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
,
34 class TwitterBaseIE(InfoExtractor
):
35 _NETRC_MACHINE
= 'twitter'
36 _API_BASE
= 'https://api.twitter.com/1.1/'
37 _GRAPHQL_API_BASE
= 'https://twitter.com/i/api/graphql/'
38 _BASE_REGEX
= r
'https?://(?:(?:www|m(?:obile)?)\.)?(?:twitter\.com|twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid\.onion)/'
39 _AUTH
= 'AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA'
40 _LEGACY_AUTH
= 'AAAAAAAAAAAAAAAAAAAAAIK1zgAAAAAA2tUWuhGZ2JceoId5GwYWU5GspY4%3DUq7gzFoCZs1QfwGoVdvSac3IniczZEYXIcDyumCauIXpcAPorE'
43 _LOGIN_INIT_DATA
= json
.dumps({
46 'debug_overrides': {},
55 'app_download_cta': 1,
56 'check_logged_in_account': 1,
57 'choice_selection': 3,
58 'contacts_live_sync_permission_prompt': 0,
60 'email_verification': 2,
70 'in_app_notification': 1,
72 'js_instrumentation': 1,
74 'notifications_permission_prompt': 2,
76 'open_home_timeline': 1,
78 'phone_verification': 4,
87 'tweet_selection_urt': 1,
90 'user_recommendations_list': 4,
91 'user_recommendations_urt': 1,
95 }, separators
=(',', ':')).encode()
97 def _extract_variant_formats(self
, variant
, video_id
):
98 variant_url
= variant
.get('url')
101 elif '.m3u8' in variant_url
:
102 return self
._extract
_m
3u8_formats
_and
_subtitles
(
103 variant_url
, video_id
, 'mp4', 'm3u8_native',
104 m3u8_id
='hls', fatal
=False)
106 tbr
= int_or_none(dict_get(variant
, ('bitrate', 'bit_rate')), 1000) or None
109 'format_id': 'http' + ('-%d' % tbr
if tbr
else ''),
112 self
._search
_dimensions
_in
_video
_url
(f
, variant_url
)
115 def _extract_formats_from_vmap_url(self
, vmap_url
, video_id
):
116 vmap_url
= url_or_none(vmap_url
)
119 vmap_data
= self
._download
_xml
(vmap_url
, video_id
)
123 for video_variant
in vmap_data
.findall('.//{http://twitter.com/schema/videoVMapV2.xsd}videoVariant'):
124 video_variant
.attrib
['url'] = compat_urllib_parse_unquote(
125 video_variant
.attrib
['url'])
126 urls
.append(video_variant
.attrib
['url'])
127 fmts
, subs
= self
._extract
_variant
_formats
(
128 video_variant
.attrib
, video_id
)
130 subtitles
= self
._merge
_subtitles
(subtitles
, subs
)
131 video_url
= strip_or_none(xpath_text(vmap_data
, './/MediaFile'))
132 if video_url
not in urls
:
133 fmts
, subs
= self
._extract
_variant
_formats
({'url': video_url}
, video_id
)
135 subtitles
= self
._merge
_subtitles
(subtitles
, subs
)
136 return formats
, subtitles
139 def _search_dimensions_in_video_url(a_format
, video_url
):
140 m
= re
.search(r
'/(?P<width>\d+)x(?P<height>\d+)/', video_url
)
143 'width': int(m
.group('width')),
144 'height': int(m
.group('height')),
148 def is_logged_in(self
):
149 return bool(self
._get
_cookies
(self
._API
_BASE
).get('auth_token'))
151 @functools.cached_property
152 def _selected_api(self
):
153 return self
._configuration
_arg
('api', ['graphql'], ie_key
='Twitter')[0]
155 def _fetch_guest_token(self
, display_id
):
156 guest_token
= traverse_obj(self
._download
_json
(
157 f
'{self._API_BASE}guest/activate.json', display_id
, 'Downloading guest token', data
=b
'',
158 headers
=self
._set
_base
_headers
(legacy
=display_id
and self
._selected
_api
== 'legacy')),
159 ('guest_token', {str}
))
161 raise ExtractorError('Could not retrieve guest token')
164 def _set_base_headers(self
, legacy
=False):
165 bearer_token
= self
._LEGACY
_AUTH
if legacy
and not self
.is_logged_in
else self
._AUTH
167 'Authorization': f
'Bearer {bearer_token}',
168 'x-csrf-token': try_call(lambda: self
._get
_cookies
(self
._API
_BASE
)['ct0'].value
),
171 def _call_login_api(self
, note
, headers
, query
={}, data
=None):
172 response
= self
._download
_json
(
173 f
'{self._API_BASE}onboarding/task.json', None, note
,
174 headers
=headers
, query
=query
, data
=data
, expected_status
=400)
175 error
= traverse_obj(response
, ('errors', 0, 'message', {str}
))
177 raise ExtractorError(f
'Login failed, Twitter API says: {error}', expected
=True)
178 elif traverse_obj(response
, 'status') != 'success':
179 raise ExtractorError('Login was unsuccessful')
181 subtask
= traverse_obj(
182 response
, ('subtasks', ..., 'subtask_id', {str}
), get_all
=False)
184 raise ExtractorError('Twitter API did not return next login subtask')
186 self
._flow
_token
= response
['flow_token']
190 def _perform_login(self
, username
, password
):
191 if self
.is_logged_in
:
194 webpage
= self
._download
_webpage
('https://twitter.com/', None, 'Downloading login page')
195 guest_token
= self
._search
_regex
(
196 r
'\.cookie\s*=\s*["\']gt
=(\d
+);', webpage, 'gt
', default=None) or self._fetch_guest_token(None)
198 **self._set_base_headers(),
199 'content
-type': 'application
/json
',
200 'x
-guest
-token
': guest_token,
201 'x
-twitter
-client
-language
': 'en
',
202 'x
-twitter
-active
-user
': 'yes
',
203 'Referer
': 'https
://twitter
.com
/',
204 'Origin
': 'https
://twitter
.com
',
207 def build_login_json(*subtask_inputs):
209 'flow_token
': self._flow_token,
210 'subtask_inputs
': subtask_inputs
211 }, separators=(',', ':')).encode()
213 def input_dict(subtask_id, text):
215 'subtask_id
': subtask_id,
222 next_subtask = self._call_login_api(
223 'Downloading flow token
', headers, query={'flow_name': 'login'}, data=self._LOGIN_INIT_DATA)
225 while not self.is_logged_in:
226 if next_subtask == 'LoginJsInstrumentationSubtask
':
227 next_subtask = self._call_login_api(
228 'Submitting JS instrumentation response
', headers, data=build_login_json({
229 'subtask_id
': next_subtask,
230 'js_instrumentation
': {
236 elif next_subtask == 'LoginEnterUserIdentifierSSO
':
237 next_subtask = self._call_login_api(
238 'Submitting username
', headers, data=build_login_json({
239 'subtask_id
': next_subtask,
241 'setting_responses
': [{
242 'key
': 'user_identifier
',
253 elif next_subtask == 'LoginEnterAlternateIdentifierSubtask
':
254 next_subtask = self._call_login_api(
255 'Submitting alternate identifier
', headers,
256 data=build_login_json(input_dict(next_subtask, self._get_tfa_info(
257 'one of username
, phone number
or email that was
not used
as --username
'))))
259 elif next_subtask == 'LoginEnterPassword
':
260 next_subtask = self._call_login_api(
261 'Submitting password
', headers, data=build_login_json({
262 'subtask_id
': next_subtask,
264 'password
': password,
269 elif next_subtask == 'AccountDuplicationCheck
':
270 next_subtask = self._call_login_api(
271 'Submitting account duplication check
', headers, data=build_login_json({
272 'subtask_id
': next_subtask,
273 'check_logged_in_account
': {
274 'link
': 'AccountDuplicationCheck_false
'
278 elif next_subtask == 'LoginTwoFactorAuthChallenge
':
279 next_subtask = self._call_login_api(
280 'Submitting
2FA token
', headers, data=build_login_json(input_dict(
281 next_subtask, self._get_tfa_info('two
-factor authentication token
'))))
283 elif next_subtask == 'LoginAcid
':
284 next_subtask = self._call_login_api(
285 'Submitting confirmation code
', headers, data=build_login_json(input_dict(
286 next_subtask, self._get_tfa_info('confirmation code sent to your email
or phone
'))))
288 elif next_subtask == 'ArkoseLogin
':
289 self.raise_login_required('Twitter
is requiring captcha
for this login attempt
', method='cookies
')
291 elif next_subtask == 'DenyLoginSubtask
':
292 self.raise_login_required('Twitter rejected this login attempt
as suspicious
', method='cookies
')
294 elif next_subtask == 'LoginSuccessSubtask
':
295 raise ExtractorError('Twitter API did
not grant auth token cookie
')
298 raise ExtractorError(f'Unrecognized subtask ID
"{next_subtask}"')
302 def _call_api(self, path, video_id, query={}, graphql=False):
303 headers = self._set_base_headers(legacy=not graphql and self._selected_api == 'legacy
')
305 'x
-twitter
-auth
-type': 'OAuth2Session
',
306 'x
-twitter
-client
-language
': 'en
',
307 'x
-twitter
-active
-user
': 'yes
',
308 } if self.is_logged_in else {
309 'x
-guest
-token
': self._fetch_guest_token(video_id)
311 allowed_status = {400, 401, 403, 404} if graphql else {403}
312 result = self._download_json(
313 (self._GRAPHQL_API_BASE if graphql else self._API_BASE) + path,
314 video_id, headers=headers, query=query, expected_status=allowed_status,
315 note=f'Downloading {"GraphQL" if graphql else "legacy API"} JSON
')
317 if result.get('errors
'):
318 errors = ', '.join(set(traverse_obj(result, ('errors
', ..., 'message
', {str}))))
319 if errors and 'not authorized
' in errors:
320 self.raise_login_required(remove_end(errors, '.'))
