+import functools
import json
+import random
import re
+import urllib.parse
from .common import InfoExtractor
from .periscope import PeriscopeBaseIE, PeriscopeIE
-from ..compat import (
- compat_parse_qs,
- compat_urllib_parse_unquote,
- compat_urllib_parse_urlparse,
-)
+from ..networking.exceptions import HTTPError
from ..utils import (
ExtractorError,
dict_get,
class TwitterBaseIE(InfoExtractor):
_NETRC_MACHINE = 'twitter'
- _API_BASE = 'https://api.twitter.com/1.1/'
- _GRAPHQL_API_BASE = 'https://twitter.com/i/api/graphql/'
- _BASE_REGEX = r'https?://(?:(?:www|m(?:obile)?)\.)?(?:twitter\.com|twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid\.onion)/'
+ _API_BASE = 'https://api.x.com/1.1/'
+ _GRAPHQL_API_BASE = 'https://x.com/i/api/graphql/'
+ _BASE_REGEX = r'https?://(?:(?:www|m(?:obile)?)\.)?(?:(?:twitter|x)\.com|twitter3e4tixl4xyajtrzo62zg5vztmjuricljdp2c5kshju4avyoid\.onion)/'
_AUTH = 'AAAAAAAAAAAAAAAAAAAAANRILgAAAAAAnNwIzUejRCOuH5E6I8xnZz4puTs%3D1Zv7ttfk8LF81IUq16cHjhLTvJu4FA33AGWWjCpTnA'
_LEGACY_AUTH = 'AAAAAAAAAAAAAAAAAAAAAIK1zgAAAAAA2tUWuhGZ2JceoId5GwYWU5GspY4%3DUq7gzFoCZs1QfwGoVdvSac3IniczZEYXIcDyumCauIXpcAPorE'
_flow_token = None
'flow_context': {
'debug_overrides': {},
'start_location': {
- 'location': 'unknown'
- }
- }
+ 'location': 'unknown',
+ },
+ },
},
'subtask_versions': {
'action_list': 2,
'user_recommendations_list': 4,
'user_recommendations_urt': 1,
'wait_spinner': 3,
- 'web_modal': 1
- }
+ 'web_modal': 1,
+ },
}, separators=(',', ':')).encode()
def _extract_variant_formats(self, variant, video_id):
if not variant_url:
return [], {}
elif '.m3u8' in variant_url:
- return self._extract_m3u8_formats_and_subtitles(
+ fmts, subs = self._extract_m3u8_formats_and_subtitles(
variant_url, video_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False)
+ for f in traverse_obj(fmts, lambda _, v: v['vcodec'] == 'none' and v.get('tbr') is None):
+ if mobj := re.match(r'hls-[Aa]udio-(?P<bitrate>\d{4,})', f['format_id']):
+ f['tbr'] = int_or_none(mobj.group('bitrate'), 1000)
+ return fmts, subs
else:
tbr = int_or_none(dict_get(variant, ('bitrate', 'bit_rate')), 1000) or None
f = {
'url': variant_url,
- 'format_id': 'http' + ('-%d' % tbr if tbr else ''),
+ 'format_id': 'http' + (f'-{tbr}' if tbr else ''),
'tbr': tbr,
}
self._search_dimensions_in_video_url(f, variant_url)
subtitles = {}
urls = []
for video_variant in vmap_data.findall('.//{http://twitter.com/schema/videoVMapV2.xsd}videoVariant'):
- video_variant.attrib['url'] = compat_urllib_parse_unquote(
+ video_variant.attrib['url'] = urllib.parse.unquote(
video_variant.attrib['url'])
urls.append(video_variant.attrib['url'])
fmts, subs = self._extract_variant_formats(
def is_logged_in(self):
return bool(self._get_cookies(self._API_BASE).get('auth_token'))
+ # XXX: Temporary workaround until twitter.com => x.com migration is completed
+ def _real_initialize(self):
+ if self.is_logged_in or not self._get_cookies('https://twitter.com/').get('auth_token'):
+ return
+ # User has not yet been migrated to x.com and has passed twitter.com cookies
+ TwitterBaseIE._API_BASE = 'https://api.twitter.com/1.1/'
+ TwitterBaseIE._GRAPHQL_API_BASE = 'https://twitter.com/i/api/graphql/'
+
+ @functools.cached_property
+ def _selected_api(self):
+ return self._configuration_arg('api', ['graphql'], ie_key='Twitter')[0]
+
def _fetch_guest_token(self, display_id):
guest_token = traverse_obj(self._download_json(
f'{self._API_BASE}guest/activate.json', display_id, 'Downloading guest token', data=b'',
- headers=self._set_base_headers(legacy=display_id and self._configuration_arg('legacy_api'))),
