'upload_date': '20190122',
519 'uploader_url': 'https://twitter.com/Twitter',
520 'comment_count': int,
527 # not available in Periscope
528 'url': 'https://twitter.com/ViviEducation/status/1136534865145286656',
530 'id': '1vOGwqejwoWxB',
532 'title': 'Vivi - Vivi founder @lior_rauchy announcing our new student feedback tool live at @EduTECH_AU #EduTECH2019',
534 'uploader_id': '1eVjYOLGkGrQL',
535 'thumbnail': r
're:^https?://.*\.jpg',
536 'tags': ['EduTECH2019'],
539 'add_ie': ['TwitterBroadcast'],
542 'url': 'https://twitter.com/BrooklynNets/status/1349794411333394432?s=20',
544 'id': '1349774757969989634',
545 'display_id': '1349794411333394432',
547 'title': 'md5:d1c4941658e4caaa6cb579260d85dcba',
548 'thumbnail': r
're:^https?://.*\.jpg',
549 'description': 'md5:71ead15ec44cee55071547d6447c6a3e',
550 'uploader': 'Brooklyn Nets',
551 'uploader_id': 'BrooklynNets',
553 'timestamp': 1610651040,
554 'upload_date': '20210114',
555 'uploader_url': 'https://twitter.com/BrooklynNets',
556 'comment_count': int,
563 'skip_download': True,
566 'url': 'https://twitter.com/oshtru/status/1577855540407197696',
568 'id': '1577855447914409984',
569 'display_id': '1577855540407197696',
571 'title': 'oshtru \U0001faac\U0001f47d - gm \u2728\ufe0f now I can post image and video. nice update.',
572 'description': 'gm \u2728\ufe0f now I can post image and video. nice update. https://t.co/cG7XgiINOm',
573 'upload_date': '20221006',
574 'uploader': 'oshtru \U0001faac\U0001f47d',
575 'uploader_id': 'oshtru',
576 'uploader_url': 'https://twitter.com/oshtru',
577 'thumbnail': r
're:^https?://.*\.jpg',
579 'timestamp': 1665025050,
580 'comment_count': int,
586 'params': {'skip_download': True}
,
588 'url': 'https://twitter.com/UltimaShadowX/status/1577719286659006464',
590 'id': '1577719286659006464',
591 'title': 'Ultima | #\u0432\u029f\u043c - Test',
592 'description': 'Test https://t.co/Y3KEZD7Dad',
593 'uploader': 'Ultima | #\u0432\u029f\u043c',
594 'uploader_id': 'UltimaShadowX',
595 'uploader_url': 'https://twitter.com/UltimaShadowX',
596 'upload_date': '20221005',
597 'timestamp': 1664992565,
598 'comment_count': int,
605 'params': {'skip_download': True}
,
607 'url': 'https://twitter.com/MesoMax919/status/1575560063510810624',
609 'id': '1575559336759263233',
610 'display_id': '1575560063510810624',
612 'title': 'md5:eec26382babd0f7c18f041db8ae1c9c9',
613 'thumbnail': r
're:^https?://.*\.jpg',
614 'description': 'md5:95aea692fda36a12081b9629b02daa92',
615 'uploader': 'Max Olson',
616 'uploader_id': 'MesoMax919',
617 'uploader_url': 'https://twitter.com/MesoMax919',
619 'timestamp': 1664477766,
620 'upload_date': '20220929',
621 'comment_count': int,
624 'tags': ['HurricaneIan'],
628 # Adult content, uses old token
629 # Fails if not logged in (GraphQL)
630 'url': 'https://twitter.com/Rizdraws/status/1575199173472927762',
632 'id': '1575199163847000068',
633 'display_id': '1575199173472927762',
638 'uploader_id': 'Rizdraws',
639 'uploader_url': 'https://twitter.com/Rizdraws',
640 'upload_date': '20220928',
641 'timestamp': 1664391723,
642 'thumbnail': 're:^https?://.*\\.jpg',
645 'comment_count': int,
649 'expected_warnings': ['404'],
651 # Description is missing one https://t.co url (GraphQL)
652 'url': 'https://twitter.com/Srirachachau/status/1395079556562706435',
653 'playlist_mincount': 2,
655 'id': '1395079556562706435',
660 'upload_date': '20210519',