321 raise ExtractorError(f'Error(s
) while querying API
: {errors or "Unknown error"}
')
325 def _build_graphql_query(self, media_id):
326 raise NotImplementedError('Method must be implemented to support GraphQL
')
328 def _call_graphql_api(self, endpoint, media_id):
329 data = self._build_graphql_query(media_id)
330 query = {key: json.dumps(value, separators=(',', ':')) for key, value in data.items()}
331 return traverse_obj(self._call_api(endpoint, media_id, query=query, graphql=True), 'data
')
334 class TwitterCardIE(InfoExtractor):
335 IE_NAME = 'twitter
:card
'
336 _VALID_URL = TwitterBaseIE._BASE_REGEX + r'i
/(?
:cards
/tfw
/v1|
videos(?
:/tweet
)?
)/(?P
<id>\d
+)'
339 'url
': 'https
://twitter
.com
/i
/cards
/tfw
/v1
/560070183650213889',
340 # MD5 checksums are different in different places
342 'id': '560070131976392705',
344 'title
': "Twitter - You can now shoot, edit and share video on Twitter. Capture life's most moving moments
from your perspective
.",
345 'description': 'md5:18d3e24bb4f6e5007487dd546e53bd96',
346 'uploader': 'Twitter',
347 'uploader_id': 'Twitter',
348 'thumbnail': r're:^https?://.*\.jpg',
350 'timestamp': 1422366112,
351 'upload_date': '20150127',
353 'comment_count': int,
357 'display_id': '560070183650213889',
358 'uploader_url': 'https://twitter.com/Twitter',
362 'url': 'https://twitter.com/i/cards/tfw/v1/623160978427936768',
363 'md5': '7137eca597f72b9abbe61e5ae0161399',
365 'id': '623160978427936768',
367 'title': "NASA
- Fly over Pluto
's icy Norgay Mountains and Sputnik Plain in this @NASANewHorizons #PlutoFlyby video.",
368 'description
': "Fly over Pluto's icy Norgay Mountains
and Sputnik Plain
in this
@NASANewHorizons #PlutoFlyby video. https://t.co/BJYgOjSeGA",
370 'uploader_id': 'NASA',
371 'timestamp': 1437408129,
372 'upload_date': '20150720',
373 'uploader_url': 'https://twitter.com/NASA',
375 'comment_count': int,
378 'tags': ['PlutoFlyby'],
380 'params': {'format': '[protocol=https]'}
383 'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
384 'md5': 'b6d9683dd3f48e340ded81c0e917ad46',
388 'title': 'Ubuntu 11.10 Overview',
389 'description': 'md5:a831e97fa384863d6e26ce48d1c43376',
390 'upload_date': '20111013',
391 'uploader': 'OMG! UBUNTU!',
392 'uploader_id': 'omgubuntu',
393 'channel_url': 'https://www.youtube.com/channel/UCIiSwcm9xiFb3Y4wjzR41eQ',
394 'channel_id': 'UCIiSwcm9xiFb3Y4wjzR41eQ',
395 'channel_follower_count': int,
396 'chapters': 'count:8',
397 'uploader_url': 'http://www.youtube.com/user/omgubuntu',
399 'categories': ['Film & Animation'],
401 'comment_count': int,
402 'availability': 'public',
404 'thumbnail': 'https://i.ytimg.com/vi/dq4Oj5quskI/maxresdefault.jpg',
407 'channel': 'OMG! UBUNTU!',
408 'playable_in_embed': True,
410 'add_ie': ['Youtube'],
413 'url': 'https://twitter.com/i/cards/tfw/v1/665289828897005568',
417 'upload_date': '20151113',
418 'uploader_id': '1189339351084113920',
419 'uploader': 'ArsenalTerje',
420 'title': 'Vine by ArsenalTerje',
421 'timestamp': 1447451307,
422 'alt_title': 'Vine by ArsenalTerje',
423 'comment_count': int,
425 'thumbnail': r
're:^https?://[^?#]+\.jpg',
430 'params': {'skip_download': 'm3u8'}
,
433 'url': 'https://twitter.com/i/videos/tweet/705235433198714880',
434 'md5': '884812a2adc8aaf6fe52b15ccbfa3b88',
436 'id': '705235433198714880',
438 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
439 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
440 'uploader': 'Brent Yarina',
441 'uploader_id': 'BTNBrentYarina',
442 'timestamp': 1456976204,
443 'upload_date': '20160303',
445 'skip': 'This content is no longer available.',
448 'url': 'https://twitter.com/i/videos/752274308186120192',
449 'only_matching': True,
453 def _real_extract(self
, url
):
454 status_id
= self
._match
_id
(url
)
455 return self
.url_result(
456 'https://twitter.com/statuses/' + status_id
,
457 TwitterIE
.ie_key(), status_id
)
460 class TwitterIE(TwitterBaseIE
):
462 _VALID_URL
= TwitterBaseIE
._BASE
_REGEX
+ r
'(?:(?:i/web|[^/]+)/status|statuses)/(?P<id>\d+)(?:/(?:video|photo)/(?P<index>\d+))?'