+ headers=self._set_base_headers(legacy=display_id and self._selected_api == 'legacy')),
('guest_token', {str}))
if not guest_token:
raise ExtractorError('Could not retrieve guest token')
if self.is_logged_in:
return
- webpage = self._download_webpage('https://twitter.com/', None, 'Downloading login page')
- guest_token = self._search_regex(
- r'\.cookie\s*=\s*["\']gt=(\d+);', webpage, 'gt', default=None) or self._fetch_guest_token(None)
+ guest_token = self._fetch_guest_token(None)
headers = {
**self._set_base_headers(),
'content-type': 'application/json',
'x-guest-token': guest_token,
'x-twitter-client-language': 'en',
'x-twitter-active-user': 'yes',
- 'Referer': 'https://twitter.com/',
- 'Origin': 'https://twitter.com',
+ 'Referer': 'https://x.com/',
+ 'Origin': 'https://x.com',
}
def build_login_json(*subtask_inputs):
return json.dumps({
'flow_token': self._flow_token,
- 'subtask_inputs': subtask_inputs
+ 'subtask_inputs': subtask_inputs,
}, separators=(',', ':')).encode()
def input_dict(subtask_id, text):
'subtask_id': subtask_id,
'enter_text': {
'text': text,
- 'link': 'next_link'
- }
+ 'link': 'next_link',
+ },
}
next_subtask = self._call_login_api(
'subtask_id': next_subtask,
'js_instrumentation': {
'response': '{}',
- 'link': 'next_link'
- }
+ 'link': 'next_link',
+ },
}))
elif next_subtask == 'LoginEnterUserIdentifierSSO':
'key': 'user_identifier',
'response_data': {
'text_data': {
- 'result': username
- }
- }
+ 'result': username,
+ },
+ },
}],
- 'link': 'next_link'
- }
+ 'link': 'next_link',
+ },
}))
elif next_subtask == 'LoginEnterAlternateIdentifierSubtask':
'subtask_id': next_subtask,
'enter_password': {
'password': password,
- 'link': 'next_link'
- }
+ 'link': 'next_link',
+ },
}))
elif next_subtask == 'AccountDuplicationCheck':
'Submitting account duplication check', headers, data=build_login_json({
'subtask_id': next_subtask,
'check_logged_in_account': {
- 'link': 'AccountDuplicationCheck_false'
- }
+ 'link': 'AccountDuplicationCheck_false',
+ },
}))
elif next_subtask == 'LoginTwoFactorAuthChallenge':
'Submitting confirmation code', headers, data=build_login_json(input_dict(
next_subtask, self._get_tfa_info('confirmation code sent to your email or phone'))))
+ elif next_subtask == 'ArkoseLogin':
+ self.raise_login_required('Twitter is requiring captcha for this login attempt', method='cookies')
+
+ elif next_subtask == 'DenyLoginSubtask':
+ self.raise_login_required('Twitter rejected this login attempt as suspicious', method='cookies')
+
elif next_subtask == 'LoginSuccessSubtask':
raise ExtractorError('Twitter API did not grant auth token cookie')
self.report_login()
def _call_api(self, path, video_id, query={}, graphql=False):
- headers = self._set_base_headers(legacy=not graphql and self._configuration_arg('legacy_api'))
+ headers = self._set_base_headers(legacy=not graphql and self._selected_api == 'legacy')
headers.update({
'x-twitter-auth-type': 'OAuth2Session',
'x-twitter-client-language': 'en',
'x-twitter-active-user': 'yes',
} if self.is_logged_in else {
- 'x-guest-token': self._fetch_guest_token(video_id)
+ 'x-guest-token': self._fetch_guest_token(video_id),
})
allowed_status = {400, 401, 403, 404} if graphql else {403}
result = self._download_json(
if result.get('errors'):
errors = ', '.join(set(traverse_obj(result, ('errors', ..., 'message', {str}))))
- raise ExtractorError(
- f'Error(s) while querying API: {errors or "Unknown error"}', expected=True)
+ if errors and 'not authorized' in errors:
+ self.raise_login_required(remove_end(errors, '.'))