663 'description': 'Here it is! Finished my gothic western cartoon. Pretty proud of it. It\'s got some goofs and lots of splashy over the top violence, something for everyone, hope you like it https://t.co/fOsG5glUnw https://t.co/kbXZrozlY7',
664 'uploader_id': 'Srirachachau',
665 'comment_count': int,
666 'uploader_url': 'https://twitter.com/Srirachachau',
667 'timestamp': 1621447860,
670 # Description is missing one https://t.co url (GraphQL)
671 'url': 'https://twitter.com/DavidToons_/status/1578353380363501568',
672 'playlist_mincount': 2,
674 'id': '1578353380363501568',
676 'uploader_id': 'DavidToons_',
680 'timestamp': 1665143744,
681 'uploader_url': 'https://twitter.com/DavidToons_',
682 'description': 'Chris sounds like Linda from Bob\'s Burgers, so as an animator: this had to be done. https://t.co/glfQdgfFXH https://t.co/WgJauwIW1w',
684 'comment_count': int,
685 'upload_date': '20221007',
689 'url': 'https://twitter.com/primevideouk/status/1578401165338976258',
692 'id': '1578401165338976258',
694 'description': 'md5:659a6b517a034b4cee5d795381a2dc41',
696 'uploader_id': 'primevideouk',
697 'timestamp': 1665155137,
698 'upload_date': '20221007',
700 'uploader_url': 'https://twitter.com/primevideouk',
701 'comment_count': int,
704 'tags': ['TheRingsOfPower'],
708 'url': 'https://twitter.com/MoniqueCamarra/status/1550101959377551360',
710 'id': '1lPJqmBeeNAJb',
712 'title': 'EuroFile@6 Ukraine Up-date-Draghi Defenestration-the West',
713 'uploader': r
're:Monique Camarra.+?',
714 'uploader_id': 'MoniqueCamarra',
715 'live_status': 'was_live',
716 'description': 'md5:acce559345fd49f129c20dbcda3f1201',
717 'timestamp': 1658407771464,
719 'add_ie': ['TwitterSpaces'],
720 'params': {'skip_download': 'm3u8'}
,
723 'url': 'https://twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid.onion/TwitterBlue/status/1484226494708662273',
724 'only_matching': True,
727 'url': 'https://twitter.com/GunB1g/status/1163218564784017422',
728 'only_matching': True,
730 # promo_video_website card
731 'url': 'https://twitter.com/GunB1g/status/1163218564784017422',
732 'only_matching': True,
734 # promo_video_convo card
735 'url': 'https://twitter.com/poco_dandy/status/1047395834013384704',
736 'only_matching': True,
739 'url': 'https://twitter.com/poco_dandy/status/1150646424461176832',
740 'only_matching': True,
742 # video_direct_message card
743 'url': 'https://twitter.com/qarev001/status/1348948114569269251',
744 'only_matching': True,
746 # poll2choice_video card
747 'url': 'https://twitter.com/CAF_Online/status/1349365911120195585',
748 'only_matching': True,
750 # poll3choice_video card
751 'url': 'https://twitter.com/SamsungMobileSA/status/1348609186725289984',
752 'only_matching': True,
754 # poll4choice_video card
755 'url': 'https://twitter.com/SouthamptonFC/status/1347577658079641604',
756 'only_matching': True,
759 def _graphql_to_legacy(self
, data
, twid
):
760 result
= traverse_obj(data
, (
761 'threaded_conversation_with_injections_v2', 'instructions', 0, 'entries',
762 lambda _
, v
: v
['entryId'] == f
'tweet-{twid}', 'content', 'itemContent',
763 'tweet_results', 'result'
764 ), expected_type
=dict, default
={}, get_all
=False)
766 if 'tombstone' in result
:
767 cause
= traverse_obj(result
, ('tombstone', 'text', 'text'), expected_type
=str)
768 raise ExtractorError(f
'Twitter API says: {cause or "Unknown error"}', expected
=True)
770 status
= result
.get('legacy', {})
771 status
.update(traverse_obj(result
, {
772 'user': ('core', 'user_results', 'result', 'legacy'),