465 'url': 'https://twitter.com/freethenipple/status/643211948184596480',
467 'id': '643211870443208704',
468 'display_id': '643211948184596480',
470 'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
471 'thumbnail': r
're:^https?://.*\.jpg',
472 'description': 'FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ',
473 'uploader': 'FREE THE NIPPLE',
474 'uploader_id': 'freethenipple',
476 'timestamp': 1442188653,
477 'upload_date': '20150913',
478 'uploader_url': 'https://twitter.com/freethenipple',
479 'comment_count': int,
487 'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
488 'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
490 'id': '657991469417025536',
492 'title': 'Gifs - tu vai cai tu vai cai tu nao eh capaz disso tu vai cai',
493 'description': 'Gifs on Twitter: "tu vai cai tu vai cai tu nao eh capaz disso tu vai cai https://t.co/tM46VHFlO5"',
494 'thumbnail': r
're:^https?://.*\.png',
496 'uploader_id': 'giphz',
498 'expected_warnings': ['height', 'width'],
499 'skip': 'Account suspended',
501 'url': 'https://twitter.com/starwars/status/665052190608723968',
503 'id': '665052190608723968',
504 'display_id': '665052190608723968',
506 'title': r
're:Star Wars.*A new beginning is coming December 18.*',
507 'description': 'A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens. https://t.co/OkSqT2fjWJ',
508 'uploader_id': 'starwars',
509 'uploader': r
're:Star Wars.*',
510 'timestamp': 1447395772,
511 'upload_date': '20151113',
512 'uploader_url': 'https://twitter.com/starwars',
513 'comment_count': int,
516 'tags': ['TV', 'StarWars', 'TheForceAwakens'],
520 'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880',
522 'id': '705235433198714880',
524 'title': "Brent Yarina - Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight.",
525 'description': "Khalil Iverson's missed highlight dunk. And made highlight dunk. In one highlight. https://t.co/OrxcJ28Bns",
526 'uploader_id': 'BTNBrentYarina',
527 'uploader': 'Brent Yarina',
528 'timestamp': 1456976204,
529 'upload_date': '20160303',
530 'uploader_url': 'https://twitter.com/BTNBrentYarina',
531 'comment_count': int,
538 # The same video as https://twitter.com/i/videos/tweet/705235433198714880
539 # Test case of TwitterCardIE
540 'skip_download': True,
542 'skip': 'Dead external link',
544 'url': 'https://twitter.com/jaydingeer/status/700207533655363584',
546 'id': '700207414000242688',
547 'display_id': '700207533655363584',
549 'title': 'jaydin donte geer - BEAT PROD: @suhmeduh #Damndaniel',
550 'description': 'BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ',
551 'thumbnail': r
're:^https?://.*\.jpg',
552 'uploader': 'jaydin donte geer',
553 'uploader_id': 'jaydingeer',
555 'timestamp': 1455777459,
556 'upload_date': '20160218',
557 'uploader_url': 'https://twitter.com/jaydingeer',
558 'comment_count': int,
562 'tags': ['Damndaniel'],
566 'url': 'https://twitter.com/Filmdrunk/status/713801302971588609',
567 'md5': '89a15ed345d13b86e9a5a5e051fa308a',
571 'title': 'Dr.Pepperの飲み方 #japanese #バカ #ドクペ #電動ガン',
572 'uploader': 'TAKUMA',
573 'uploader_id': '1004126642786242560',
574 'timestamp': 1402826626,
575 'upload_date': '20140615',
576 'thumbnail': r
're:^https?://.*\.jpg',
577 'alt_title': 'Vine by TAKUMA',
578 'comment_count': int,
585 'url': 'https://twitter.com/captainamerica/status/719944021058060289',
587 'id': '717462543795523584',
588 'display_id': '719944021058060289',
590 'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.',
591 'description': '@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI',
592 'uploader_id': 'CaptainAmerica',
593 'uploader': 'Captain America',
595 'timestamp': 1460483005,
596 'upload_date': '20160412',
597 'uploader_url': 'https://twitter.com/CaptainAmerica',
598 'thumbnail': r
're:^https?://.*\.jpg',
599 'comment_count': int,
607 'url': 'https://twitter.com/OPP_HSD/status/779210622571536384',
609 'id': '1zqKVVlkqLaKB',
611 'title': 'Sgt Kerry Schmidt - Ontario Provincial Police - Road rage, mischief, assault, rollover and fire in one occurrence',
612 'upload_date': '20160923',
613 'uploader_id': '1PmKqpJdOJQoY',
614 'uploader': 'Sgt Kerry Schmidt - Ontario Provincial Police',
615 'timestamp': 1474613214,
616 'thumbnail': r
're:^https?://.*\.jpg',
618 'add_ie': ['Periscope'],
620 # has mp4 formats via mobile API
621 'url': 'https://twitter.com/news_al3alm/status/852138619213144067',
623 'id': '852077943283097602',
625 'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة',
626 'description': 'كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN',
627 'uploader': 'عالم الأخبار',
628 'uploader_id': 'news_al3alm',
630 'timestamp': 1492000653,
631 'upload_date': '20170412',
632 'display_id': '852138619213144067',
634 'uploader_url': 'https://twitter.com/news_al3alm',
635 'thumbnail': r
're:^https?://.*\.jpg',
640 'comment_count': int,
643 'url': 'https://twitter.com/i/web/status/910031516746514432',
645 'id': '910030238373089285',
646 'display_id': '910031516746514432',
648 '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.',
649 'thumbnail': r
're:^https?://.*\.jpg',
650 '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',
651 'uploader': 'Préfet de Guadeloupe',
652 'uploader_id': 'Prefet971',
654 'timestamp': 1505803395,
655 'upload_date': '20170919',
656 'uploader_url': 'https://twitter.com/Prefet971',
657 'comment_count': int,
665 'skip_download': True, # requires ffmpeg
668 # card via api.twitter.com/1.1/videos/tweet/config
669 'url': 'https://twitter.com/LisPower1/status/1001551623938805763',
671 'id': '1001551417340022785',
672 'display_id': '1001551623938805763',
674 'title': 're:.*?Shep is on a roll today.*?',
675 'thumbnail': r
're:^https?://.*\.jpg',
676 'description': 'md5:37b9f2ff31720cef23b2bd42ee8a0f09',
677 'uploader': 'Lis Power',
678 'uploader_id': 'LisPower1',
680 'timestamp': 1527623489,
681 'upload_date': '20180529',
682 'uploader_url': 'https://twitter.com/LisPower1',
683 'comment_count': int,
691 'skip_download': True, # requires ffmpeg
694 'url': 'https://twitter.com/foobar/status/1087791357756956680',
696 'id': '1087791272830607360',
697 'display_id': '1087791357756956680',
699 'title': 'X - 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!',
700 'thumbnail': r
're:^https?://.*\.jpg',
701 'description': 'md5:6dfd341a3310fb97d80d2bf7145df976',
705 'timestamp': 1548184644,
706 'upload_date': '20190122',
707 'uploader_url': 'https://twitter.com/X',
708 'comment_count': int,
715 'skip': 'This Tweet is unavailable',
717 # not available in Periscope
718 'url': 'https://twitter.com/ViviEducation/status/1136534865145286656',
720 'id': '1vOGwqejwoWxB',
722 'title': 'Vivi - Vivi founder @lior_rauchy announcing our new student feedback tool live at @EduTECH_AU #EduTECH2019',
724 'uploader_id': '1eVjYOLGkGrQL',
725 'thumbnail': r
're:^https?://.*\.jpg',
726 'tags': ['EduTECH2019'],
729 'add_ie': ['TwitterBroadcast'],
730 'skip': 'Broadcast no longer exists',
733 'url': 'https://twitter.com/BrooklynNets/status/1349794411333394432?s=20',
735 'id': '1349774757969989634',
736 'display_id': '1349794411333394432',
738 'title': 'md5:d1c4941658e4caaa6cb579260d85dcba',
739 'thumbnail': r
're:^https?://.*\.jpg',