+ raise ExtractorError(f'Error(s) while querying API: {errors or "Unknown error"}')
return result
'repost_count': int,
'tags': ['PlutoFlyby'],
},
- 'params': {'format': '[protocol=https]'}
+ 'params': {'format': '[protocol=https]'},
},
{
'url': 'https://twitter.com/i/cards/tfw/v1/654001591733886977',
'title': 'FREE THE NIPPLE - FTN supporters on Hollywood Blvd today!',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'FTN supporters on Hollywood Blvd today! http://t.co/c7jHH749xJ',
+ 'channel_id': '549749560',
'uploader': 'FREE THE NIPPLE',
'uploader_id': 'freethenipple',
'duration': 12.922,
'comment_count': int,
'repost_count': int,
'like_count': int,
- 'view_count': int,
'tags': [],
'age_limit': 18,
+ '_old_archive_ids': ['twitter 643211948184596480'],
},
+ 'skip': 'Requires authentication',
}, {
'url': 'https://twitter.com/giphz/status/657991469417025536/photo/1',
'md5': 'f36dcd5fb92bf7057f155e7d927eeb42',
'ext': 'mp4',
'title': r're:Star Wars.*A new beginning is coming December 18.*',
'description': 'A new beginning is coming December 18. Watch the official 60 second #TV spot for #StarWars: #TheForceAwakens. https://t.co/OkSqT2fjWJ',
+ 'channel_id': '20106852',
'uploader_id': 'starwars',
'uploader': r're:Star Wars.*',
'timestamp': 1447395772,
'like_count': int,
'tags': ['TV', 'StarWars', 'TheForceAwakens'],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 665052190608723968'],
},
}, {
'url': 'https://twitter.com/BTNBrentYarina/status/705235433198714880',
'title': 'jaydin donte geer - BEAT PROD: @suhmeduh #Damndaniel',
'description': 'BEAT PROD: @suhmeduh https://t.co/HBrQ4AfpvZ #Damndaniel https://t.co/byBooq2ejZ',
'thumbnail': r're:^https?://.*\.jpg',
+ 'channel_id': '1383165541',
'uploader': 'jaydin donte geer',
'uploader_id': 'jaydingeer',
'duration': 30.0,
'comment_count': int,
'repost_count': int,
'like_count': int,
- 'view_count': int,
'tags': ['Damndaniel'],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 700207533655363584'],
},
}, {
'url': 'https://twitter.com/Filmdrunk/status/713801302971588609',
'ext': 'mp4',
'title': 'Captain America - @King0fNerd Are you sure you made the right choice? Find out in theaters.',
'description': '@King0fNerd Are you sure you made the right choice? Find out in theaters. https://t.co/GpgYi9xMJI',
+ 'channel_id': '701615052',
'uploader_id': 'CaptainAmerica',
'uploader': 'Captain America',
'duration': 3.17,
'comment_count': int,
'repost_count': int,
'like_count': int,
- 'view_count': int,
'tags': [],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 719944021058060289'],
},
}, {
'url': 'https://twitter.com/OPP_HSD/status/779210622571536384',
'thumbnail': r're:^https?://.*\.jpg',
},
'add_ie': ['Periscope'],
+ 'skip': 'Broadcast not found',
}, {
# has mp4 formats via mobile API
'url': 'https://twitter.com/news_al3alm/status/852138619213144067',
'info_dict': {
- 'id': '852138619213144067',
+ 'id': '852077943283097602',
'ext': 'mp4',
'title': 'عالم الأخبار - كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة',
'description': 'كلمة تاريخية بجلسة الجناسي التاريخية.. النائب خالد مؤنس العتيبي للمعارضين : اتقوا الله .. الظلم ظلمات يوم القيامة https://t.co/xg6OhpyKfN',
+ 'channel_id': '2526757026',
'uploader': 'عالم الأخبار',
'uploader_id': 'news_al3alm',
'duration': 277.4,