773 'card': ('card', 'legacy'),
774 'quoted_status': ('quoted_status_result', 'result', 'legacy'),
775 }, expected_type
=dict, default
={}))
777 # extra transformation is needed since result does not match legacy format
779 binding_value
.get('key'): binding_value
.get('value')
780 for binding_value
in traverse_obj(status
, ('card', 'binding_values', ...), expected_type
=dict)
783 status
['card']['binding_values'] = binding_values
787 def _build_graphql_query(self
, media_id
):
790 'focalTweetId': media_id
,
791 'includePromotedContent': True,
792 'with_rux_injections': False,
793 'withBirdwatchNotes': True,
794 'withCommunity': True,
795 'withDownvotePerspective': False,
796 'withQuickPromoteEligibilityTweetFields': True,
797 'withReactionsMetadata': False,
798 'withReactionsPerspective': False,
799 'withSuperFollowsTweetFields': True,
800 'withSuperFollowsUserFields': True,
801 'withV2Timeline': True,
805 'graphql_is_translatable_rweb_tweet_is_translatable_enabled': False,
806 'interactive_text_enabled': True,
807 'responsive_web_edit_tweet_api_enabled': True,
808 'responsive_web_enhance_cards_enabled': True,
809 'responsive_web_graphql_timeline_navigation_enabled': False,
810 'responsive_web_text_conversations_enabled': False,
811 'responsive_web_uc_gql_enabled': True,
812 'standardized_nudges_misinfo': True,
813 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': False,
814 'tweetypie_unmention_optimization_enabled': True,
815 'unified_cards_ad_metadata_container_dynamic_card_content_query_enabled': True,
816 'verified_phone_label_enabled': False,
817 'vibe_api_enabled': True,
821 def _real_extract(self
, url
):
822 twid
= self
._match
_id
(url
)
823 if self
.is_logged_in
or self
._configuration
_arg
('force_graphql'):
824 self
.write_debug(f
'Using GraphQL API (Auth = {self.is_logged_in})')
825 result
= self
._call
_graphql
_api
('zZXycP0V6H7m-2r0mOnFcA/TweetDetail', twid
)
826 status
= self
._graphql
_to
_legacy
(result
, twid
)
829 status
= self
._call
_api
(f
'statuses/show/{twid}.json', twid
, {
830 'cards_platform': 'Web-12',
832 'include_reply_count': 1,
833 'include_user_entities': 0,
834 'tweet_mode': 'extended',
837 title
= description
= status
['full_text'].replace('\n', ' ')
838 # strip 'https -_t.co_BJYgOjSeGA' junk from filenames
839 title
= re
.sub(r
'\s+(https?://[^ ]+)', '', title
)
840 user
= status
.get('user') or {}
841 uploader
= user
.get('name')
843 title
= f
'{uploader} - {title}'
844 uploader_id
= user
.get('screen_name')
847 for hashtag
in (try_get(status
, lambda x
: x
['entities']['hashtags'], list) or []):
848 hashtag_text
= hashtag
.get('text')
851 tags
.append(hashtag_text
)
856 'description': description
,
857 'uploader': uploader
,
858 'timestamp': unified_timestamp(status
.get('created_at')),
859 'uploader_id': uploader_id
,
860 'uploader_url': format_field(uploader_id
, None, 'https://twitter.com/%s'),
861 'like_count': int_or_none(status
.get('favorite_count')),
862 'repost_count': int_or_none(status
.get('retweet_count')),
863 'comment_count': int_or_none(status
.get('reply_count')),
864 'age_limit': 18 if status
.get('possibly_sensitive') else 0,
868 def extract_from_video_info(media
):
869 media_id
= traverse_obj(media
, 'id_str', 'id', expected_type
=str_or_none
)
870 self
.write_debug(f
'Extracting from video info: {media_id}')
871 video_info
= media
.get('video_info') or {}
875 for variant
in video_info
.get('variants', []):
876 fmts
, subs
= self
._extract
_variant