740 'description': 'md5:71ead15ec44cee55071547d6447c6a3e',
741 'uploader': 'Brooklyn Nets',
742 'uploader_id': 'BrooklynNets',
744 'timestamp': 1610651040,
745 'upload_date': '20210114',
746 'uploader_url': 'https://twitter.com/BrooklynNets',
747 'comment_count': int,
754 'skip_download': True,
757 'url': 'https://twitter.com/oshtru/status/1577855540407197696',
759 'id': '1577855447914409984',
760 'display_id': '1577855540407197696',
762 'title': 'md5:9d198efb93557b8f8d5b78c480407214',
763 'description': 'md5:b9c3699335447391d11753ab21c70a74',
764 'upload_date': '20221006',
765 'uploader': 'oshtru',
766 'uploader_id': 'oshtru',
767 'uploader_url': 'https://twitter.com/oshtru',
768 'thumbnail': r
're:^https?://.*\.jpg',
770 'timestamp': 1665025050,
771 'comment_count': int,
778 'params': {'skip_download': True}
,
780 'url': 'https://twitter.com/UltimaShadowX/status/1577719286659006464',
782 'id': '1577719286659006464',
783 'title': 'Ultima📛| New Era - Test',
784 'description': 'Test https://t.co/Y3KEZD7Dad',
785 'uploader': 'Ultima📛| New Era',
786 'uploader_id': 'UltimaShadowX',
787 'uploader_url': 'https://twitter.com/UltimaShadowX',
788 'upload_date': '20221005',
789 'timestamp': 1664992565,
790 'comment_count': int,
797 'params': {'skip_download': True}
,
799 'url': 'https://twitter.com/MesoMax919/status/1575560063510810624',
801 'id': '1575559336759263233',
802 'display_id': '1575560063510810624',
804 'title': 'md5:eec26382babd0f7c18f041db8ae1c9c9',
805 'thumbnail': r
're:^https?://.*\.jpg',
806 'description': 'md5:95aea692fda36a12081b9629b02daa92',
807 'uploader': 'Max Olson',
808 'uploader_id': 'MesoMax919',
809 'uploader_url': 'https://twitter.com/MesoMax919',
811 'timestamp': 1664477766,
812 'upload_date': '20220929',
813 'comment_count': int,
817 'tags': ['HurricaneIan'],
821 # Adult content, fails if not logged in
822 'url': 'https://twitter.com/Rizdraws/status/1575199173472927762',
824 'id': '1575199163847000068',
825 'display_id': '1575199173472927762',
830 'uploader_id': 'Rizdraws',
831 'uploader_url': 'https://twitter.com/Rizdraws',
832 'upload_date': '20220928',
833 'timestamp': 1664391723,
834 'thumbnail': r
're:^https?://.+\.jpg',
837 'comment_count': int,
841 'params': {'skip_download': 'The media could not be played'}
,
842 'skip': 'Requires authentication',
844 # Playlist result only with graphql API
845 'url': 'https://twitter.com/Srirachachau/status/1395079556562706435',
846 'playlist_mincount': 2,
848 'id': '1395079556562706435',
853 'upload_date': '20210519',
856 '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',
857 'uploader_id': 'Srirachachau',
858 'comment_count': int,
859 'uploader_url': 'https://twitter.com/Srirachachau',
860 'timestamp': 1621447860,
863 'url': 'https://twitter.com/DavidToons_/status/1578353380363501568',
864 'playlist_mincount': 2,
866 'id': '1578353380363501568',
868 'uploader_id': 'DavidToons_',
872 'timestamp': 1665143744,
873 'uploader_url': 'https://twitter.com/DavidToons_',
874 'description': 'Chris sounds like Linda from Bob\'s Burgers, so as an animator: this had to be done. https://t.co/WgJauwIW1w',
876 'comment_count': int,
877 'upload_date': '20221007',
881 'url': 'https://twitter.com/primevideouk/status/1578401165338976258',
884 'id': '1578401165338976258',
886 'description': 'md5:659a6b517a034b4cee5d795381a2dc41',
888 'uploader_id': 'primevideouk',
889 'timestamp': 1665155137,
890 'upload_date': '20221007',
892 'uploader_url': 'https://twitter.com/primevideouk',
893 'comment_count': int,
896 'tags': ['TheRingsOfPower'],
900 'url': 'https://twitter.com/MoniqueCamarra/status/1550101959377551360',
902 'id': '1lPJqmBeeNAJb',
904 'title': 'EuroFile@6 Ukraine Up-date-Draghi Defenestration-the West',
905 'uploader': r
're:Monique Camarra.+?',
906 'uploader_id': 'MoniqueCamarra',
907 'live_status': 'was_live',
908 'release_timestamp': 1658417414,
909 'description': 'md5:acce559345fd49f129c20dbcda3f1201',
910 'timestamp': 1658407771,
911 'release_date': '20220721',
912 'upload_date': '20220721',
914 'add_ie': ['TwitterSpaces'],
915 'params': {'skip_download': 'm3u8'}
,
916 'skip': 'Requires authentication',
918 # URL specifies video number but --yes-playlist
919 'url': 'https://twitter.com/CTVJLaidlaw/status/1600649710662213632/video/1',
920 'playlist_mincount': 2,
922 'id': '1600649710662213632',
923 'title': 'md5:be05989b0722e114103ed3851a0ffae2',
924 'timestamp': 1670459604.0,
925 'description': 'md5:591c19ce66fadc2359725d5cd0d1052c',
926 'comment_count': int,
927 'uploader_id': 'CTVJLaidlaw',
929 'tags': ['colorectalcancer', 'cancerjourney', 'imnotaquitter'],
930 'upload_date': '20221208',
932 'uploader': 'Jocelyn Laidlaw',
933 'uploader_url': 'https://twitter.com/CTVJLaidlaw',
937 # URL specifies video number and --no-playlist
938 'url': 'https://twitter.com/CTVJLaidlaw/status/1600649710662213632/video/2',
940 'id': '1600649511827013632',
942 'title': 'md5:7662a0a27ce6faa3e5b160340f3cfab1',
943 'thumbnail': r
're:^https?://.+\.jpg',
944 'timestamp': 1670459604.0,
945 'uploader_id': 'CTVJLaidlaw',
946 'uploader': 'Jocelyn Laidlaw',
948 'comment_count': int,
949 'tags': ['colorectalcancer', 'cancerjourney', 'imnotaquitter'],
951 'uploader_url': 'https://twitter.com/CTVJLaidlaw',
952 'display_id': '1600649710662213632',
955 'description': 'md5:591c19ce66fadc2359725d5cd0d1052c',
956 'upload_date': '20221208',
959 'params': {'noplaylist': True}
,
961 # id pointing to TweetWithVisibilityResults type entity which wraps the actual Tweet over
962 # note the id different between extraction and url
963 'url': 'https://twitter.com/s2FAKER/status/1621117700482416640',
965 'id': '1621117577354424321',
966 'display_id': '1621117700482416640',
968 'title': '뽀 - 아 최우제 이동속도 봐',
969 'description': '아 최우제 이동속도 봐 https://t.co/dxu2U5vXXB',
972 'uploader_id': 's2FAKER',
973 'uploader_url': 'https://twitter.com/s2FAKER',
974 'upload_date': '20230202',
975 'timestamp': 1675339553.0,
976 'thumbnail': r
're:https?://pbs\.twimg\.com/.+',
981 'comment_count': int,
985 'url': 'https://twitter.com/hlo_again/status/1599108751385972737/video/2',
987 'id': '1599108643743473680',
988 'display_id': '1599108751385972737',
990 'title': '\u06ea - \U0001F48B',
991 'uploader_url': 'https://twitter.com/hlo_again',
993 'uploader_id': 'hlo_again',
994 'thumbnail': 'https://pbs.twimg.com/ext_tw_video_thumb/1599108643743473680/pu/img/UG3xjov4rgg5sbYM.jpg?name=orig',
997 'comment_count': int,
999 'upload_date': '20221203',
1001 'timestamp': 1670092210.0,
1003 'uploader': '\u06ea',
1004 'description': '\U0001F48B https://t.co/bTj9Qz7vQP',
1006 'params': {'noplaylist': True}
,
1008 'url': 'https://twitter.com/MunTheShinobi/status/1600009574919962625',
1010 'id': '1600009362759733248',
1011 'display_id': '1600009574919962625',
1013 'uploader_url': 'https://twitter.com/MunTheShinobi',
1014 'description': 'This is a genius ad by Apple. \U0001f525\U0001f525\U0001f525\U0001f525\U0001f525 https://t.co/cNsA0MoOml',
1016 'thumbnail': 'https://pbs.twimg.com/ext_tw_video_thumb/1600009362759733248/pu/img/XVhFQivj75H_YxxV.jpg?name=orig',
1019 'repost_count': int,
1020 'upload_date': '20221206',
1021 'title': 'Mün - This is a genius ad by Apple. \U0001f525\U0001f525\U0001f525\U0001f525\U0001f525',