'timestamp': 1492000653,
'upload_date': '20170412',
+ 'display_id': '852138619213144067',
+ 'age_limit': 0,
+ 'uploader_url': 'https://twitter.com/news_al3alm',
+ 'thumbnail': r're:^https?://.*\.jpg',
+ 'tags': [],
+ 'repost_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ '_old_archive_ids': ['twitter 852138619213144067'],
},
- 'skip': 'Account suspended',
}, {
'url': 'https://twitter.com/i/web/status/910031516746514432',
'info_dict': {
'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.',
'thumbnail': r're:^https?://.*\.jpg',
'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',
+ 'channel_id': '2319432498',
'uploader': 'Préfet de Guadeloupe',
'uploader_id': 'Prefet971',
'duration': 47.48,
'comment_count': int,
'repost_count': int,
'like_count': int,
- 'view_count': int,
'tags': ['Maria'],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 910031516746514432'],
},
'params': {
'skip_download': True, # requires ffmpeg
'title': 're:.*?Shep is on a roll today.*?',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'md5:37b9f2ff31720cef23b2bd42ee8a0f09',
+ 'channel_id': '255036353',
'uploader': 'Lis Power',
'uploader_id': 'LisPower1',
'duration': 111.278,
'comment_count': int,
'repost_count': int,
'like_count': int,
- 'view_count': int,
'tags': [],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 1001551623938805763'],
},
'params': {
'skip_download': True, # requires ffmpeg
'id': '1087791272830607360',
'display_id': '1087791357756956680',
'ext': 'mp4',
- '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!',
+ '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!',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'md5:6dfd341a3310fb97d80d2bf7145df976',
- 'uploader': 'Twitter',
- 'uploader_id': 'Twitter',
+ 'uploader': 'X',
+ 'uploader_id': 'X',
'duration': 61.567,
'timestamp': 1548184644,
'upload_date': '20190122',
- 'uploader_url': 'https://twitter.com/Twitter',
+ 'uploader_url': 'https://twitter.com/X',
'comment_count': int,
'repost_count': int,
'like_count': int,
'tags': [],
'age_limit': 0,
},
+ 'skip': 'This Tweet is unavailable',
}, {
# not available in Periscope
'url': 'https://twitter.com/ViviEducation/status/1136534865145286656',
'view_count': int,
},
'add_ie': ['TwitterBroadcast'],
+ 'skip': 'Broadcast no longer exists',
}, {
# unified card
'url': 'https://twitter.com/BrooklynNets/status/1349794411333394432?s=20',
'title': 'md5:d1c4941658e4caaa6cb579260d85dcba',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'md5:71ead15ec44cee55071547d6447c6a3e',
+ 'channel_id': '18552281',
'uploader': 'Brooklyn Nets',
'uploader_id': 'BrooklynNets',
'duration': 324.484,
'like_count': int,
'tags': [],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 1349794411333394432'],
},
'params': {
'skip_download': True,
'id': '1577855447914409984',
'display_id': '1577855540407197696',
'ext': 'mp4',
- 'title': 'md5:9d198efb93557b8f8d5b78c480407214',
+ 'title': 'md5:466a3a8b049b5f5a13164ce915484b51',
'description': 'md5:b9c3699335447391d11753ab21c70a74',
'upload_date': '20221006',
- 'uploader': 'oshtru',
+ 'channel_id': '143077138',
+ 'uploader': 'Oshtru',
'uploader_id': 'oshtru',
'uploader_url': 'https://twitter.com/oshtru',
'thumbnail': r're:^https?://.*\.jpg',
'comment_count': int,
'repost_count': int,
'like_count': int,