_formats
(variant
, twid
)
877 subtitles
= self
._merge
_subtitles
(subtitles
, subs
)
881 media_url
= media
.get('media_url_https') or media
.get('media_url')
883 def add_thumbnail(name
, size
):
886 'url': update_url_query(media_url
, {'name': name}
),
887 'width': int_or_none(size
.get('w') or size
.get('width')),
888 'height': int_or_none(size
.get('h') or size
.get('height')),
890 for name
, size
in media
.get('sizes', {}).items():
891 add_thumbnail(name
, size
)
892 add_thumbnail('orig', media
.get('original_info') or {})
897 'subtitles': subtitles
,
898 'thumbnails': thumbnails
,
899 'duration': float_or_none(video_info
.get('duration_millis'), 1000),
900 # The codec of http formats are unknown
901 '_format_sort_fields': ('res', 'br', 'size', 'proto'),
904 def extract_from_card_info(card
):
908 self
.write_debug(f
'Extracting from card info: {card.get("url")}')
909 binding_values
= card
['binding_values']
911 def get_binding_value(k
):
912 o
= binding_values
.get(k
) or {}
913 return try_get(o
, lambda x
: x
[x
['type'].lower() + '_value'])
915 card_name
= card
['name'].split(':')[-1]
916 if card_name
== 'player':
919 'url': get_binding_value('player_url'),
921 elif card_name
== 'periscope_broadcast':
924 'url': get_binding_value('url') or get_binding_value('player_url'),
925 'ie_key': PeriscopeIE
.ie_key(),
927 elif card_name
== 'broadcast':
930 'url': get_binding_value('broadcast_url'),
931 'ie_key': TwitterBroadcastIE
.ie_key(),
933 elif card_name
== 'audiospace':
936 'url': f
'https://twitter.com/i/spaces/{get_binding_value("id")}',
937 'ie_key': TwitterSpacesIE
.ie_key(),
939 elif card_name
== 'summary':
942 'url': get_binding_value('card_url'),
944 elif card_name
== 'unified_card':
945 unified_card
= self
._parse
_json
(get_binding_value('unified_card'), twid
)
946 yield from map(extract_from_video_info
, traverse_obj(
947 unified_card
, ('media_entities', ...), expected_type
=dict))
948 # amplify, promo_video_website, promo_video_convo, appplayer,
949 # video_direct_message, poll2choice_video, poll3choice_video,
950 # poll4choice_video, ...
952 is_amplify
= card_name
== 'amplify'
953 vmap_url
= get_binding_value('amplify_url_vmap') if is_amplify
else get_binding_value('player_stream_url')
954 content_id
= get_binding_value('%s_content_id' % (card_name
if is_amplify
else 'player'))
955 formats
, subtitles
= self
._extract
_formats
_from
_vmap
_url
(vmap_url
, content_id
or twid
)
958 for suffix
in ('_small', '', '_large', '_x_large', '_original'):
959 image
= get_binding_value('player_image' + suffix
) or {}
960 image_url
= image
.get('url')
961 if not image_url
or '/player-placeholder' in image_url
:
964 'id': suffix
[1:] if suffix
else 'medium',
966 'width': int_or_none(image
.get('width')),
967 'height': int_or_none(image
.get('height')),
972 'subtitles': subtitles
,
973 'thumbnails': thumbnails
,
974 'duration': int_or_none(get_binding_value(
975 'content_duration_seconds')),
978 media_path
= ((None, 'quoted_status'), 'extended_entities', 'media', lambda _
, m
: m
['type'] != 'photo')
979 videos
= map(extract_from_video_info
, traverse_obj(status
, media_path
, expected_type
=dict))
980 cards
= extract_from_card_info(status
.get('card'))
981 entries
= [{**info, **data, 'display_id': twid}
for data
in (*videos
, *cards
)]
984 expanded_url
= traverse_obj(status
, ('entities', 'urls', 0, 'expanded_url'), expected_type
=url_or_none
)
985 if not expanded_url
or expanded_url
== url
:
986 raise ExtractorError('No video could be found in this tweet', expected
=True)
988 return self
.url_result(expanded_url
, display_id
=twid
, **info
)
990 entries
[0]['_old_archive_ids'] = [make_archive_id(self
, twid
)]
992 if len(entries
) == 1:
995 for index
, entry
in enumerate(entries
, 1):
996 entry
['title'] += f
' #{index}'
998 return self
.playlist_result(entries
, **info
)
1001 class TwitterAmplifyIE(TwitterBaseIE
):
1002 IE_NAME
= 'twitter:amplify'
1003 _VALID_URL
= r
'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})'
1006 'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
1007 'md5': 'fec25801d18a4557c5c9f33d2c379ffa',
1009 'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
1011 'title': 'Twitter Video',
1012 'thumbnail': 're:^https?://.*',
1014 'params': {'format': '[protocol=https]'}
,
1017 def _real_extract(self
, url
):
1018 video_id
= self
._match
_id
(url
)
1019 webpage
= self
._download
_webpage
(url
, video_id
)
1021 vmap_url
= self
._html
_search
_meta
(
1022 'twitter:amplify:vmap', webpage
, 'vmap url')
1023 formats
, _
= self
._extract
_formats
_from
_vmap
_url
(vmap_url
, video_id
)
1026 thumbnail
= self
._html
_search
_meta
(
1027 'twitter:image:src', webpage
, 'thumbnail', fatal
=False)
1029 def _find_dimension(target
):
1030 w
= int_or_none(self
._html
_search
_meta
(
1031 'twitter:%s:width' % target
, webpage
, fatal
=False))
1032 h
= int_or_none(self
._html
_search
_meta
(
1033 'twitter:%s:height' % target
, webpage
, fatal
=False))
1037 thumbnail_w
, thumbnail_h
= _find_dimension('image')
1040 'width': thumbnail_w
,
1041 'height': thumbnail_h
,
1044 video_w
, video_h
= _find_dimension('player')
1052 'title': 'Twitter Video',
1054 'thumbnails': thumbnails
,
1058 class TwitterBroadcastIE(TwitterBaseIE
, PeriscopeBaseIE
):
1059 IE_NAME
= 'twitter:broadcast'
1060 _VALID_URL
= TwitterBaseIE
._BASE
_REGEX
+ r
'i/broadcasts/(?P<id>[0-9a-zA-Z]{13})'
1063 # untitled Periscope video
1064 'url': 'https://twitter.com/i/broadcasts/1yNGaQLWpejGj',
1066 'id': '1yNGaQLWpejGj',
1068 'title': 'Andrea May Sahouri - Periscope Broadcast',
1069 'uploader': 'Andrea May Sahouri',
1070 'uploader_id': '1PXEdBZWpGwKe',
1071 'thumbnail': r
're:^https?://[^?#]+\.jpg\?token=',
1076 def _real_extract(self
, url
):
1077 broadcast_id
= self
._match
_id
(url
)
1078 broadcast
= self
._call
_api
(
1079 'broadcasts/show.json', broadcast_id
,
1080 {'ids': broadcast_id}
)['broadcasts'][broadcast_id
]
1081 info
= self
._parse
_broadcast
_data
(broadcast
, broadcast_id
)
1082 media_key
= broadcast
['media_key']
1083 source
= self
._call
_api
(
1084 f
'live_video_stream/status/{media_key}', media_key
)['source']
1085 m3u8_url
= source
.get('noRedirectPlaybackUrl') or source
['location']
1086 if '/live_video_stream/geoblocked/' in m3u8_url
:
1087 self
.raise_geo_restricted()
1088 m3u8_id
= compat_parse_qs(compat_urllib_parse_urlparse(
1089 m3u8_url
).query
).get('type', [None])[0]
1090 state
, width
, height
= self
._extract
_common
_format
_info
(broadcast
)
1091 info
['formats'] = self
._extract
_pscp
_m
3u8_formats
(
1092 m3u8_url
, broadcast_id
, m3u8_id
, state
, width
, height
)
1096 class TwitterSpacesIE(TwitterBaseIE
):
1097 IE_NAME
= 'twitter:spaces'
1098 _VALID_URL
= TwitterBaseIE
._BASE
_REGEX
+ r
'i/spaces/(?P<id>[0-9a-zA-Z]{13})'
1099 _TWITTER_GRAPHQL
= 'https://twitter.com/i/api/graphql/HPEisOmj1epUNLCWTYhUWw/'
1102 'url': 'https://twitter.com/i/spaces/1RDxlgyvNXzJL',
1104 'id': '1RDxlgyvNXzJL',