1022 'comment_count': int,
1025 'uploader_id': 'MunTheShinobi',
1026 'duration': 139.987,
1027 'timestamp': 1670306984.0,
1030 # retweeted_status (private)
1031 'url': 'https://twitter.com/liberdalau/status/1623739803874349067',
1033 'id': '1623274794488659969',
1034 'display_id': '1623739803874349067',
1036 'title': 'Johnny Bullets - Me after going viral to over 30million people: Whoopsie-daisy',
1037 'description': 'md5:b06864cd3dc2554821cc327f5348485a',
1038 'uploader': 'Johnny Bullets',
1039 'uploader_id': 'Johnnybull3ts',
1040 'uploader_url': 'https://twitter.com/Johnnybull3ts',
1044 'timestamp': 1675853859.0,
1045 'upload_date': '20230208',
1046 'thumbnail': r
're:https://pbs\.twimg\.com/ext_tw_video_thumb/.+',
1048 'repost_count': int,
1050 'skip': 'Protected tweet',
1053 'url': 'https://twitter.com/playstrumpcard/status/1695424220702888009',
1055 'id': '1694928337846538240',
1057 'display_id': '1695424220702888009',
1058 'title': 'md5:e8daa9527bc2b947121395494f786d9d',
1059 'description': 'md5:004f2d37fd58737724ec75bc7e679938',
1060 'uploader': 'Benny Johnson',
1061 'uploader_id': 'bennyjohnson',
1062 'uploader_url': 'https://twitter.com/bennyjohnson',
1066 'timestamp': 1692962814.0,
1067 'upload_date': '20230825',
1068 'thumbnail': r
're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
1070 'repost_count': int,
1072 'comment_count': int,
1075 # retweeted_status w/ legacy API
1076 'url': 'https://twitter.com/playstrumpcard/status/1695424220702888009',
1078 'id': '1694928337846538240',
1080 'display_id': '1695424220702888009',
1081 'title': 'md5:e8daa9527bc2b947121395494f786d9d',
1082 'description': 'md5:004f2d37fd58737724ec75bc7e679938',
1083 'uploader': 'Benny Johnson',
1084 'uploader_id': 'bennyjohnson',
1085 'uploader_url': 'https://twitter.com/bennyjohnson',
1089 'timestamp': 1692962814.0,
1090 'upload_date': '20230825',
1091 'thumbnail': r
're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
1093 'repost_count': int,
1095 'params': {'extractor_args': {'twitter': {'api': ['legacy']}
}},
1097 # Broadcast embedded in tweet
1098 'url': 'https://twitter.com/JessicaDobsonWX/status/1693057346933600402',
1100 'id': '1yNGaNLjEblJj',
1102 'title': 'Jessica Dobson - WAVE Weather Now - Saturday 8/19/23 Update',
1103 'uploader': 'Jessica Dobson',
1104 'uploader_id': '1DZEoDwDovRQa',
1105 'thumbnail': r
're:^https?://.*\.jpg',
1108 'add_ie': ['TwitterBroadcast'],
1110 # Animated gif and quote tweet video, with syndication API
1111 'url': 'https://twitter.com/BAKKOOONN/status/1696256659889565950',
1112 'playlist_mincount': 2,
1114 'id': '1696256659889565950',
1115 'title': 'BAKOON - https://t.co/zom968d0a0',
1116 'description': 'https://t.co/zom968d0a0',
1118 'uploader': 'BAKOON',
1119 'uploader_id': 'BAKKOOONN',
1120 'uploader_url': 'https://twitter.com/BAKKOOONN',
1122 'timestamp': 1693254077.0,
1123 'upload_date': '20230828',
1126 'params': {'extractor_args': {'twitter': {'api': ['syndication']}
}},
1127 'expected_warnings': ['Not all metadata'],
1130 'url': 'https://twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid.onion/TwitterBlue/status/1484226494708662273',
1131 'only_matching': True,
1134 'url': 'https://twitter.com/GunB1g/status/1163218564784017422',
1135 'only_matching': True,
1137 # promo_video_website card
1138 'url': 'https://twitter.com/GunB1g/status/1163218564784017422',
1139 'only_matching': True,
1141 # promo_video_convo card
1142 'url': 'https://twitter.com/poco_dandy/status/1047395834013384704',
1143 'only_matching': True,
1146 'url': 'https://twitter.com/poco_dandy/status/1150646424461176832',
1147 'only_matching': True,
1149 # video_direct_message card
1150 'url': 'https://twitter.com/qarev001/status/1348948114569269251',
1151 'only_matching': True,
1153 # poll2choice_video card
1154 'url': 'https://twitter.com/CAF_Online/status/1349365911120195585',
1155 'only_matching': True,
1157 # poll3choice_video card
1158 'url': 'https://twitter.com/SamsungMobileSA/status/1348609186725289984',
1159 'only_matching': True,
1161 # poll4choice_video card
1162 'url': 'https://twitter.com/SouthamptonFC/status/1347577658079641604',
1163 'only_matching': True,
1166 _MEDIA_ID_RE
= re
.compile(r
'_video/(\d+)/')
1169 def _GRAPHQL_ENDPOINT(self
):
1170 if self
.is_logged_in
:
1171 return 'zZXycP0V6H7m-2r0mOnFcA/TweetDetail'
1172 return '2ICDjqPd81tulZcYrtpTuQ/TweetResultByRestId'
1174 def _graphql_to_legacy(self
, data
, twid
):
1175 result
= traverse_obj(data
, (
1176 'threaded_conversation_with_injections_v2', 'instructions', 0, 'entries',
1177 lambda _
, v
: v
['entryId'] == f
'tweet-{twid}', 'content', 'itemContent',
1178 'tweet_results', 'result', ('tweet', None), {dict}
,
1179 ), default
={}, get_all
=False) if self
.is_logged_in
else traverse_obj(
1180 data
, ('tweetResult', 'result', {dict}
), default
={})
1182 if result
.get('__typename') not in ('Tweet', 'TweetTombstone', 'TweetUnavailable', None):
1183 self
.report_warning(f
'Unknown typename: {result.get("__typename")}', twid
, only_once
=True)
1185 if 'tombstone' in result
:
1186 cause
= remove_end(traverse_obj(result
, ('tombstone', 'text', 'text', {str}
)), '. Learn more')
1187 raise ExtractorError(f
'Twitter API says: {cause or "Unknown error"}', expected
=True)
1188 elif result
.get('__typename') == 'TweetUnavailable':
1189 reason
= result
.get('reason')
1190 if reason
== 'NsfwLoggedOut':
1191 self
.raise_login_required('NSFW tweet requires authentication')
1192 elif reason
== 'Protected':
1193 self
.raise_login_required('You are not authorized to view this protected tweet')
1194 raise ExtractorError(reason
or 'Requested tweet is unavailable', expected
=True)
1196 status
= result
.get('legacy', {})
1197 status
.update(traverse_obj(result
, {
1198 'user': ('core', 'user_results', 'result', 'legacy'),
1199 'card': ('card', 'legacy'),
1200 'quoted_status': ('quoted_status_result', 'result', 'legacy'),
1201 'retweeted_status': ('legacy', 'retweeted_status_result', 'result', 'legacy'),
1202 }, expected_type
=dict, default
={}))
1204 # extra transformations needed since result does not match legacy format
1205 if status
.get('retweeted_status'):
1206 status
['retweeted_status']['user'] = traverse_obj(status
, (
1207 'retweeted_status_result', 'result', 'core', 'user_results', 'result', 'legacy', {dict}
)) or {}
1210 binding_value
.get('key'): binding_value
.get('value')
1211 for binding_value
in traverse_obj(status
, ('card', 'binding_values', ..., {dict}
))
1214 status
['card']['binding_values'] = binding_values
1218 def _build_graphql_query(self
, media_id
):
1221 'focalTweetId': media_id
,
1222 'includePromotedContent': True,
1223 'with_rux_injections': False,
1224 'withBirdwatchNotes': True,
1225 'withCommunity': True,
1226 'withDownvotePerspective': False,
1227 'withQuickPromoteEligibilityTweetFields': True,
1228 'withReactionsMetadata': False,
1229 'withReactionsPerspective': False,
1230 'withSuperFollowsTweetFields': True,
1231 'withSuperFollowsUserFields': True,
1232 'withV2Timeline': True,
1236 'graphql_is_translatable_rweb_tweet_is_translatable_enabled': False,
1237 'interactive_text_enabled': True,
1238 'responsive_web_edit_tweet_api_enabled': True,