- 'view_count': int,
'tags': [],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 1577855540407197696'],
},
'params': {'skip_download': True},
}, {
'url': 'https://twitter.com/UltimaShadowX/status/1577719286659006464',
'info_dict': {
'id': '1577719286659006464',
- 'title': 'Ultima📛 | #вʟм - Test',
+ 'title': 'Ultima Reload - Test',
'description': 'Test https://t.co/Y3KEZD7Dad',
- 'uploader': 'Ultima📛 | #вʟм',
+ 'channel_id': '168922496',
+ 'uploader': 'Ultima Reload',
'uploader_id': 'UltimaShadowX',
'uploader_url': 'https://twitter.com/UltimaShadowX',
'upload_date': '20221005',
'title': 'md5:eec26382babd0f7c18f041db8ae1c9c9',
'thumbnail': r're:^https?://.*\.jpg',
'description': 'md5:95aea692fda36a12081b9629b02daa92',
+ 'channel_id': '1094109584',
'uploader': 'Max Olson',
'uploader_id': 'MesoMax919',
'uploader_url': 'https://twitter.com/MesoMax919',
'comment_count': int,
'repost_count': int,
'like_count': int,
- 'view_count': int,
'tags': ['HurricaneIan'],
'age_limit': 0,
+ '_old_archive_ids': ['twitter 1575560063510810624'],
},
}, {
- # Adult content, fails if not logged in (GraphQL)
+ # Adult content, fails if not logged in
'url': 'https://twitter.com/Rizdraws/status/1575199173472927762',
'info_dict': {
'id': '1575199163847000068',
'ext': 'mp4',
'title': str,
'description': str,
+ 'channel_id': '1217167793541480450',
'uploader': str,
'uploader_id': 'Rizdraws',
'uploader_url': 'https://twitter.com/Rizdraws',
'repost_count': int,
'comment_count': int,
'age_limit': 18,
- 'tags': []
+ 'tags': [],
+ '_old_archive_ids': ['twitter 1575199173472927762'],
},
+ 'params': {'skip_download': 'The media could not be played'},
'skip': 'Requires authentication',
}, {
- # Playlist result only with auth
+ # Playlist result only with graphql API
'url': 'https://twitter.com/Srirachachau/status/1395079556562706435',
'playlist_mincount': 2,
'info_dict': {
'id': '1395079556562706435',
'title': str,
'tags': [],
+ 'channel_id': '21539378',
'uploader': str,
'like_count': int,
'upload_date': '20210519',
'info_dict': {
'id': '1578353380363501568',
'title': str,
+ 'channel_id': '2195866214',
'uploader_id': 'DavidToons_',
'repost_count': int,
'like_count': int,
'id': '1578401165338976258',
'title': str,
'description': 'md5:659a6b517a034b4cee5d795381a2dc41',
+ 'channel_id': '19338359',
'uploader': str,
'uploader_id': 'primevideouk',
'timestamp': 1665155137,
'uploader_id': 'MoniqueCamarra',
'live_status': 'was_live',
'release_timestamp': 1658417414,
- 'description': 'md5:4dc8e972f1d8b3c6580376fabb02a3ad',
+ 'description': 'md5:acce559345fd49f129c20dbcda3f1201',
'timestamp': 1658407771,
'release_date': '20220721',
'upload_date': '20220721',
'description': 'md5:591c19ce66fadc2359725d5cd0d1052c',
'comment_count': int,
'uploader_id': 'CTVJLaidlaw',
+ 'channel_id': '80082014',
'repost_count': int,
'tags': ['colorectalcancer', 'cancerjourney', 'imnotaquitter'],
'upload_date': '20221208',
'title': 'md5:7662a0a27ce6faa3e5b160340f3cfab1',
'thumbnail': r're:^https?://.+\.jpg',
'timestamp': 1670459604.0,
+ 'channel_id': '80082014',
'uploader_id': 'CTVJLaidlaw',
'uploader': 'Jocelyn Laidlaw',
'repost_count': int,
'uploader_url': 'https://twitter.com/CTVJLaidlaw',
'display_id': '1600649710662213632',
'like_count': int,
- 'view_count': int,
'description': 'md5:591c19ce66fadc2359725d5cd0d1052c',