1106 'title': 'King Carlo e la mossa Kansas City per fare il Grande Centro',
1107 'description': 'Twitter Space participated by annarita digiorgio, Signor Ernesto, Raffaello Colosimo, Simone M. Sepe',
1108 'uploader': r
're:Lucio Di Gaetano.*?',
1109 'uploader_id': 'luciodigaetano',
1110 'live_status': 'was_live',
1111 'timestamp': 1659877956397,
1113 'params': {'skip_download': 'm3u8'}
,
1117 'notstarted': 'is_upcoming',
1118 'ended': 'was_live',
1119 'running': 'is_live',
1120 'timedout': 'post_live',
1123 def _build_graphql_query(self
, space_id
):
1127 'isMetatagsQuery': True,
1128 'withDownvotePerspective': False,
1129 'withReactionsMetadata': False,
1130 'withReactionsPerspective': False,
1131 'withReplays': True,
1132 'withSuperFollowsUserFields': True,
1133 'withSuperFollowsTweetFields': True,
1136 'dont_mention_me_view_api_enabled': True,
1137 'interactive_text_enabled': True,
1138 'responsive_web_edit_tweet_api_enabled': True,
1139 'responsive_web_enhance_cards_enabled': True,
1140 'responsive_web_uc_gql_enabled': True,
1141 'spaces_2022_h2_clipping': True,
1142 'spaces_2022_h2_spaces_communities': False,
1143 'standardized_nudges_misinfo': True,
1144 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': False,
1145 'vibe_api_enabled': True,
1149 def _real_extract(self
, url
):
1150 space_id
= self
._match
_id
(url
)
1151 space_data
= self
._call
_graphql
_api
('HPEisOmj1epUNLCWTYhUWw/AudioSpaceById', space_id
)['audioSpace']
1153 raise ExtractorError('Twitter Space not found', expected
=True)
1155 metadata
= space_data
['metadata']
1156 live_status
= try_call(lambda: self
.SPACE_STATUS
[metadata
['state'].lower()])
1159 if live_status
== 'is_upcoming':
1160 self
.raise_no_formats('Twitter Space not started yet', expected
=True)
1161 elif live_status
== 'post_live':
1162 self
.raise_no_formats('Twitter Space ended but not downloadable yet', expected
=True)
1164 source
= self
._call
_api
(
1165 f
'live_video_stream/status/{metadata["media_key"]}', metadata
['media_key'])['source']
1167 # XXX: Native downloader does not work
1168 formats
= self
._extract
_m
3u8_formats
(
1169 traverse_obj(source
, 'noRedirectPlaybackUrl', 'location'),
1170 metadata
['media_key'], 'm4a', 'm3u8', live
=live_status
== 'is_live')
1172 fmt
.update({'vcodec': 'none', 'acodec': 'aac'}
)
1174 participants
= ', '.join(traverse_obj(
1175 space_data
, ('participants', 'speakers', ..., 'display_name'))) or 'nobody yet'
1178 'title': metadata
.get('title'),
1179 'description': f
'Twitter Space participated by {participants}',
1180 'uploader': traverse_obj(
1181 metadata
, ('creator_results', 'result', 'legacy', 'name')),
1182 'uploader_id': traverse_obj(
1183 metadata
, ('creator_results', 'result', 'legacy', 'screen_name')),
1184 'live_status': live_status
,
1185 'timestamp': metadata
.get('created_at'),
1190 class TwitterShortenerIE(TwitterBaseIE
):
1191 IE_NAME
= 'twitter:shortener'
1192 _VALID_URL
= r
'https?://t.co/(?P<id>[^?]+)|tco:(?P<eid>[^?]+)'
1193 _BASE_URL
= 'https://t.co/'
1195 def _real_extract(self
, url
):
1196 mobj
= self
._match
_valid
_url
(url
)
1197 eid
, id = mobj
.group('eid', 'id')
1200 url
= self
._BASE
_URL
+ id
1201 new_url
= self
._request
_webpage
(url
, id, headers
={'User-Agent': 'curl'}
).geturl()
1202 __UNSAFE_LINK
= "https://twitter.com/safety/unsafe_link_warning?unsafe_link="
1203 if new_url
.startswith(__UNSAFE_LINK
):
1204 new_url
= new_url
.replace(__UNSAFE_LINK
, "")
1205 return self
.url_result(new_url
)