1239 'responsive_web_enhance_cards_enabled': True,
1240 'responsive_web_graphql_timeline_navigation_enabled': False,
1241 'responsive_web_text_conversations_enabled': False,
1242 'responsive_web_uc_gql_enabled': True,
1243 'standardized_nudges_misinfo': True,
1244 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': False,
1245 'tweetypie_unmention_optimization_enabled': True,
1246 'unified_cards_ad_metadata_container_dynamic_card_content_query_enabled': True,
1247 'verified_phone_label_enabled': False,
1248 'vibe_api_enabled': True,
1250 } if self
.is_logged_in
else {
1252 'tweetId': media_id
,
1253 'withCommunity': False,
1254 'includePromotedContent': False,
1258 'creator_subscriptions_tweet_preview_api_enabled': True,
1259 'tweetypie_unmention_optimization_enabled': True,
1260 'responsive_web_edit_tweet_api_enabled': True,
1261 'graphql_is_translatable_rweb_tweet_is_translatable_enabled': True,
1262 'view_counts_everywhere_api_enabled': True,
1263 'longform_notetweets_consumption_enabled': True,
1264 'responsive_web_twitter_article_tweet_consumption_enabled': False,
1265 'tweet_awards_web_tipping_enabled': False,
1266 'freedom_of_speech_not_reach_fetch_enabled': True,
1267 'standardized_nudges_misinfo': True,
1268 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': True,
1269 'longform_notetweets_rich_text_read_enabled': True,
1270 'longform_notetweets_inline_media_enabled': True,
1271 'responsive_web_graphql_exclude_directive_enabled': True,
1272 'verified_phone_label_enabled': False,
1273 'responsive_web_media_download_video_enabled': False,
1274 'responsive_web_graphql_skip_user_profile_image_extensions_enabled': False,
1275 'responsive_web_graphql_timeline_navigation_enabled': True,
1276 'responsive_web_enhance_cards_enabled': False
1279 'withArticleRichContentState': False
1283 def _extract_status(self
, twid
):
1284 if self
.is_logged_in
or self
._selected
_api
== 'graphql':
1285 status
= self
._graphql
_to
_legacy
(self
._call
_graphql
_api
(self
._GRAPHQL
_ENDPOINT
, twid
), twid
)
1287 elif self
._selected
_api
== 'legacy':
1288 status
= self
._call
_api
(f
'statuses/show/{twid}.json', twid
, {
1289 'cards_platform': 'Web-12',
1291 'include_reply_count': 1,
1292 'include_user_entities': 0,
1293 'tweet_mode': 'extended',
1296 elif self
._selected
_api
== 'syndication':
1297 self
.report_warning(
1298 'Not all metadata or media is available via syndication endpoint', twid
, only_once
=True)
1299 status
= self
._download
_json
(
1300 'https://cdn.syndication.twimg.com/tweet-result', twid
, 'Downloading syndication JSON',
1301 headers
={'User-Agent': 'Googlebot'}
, query
={
1303 # TODO: token = ((Number(twid) / 1e15) * Math.PI).toString(36).replace(/(0+|\.)/g, '')
1304 'token': ''.join(random
.choices('123456789abcdefghijklmnopqrstuvwxyz', k
=10)),
1307 raise ExtractorError('Syndication endpoint returned empty JSON response')
1308 # Transform the result so its structure matches that of legacy/graphql
1310 for detail
in traverse_obj(status
, ((None, 'quoted_tweet'), 'mediaDetails', ..., {dict}
)):
1311 detail
['id_str'] = traverse_obj(detail
, (
1312 'video_info', 'variants', ..., 'url', {self._MEDIA_ID_RE.search}
, 1), get_all
=False) or twid
1313 media
.append(detail
)
1314 status
['extended_entities'] = {'media': media}
1317 raise ExtractorError(f
'"{self._selected_api}" is not a valid API selection', expected
=True)
1319 return traverse_obj(status
, 'retweeted_status', None, expected_type
=dict) or {}
1321 def _real_extract(self
, url
):
1322 twid
, selected_index
= self
._match
_valid
_url
(url
).group('id', 'index')
1323 status
= self
._extract
_status
(twid
)
1325 title
= description
= traverse_obj(
1326 status
, (('full_text', 'text'), {lambda x: x.replace('\n', ' ')}
), get_all
=False) or ''
1327 # strip 'https -_t.co_BJYgOjSeGA' junk from filenames
1328 title
= re
.sub(r
'\s+(https?://[^ ]+)', '', title
)
1329 user
= status
.get('user') or {}
1330 uploader
= user
.get('name')
1332 title
= f
'{uploader} - {title}'
1333 uploader_id
= user
.get('screen_name')
1338 'description': description
,
1339 'uploader': uploader
,
1340 'timestamp': unified_timestamp(status
.get('created_at')),
1341 'uploader_id': uploader_id
,
1342 'uploader_url': format_field(uploader_id
, None, 'https://twitter.com/%s'),
1343 'like_count': int_or_none(status
.get('favorite_count')),
1344 'repost_count': int_or_none(status
.get('retweet_count')),
1345 'comment_count': int_or_none(status
.get('reply_count')),
1346 'age_limit': 18 if status
.get('possibly_sensitive') else 0,
1347 'tags': traverse_obj(status
, ('entities', 'hashtags', ..., 'text')),
1350 def extract_from_video_info(media
):
1351 media_id
= traverse_obj(media
, 'id_str', 'id', expected_type
=str_or_none
)
1352 self
.write_debug(f
'Extracting from video info: {media_id}')
1356 for variant
in traverse_obj(media
, ('video_info', 'variants', ...)):
1357 fmts
, subs
= self
._extract
_variant
_formats
(variant
, twid
)
1358 subtitles
= self
._merge
_subtitles
(subtitles
, subs
)
1359 formats
.extend(fmts
)
1362 media_url
= media
.get('media_url_https') or media
.get('media_url')
1364 def add_thumbnail(name
, size
):
1367 'url': update_url_query(media_url
, {'name': name}
),
1368 'width': int_or_none(size
.get('w') or size
.get('width')),
1369 'height': int_or_none(size
.get('h') or size
.get('height')),
1371 for name
, size
in media
.get('sizes', {}).items():
1372 add_thumbnail(name
, size
)
1373 add_thumbnail('orig', media
.get('original_info') or {})
1378 'subtitles': subtitles
,
1379 'thumbnails': thumbnails
,
1380 'view_count': traverse_obj(media
, ('mediaStats', 'viewCount', {int_or_none}
)),
1381 'duration': float_or_none(traverse_obj(media
, ('video_info', 'duration_millis')), 1000),
1382 # The codec of http formats are unknown
1383 '_format_sort_fields': ('res', 'br', 'size', 'proto'),
1386 def extract_from_card_info(card
):
1390 self
.write_debug(f
'Extracting from card info: {card.get("url")}')
1391 binding_values
= card
['binding_values']
1393 def get_binding_value(k
):
1394 o
= binding_values
.get(k
) or {}
1395 return try_get(o
, lambda x
: x
[x
['type'].lower() + '_value'])
1397 card_name
= card
['name'].split(':')[-1]
1398 if card_name
== 'player':
1401 'url': get_binding_value('player_url'),
1403 elif card_name
== 'periscope_broadcast':
1406 'url': get_binding_value('url') or get_binding_value('player_url'),
1407 'ie_key': PeriscopeIE
.ie_key(),
1409 elif card_name
== 'broadcast':
1412 'url': get_binding_value('broadcast_url'),
1413 'ie_key': TwitterBroadcastIE
.ie_key(),
1415 elif card_name
== 'audiospace':
1418 'url': f
'https://twitter.com/i/spaces/{get_binding_value("id")}',
1419 'ie_key': TwitterSpacesIE
.ie_key(),
1421 elif card_name
== 'summary':
1424 'url': get_binding_value('card_url'),
1426 elif card_name
== 'unified_card':
1427 unified_card
= self
._parse
_json
(get_binding_value('unified_card'), twid
)
1428 yield from map(extract_from_video_info
, traverse_obj(
1429 unified_card
, ('media_entities', ...), expected_type
=dict))
1430 # amplify, promo_video_website, promo_video_convo, appplayer,
1431 # video_direct_message, poll2choice_video, poll3choice_video,
1432 # poll4choice_video, ...
1434 is_amplify
= card_name
== 'amplify'
1435 vmap_url
= get_binding_value('amplify_url_vmap') if is_amplify
else get_binding_value('player_stream_url')
1436 content_id
= get_binding_value('%s_content_id' % (card_name
if is_amplify
else 'player'))
1437 formats
, subtitles
= self
._extract
_formats
_from
_vmap
_url
(vmap_url
, content_id
or twid
)
1440 for suffix
in ('_small', '', '_large', '_x_large', '_original'):
1441 image
= get_binding_value('player_image' + suffix
) or {}
1442 image_url
= image
.get('url')
1443 if not image_url
or '/player-placeholder' in image_url
:
1446 'id': suffix
[1:] if suffix
else 'medium',
1448 'width': int_or_none(image
.get('width')),
1449 'height': int_or_none(image
.get('height')),
1454 'subtitles': subtitles
,
1455 'thumbnails': thumbnails
,
1456 'duration': int_or_none(get_binding_value(
1457 'content_duration_seconds')),
1460 videos
= traverse_obj(status
, (
1461 (None, 'quoted_status'), 'extended_entities', 'media', lambda _
, m
: m
['type'] != 'photo', {dict}
))
1463 if self
._yes
_playlist
(twid
, selected_index
, video_label
='URL-specified video number'):
1464 selected_entries
= (*map(extract_from_video_info
, videos
), *extract_from_card_info(status
.get('card')))
1466 desired_obj
= traverse_obj(status
, (
1467 (None, 'quoted_status'), 'extended_entities', 'media', int(selected_index
) - 1, {dict}
), get_all
=False)
1469 raise ExtractorError(f
'Video #{selected_index} is unavailable', expected
=True)
1470 elif desired_obj
.get('type') != 'video':
1471 raise ExtractorError(f
'Media #{selected_index} is not a video', expected
=True)
1473 # Restore original archive id and video index in title
1474 for index
, entry
in enumerate(videos
, 1):
1475 if entry
.get('id') != desired_obj
.get('id'):
1478 info
['_old_archive_ids'] = [make_archive_id(self
, twid
)]
1479 if len(videos
) != 1:
1480 info
['title'] += f
' #{index}'
1483 return {**info, **extract_from_video_info(desired_obj), 'display_id': twid}
1485 entries
= [{**info, **data, 'display_id': twid}
for data
in selected_entries
]
1487 expanded_url
= traverse_obj(status
, ('entities', 'urls', 0, 'expanded_url'), expected_type
=url_or_none
)
1488 if not expanded_url
or expanded_url
== url
:
1489 self
.raise_no_formats('No video could be found in this tweet', expected
=True)
1492 return self
.url_result(expanded_url
, display_id
=twid
, **info
)
1494 entries
[0]['_old_archive_ids'] = [make_archive_id(self
, twid
)]
1496 if len(entries
) == 1:
1499 for index
, entry
in enumerate(entries
, 1):
1500 entry
['title'] += f
' #{index}'
1502 return self
.playlist_result(entries
, **info
)
1505 class TwitterAmplifyIE(TwitterBaseIE
):
1506 IE_NAME
= 'twitter:amplify'
1507 _VALID_URL
= r
'https?://amp\.twimg\.com/v/(?P<id>[0-9a-f\-]{36})'
1510 'url': 'https://amp.twimg.com/v/0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
1511 'md5': 'fec25801d18a4557c5c9f33d2c379ffa',
1513 'id': '0ba0c3c7-0af3-4c0a-bed5-7efd1ffa2951',
1515 'title': 'Twitter Video',
1516 'thumbnail': 're:^https?://.*',
1518 'params': {'format': '[protocol=https]'}
,
1521 def _real_extract(self
, url
):
1522 video_id
= self
._match
_id
(url
)
1523 webpage
= self
._download
_webpage
(url
, video_id
)
1525 vmap_url
= self
._html
_search
_meta
(
1526 'twitter:amplify:vmap', webpage
, 'vmap url')
1527 formats
, _
= self
._extract
_formats
_from
_vmap
_url
(vmap_url
, video_id
)
1530 thumbnail
= self
._html
_search
_meta
(
1531 'twitter:image:src', webpage
, 'thumbnail', fatal
=False)
1533 def _find_dimension(target
):
1534 w
= int_or_none(self
._html
_search
_meta
(
1535 'twitter:%s:width' % target
, webpage
, fatal
=False))
1536 h
= int_or_none(self
._html
_search
_meta
(
1537 'twitter:%s:height' % target
, webpage
, fatal
=False))
1541 thumbnail_w
, thumbnail_h
= _find_dimension('image')
1544 'width': thumbnail_w
,
1545 'height': thumbnail_h
,
1548 video_w
, video_h
= _find_dimension('player')
1556 'title': 'Twitter Video',
1558 'thumbnails': thumbnails
,
1562 class TwitterBroadcastIE(TwitterBaseIE
, PeriscopeBaseIE
):
1563 IE_NAME
= 'twitter:broadcast'
1564 _VALID_URL
= TwitterBaseIE
._BASE
_REGEX
+ r
'i/broadcasts/(?P<id>[0-9a-zA-Z]{13})'
1567 # untitled Periscope video
1568 'url': 'https://twitter.com/i/broadcasts/1yNGaQLWpejGj',
1570 'id': '1yNGaQLWpejGj',
1572 'title': 'Andrea May Sahouri - Periscope Broadcast',
1573 'uploader': 'Andrea May Sahouri',
1574 'uploader_id': '1PXEdBZWpGwKe',
1575 'thumbnail': r
're:^https?://[^?#]+\.jpg\?token=',
1580 def _real_extract(self
, url
):
1581 broadcast_id
= self
._match
_id
(url
)
1582 broadcast
= self
._call
_api
(
1583 'broadcasts/show.json', broadcast_id
,
1584 {'ids': broadcast_id}
)['broadcasts'][broadcast_id
]
1586 raise ExtractorError('Broadcast no longer exists', expected
=True)
1587 info
= self
._parse
_broadcast
_data
(broadcast
, broadcast_id
)
1588 media_key
= broadcast
['media_key']
1589 source
= self
._call
_api
(
1590 f
'live_video_stream/status/{media_key}', media_key
)['source']
1591 m3u8_url
= source
.get('noRedirectPlaybackUrl') or source
['location']
1592 if '/live_video_stream/geoblocked/' in m3u8_url
:
1593 self
.raise_geo_restricted()
1594 m3u8_id
= compat_parse_qs(compat_urllib_parse_urlparse(
1595 m3u8_url
).query
).get('type', [None])[0]
1596 state
, width
, height
= self
._extract
_common
_format
_info
(broadcast
)
1597 info
['formats'] = self
._extract
_pscp
_m
3u8_formats
(
1598 m3u8_url
, broadcast_id
, m3u8_id
, state
, width
, height
)
1602 class TwitterSpacesIE(TwitterBaseIE
):
1603 IE_NAME
= 'twitter:spaces'
1604 _VALID_URL
= TwitterBaseIE
._BASE
_REGEX
+ r
'i/spaces/(?P<id>[0-9a-zA-Z]{13})'
1607 'url': 'https://twitter.com/i/spaces/1RDxlgyvNXzJL',
1609 'id': '1RDxlgyvNXzJL',
1611 'title': 'King Carlo e la mossa Kansas City per fare il Grande Centro',
1612 'description': 'Twitter Space participated by annarita digiorgio, Signor Ernesto, Raffaello Colosimo, Simone M. Sepe',
1613 'uploader': r
're:Lucio Di Gaetano.*?',
1614 'uploader_id': 'luciodigaetano',
1615 'live_status': 'was_live',
1616 'timestamp': 1659877956,
1617 'upload_date': '20220807',
1618 'release_timestamp': 1659904215,
1619 'release_date': '20220807',
1621 'params': {'skip_download': 'm3u8'}
,
1623 # post_live/TimedOut but downloadable
1624 'url': 'https://twitter.com/i/spaces/1vAxRAVQWONJl',
1626 'id': '1vAxRAVQWONJl',
1628 'title': 'Framing Up FinOps: Billing Tools',
1629 'description': 'Twitter Space participated by rupa, Alfonso Hernandez',
1630 'uploader': 'Google Cloud',
1631 'uploader_id': 'googlecloud',
1632 'live_status': 'post_live',
1633 'timestamp': 1681409554,
1634 'upload_date': '20230413',
1635 'release_timestamp': 1681839000,
1636 'release_date': '20230418',
1638 'params': {'skip_download': 'm3u8'}
,
1640 # Needs ffmpeg as downloader, see: https://github.com/yt-dlp/yt-dlp/issues/7536
1641 'url': 'https://twitter.com/i/spaces/1eaKbrQbjoRKX',
1643 'id': '1eaKbrQbjoRKX',
1646 'description': 'Twitter Space participated by nobody yet',
1647 'uploader': '息根とめる🔪Twitchで復活',
1648 'uploader_id': 'tomeru_ikinone',
1649 'live_status': 'was_live',
1650 'timestamp': 1685617198,
1651 'upload_date': '20230601',
1653 'params': {'skip_download': 'm3u8'}
,
1657 'notstarted': 'is_upcoming',
1658 'ended': 'was_live',
1659 'running': 'is_live',
1660 'timedout': 'post_live',
1663 def _build_graphql_query(self
, space_id
):
1667 'isMetatagsQuery': True,
1668 'withDownvotePerspective': False,
1669 'withReactionsMetadata': False,
1670 'withReactionsPerspective': False,
1671 'withReplays': True,
1672 'withSuperFollowsUserFields': True,
1673 'withSuperFollowsTweetFields': True,
1676 'dont_mention_me_view_api_enabled': True,
1677 'interactive_text_enabled': True,
1678 'responsive_web_edit_tweet_api_enabled': True,
1679 'responsive_web_enhance_cards_enabled': True,
1680 'responsive_web_uc_gql_enabled': True,
1681 'spaces_2022_h2_clipping': True,
1682 'spaces_2022_h2_spaces_communities': False,
1683 'standardized_nudges_misinfo': True,
1684 'tweet_with_visibility_results_prefer_gql_limited_actions_policy_enabled': False,
1685 'vibe_api_enabled': True,
1689 def _real_extract(self
, url
):
1690 space_id
= self
._match
_id
(url
)
1691 if not self
.is_logged_in
:
1692 self
.raise_login_required('Twitter Spaces require authentication')
1693 space_data
= self
._call
_graphql
_api
('HPEisOmj1epUNLCWTYhUWw/AudioSpaceById', space_id
)['audioSpace']
1695 raise ExtractorError('Twitter Space not found', expected
=True)
1697 metadata
= space_data
['metadata']
1698 live_status
= try_call(lambda: self
.SPACE_STATUS
[metadata
['state'].lower()])
1699 is_live
= live_status
== 'is_live'
1702 headers
= {'Referer': 'https://twitter.com/'}
1703 if live_status
== 'is_upcoming':
1704 self
.raise_no_formats('Twitter Space not started yet', expected
=True)
1705 elif not is_live
and not metadata
.get('is_space_available_for_replay'):
1706 self
.raise_no_formats('Twitter Space ended and replay is disabled', expected
=True)
1707 elif metadata
.get('media_key'):
1708 source
= traverse_obj(
1709 self
._call
_api
(f
'live_video_stream/status/{metadata["media_key"]}', metadata
['media_key']),
1710 ('source', ('noRedirectPlaybackUrl', 'location'), {url_or_none}
), get_all
=False)
1711 formats
= self
._extract
_m
3u8_formats
( # XXX: Some Spaces need ffmpeg as downloader
1712 source
, metadata
['media_key'], 'm4a', entry_protocol
='m3u8', live
=is_live
,
1713 headers
=headers
, fatal
=False) if source
else []
1715 fmt
.update({'vcodec': 'none', 'acodec': 'aac'}
)
1717 fmt
['container'] = 'm4a_dash'
1719 participants
= ', '.join(traverse_obj(
1720 space_data
, ('participants', 'speakers', ..., 'display_name'))) or 'nobody yet'
1722 if not formats
and live_status
== 'post_live':
1723 self
.raise_no_formats('Twitter Space ended but not downloadable yet', expected
=True)
1727 'title': metadata
.get('title'),
1728 'description': f
'Twitter Space participated by {participants}',
1729 'uploader': traverse_obj(
1730 metadata
, ('creator_results', 'result', 'legacy', 'name')),
1731 'uploader_id': traverse_obj(
1732 metadata
, ('creator_results', 'result', 'legacy', 'screen_name')),
1733 'live_status': live_status
,
1734 'release_timestamp': try_call(
1735 lambda: int_or_none(metadata
['scheduled_start'], scale
=1000)),
1736 'timestamp': int_or_none(metadata
.get('created_at'), scale
=1000),
1738 'http_headers': headers
,
1742 class TwitterShortenerIE(TwitterBaseIE
):
1743 IE_NAME
= 'twitter:shortener'
1744 _VALID_URL
= r
'https?://t.co/(?P<id>[^?]+)|tco:(?P<eid>[^?]+)'
1745 _BASE_URL
= 'https://t.co/'
1747 def _real_extract(self
, url
):
1748 mobj
= self
._match
_valid
_url
(url
)
1749 eid
, id = mobj
.group('eid', 'id')
1752 url
= self
._BASE
_URL
+ id
1753 new_url
= self
._request
_webpage
(url
, id, headers
={'User-Agent': 'curl'}
).url
1754 __UNSAFE_LINK
= "https://twitter.com/safety/unsafe_link_warning?unsafe_link="
1755 if new_url
.startswith(__UNSAFE_LINK
):
1756 new_url
= new_url
.replace(__UNSAFE_LINK
, "")
1757 return self
.url_result(new_url
)