'upload_date': '20221208',
'age_limit': 0,
+ '_old_archive_ids': ['twitter 1600649710662213632'],
},
'params': {'noplaylist': True},
}, {
'title': '뽀 - 아 최우제 이동속도 봐',
'description': '아 최우제 이동속도 봐 https://t.co/dxu2U5vXXB',
'duration': 24.598,
+ 'channel_id': '1281839411068432384',
'uploader': '뽀',
'uploader_id': 's2FAKER',
'uploader_url': 'https://twitter.com/s2FAKER',
'like_count': int,
'repost_count': int,
'comment_count': int,
- 'view_count': int,
+ '_old_archive_ids': ['twitter 1621117700482416640'],
},
+ 'skip': 'Requires authentication',
}, {
'url': 'https://twitter.com/hlo_again/status/1599108751385972737/video/2',
'info_dict': {
'display_id': '1599108751385972737',
'ext': 'mp4',
'title': '\u06ea - \U0001F48B',
+ 'channel_id': '1347791436809441283',
'uploader_url': 'https://twitter.com/hlo_again',
'like_count': int,
'uploader_id': 'hlo_again',
'repost_count': int,
'duration': 9.531,
'comment_count': int,
- 'view_count': int,
'upload_date': '20221203',
'age_limit': 0,
'timestamp': 1670092210.0,
'tags': [],
'uploader': '\u06ea',
'description': '\U0001F48B https://t.co/bTj9Qz7vQP',
+ '_old_archive_ids': ['twitter 1599108751385972737'],
},
'params': {'noplaylist': True},
}, {
'id': '1600009362759733248',
'display_id': '1600009574919962625',
'ext': 'mp4',
+ 'channel_id': '211814412',
'uploader_url': 'https://twitter.com/MunTheShinobi',
'description': 'This is a genius ad by Apple. \U0001f525\U0001f525\U0001f525\U0001f525\U0001f525 https://t.co/cNsA0MoOml',
- 'view_count': int,
'thumbnail': 'https://pbs.twimg.com/ext_tw_video_thumb/1600009362759733248/pu/img/XVhFQivj75H_YxxV.jpg?name=orig',
'age_limit': 0,
- 'uploader': 'Mün The Shinobi',
+ 'uploader': 'Mün',
'repost_count': int,
'upload_date': '20221206',
- 'title': 'Mün The Shinobi - This is a genius ad by Apple. \U0001f525\U0001f525\U0001f525\U0001f525\U0001f525',
+ 'title': 'Mün - This is a genius ad by Apple. \U0001f525\U0001f525\U0001f525\U0001f525\U0001f525',
'comment_count': int,
'like_count': int,
'tags': [],
'uploader_id': 'MunTheShinobi',
'duration': 139.987,
'timestamp': 1670306984.0,
+ '_old_archive_ids': ['twitter 1600009574919962625'],
},
}, {
- # url to retweet id w/ legacy api
+ # retweeted_status (private)
'url': 'https://twitter.com/liberdalau/status/1623739803874349067',
'info_dict': {
'id': '1623274794488659969',
'like_count': int,
'repost_count': int,
},
- 'params': {'extractor_args': {'twitter': {'legacy_api': ['']}}},
+ 'skip': 'Protected tweet',
}, {
- # orig tweet w/ graphql
- 'url': 'https://twitter.com/liberdalau/status/1623739803874349067',
+ # retweeted_status
+ 'url': 'https://twitter.com/playstrumpcard/status/1695424220702888009',
'info_dict': {
- 'id': '1623274794488659969',
- 'display_id': '1623739803874349067',
+ 'id': '1694928337846538240',
'ext': 'mp4',
- 'title': '@selfisekai@hackerspace.pl 🐀 - RT @Johnnybull3ts: Me after going viral to over 30million people: Whoopsie-daisy',
- 'description': 'md5:9258bdbb54793bdc124fe1cd47e96c6a',
- 'uploader': '@selfisekai@hackerspace.pl 🐀',
- 'uploader_id': 'liberdalau',
- 'uploader_url': 'https://twitter.com/liberdalau',
+ 'display_id': '1695424220702888009',
+ 'title': 'md5:e8daa9527bc2b947121395494f786d9d',
+ 'description': 'md5:004f2d37fd58737724ec75bc7e679938',
+ 'channel_id': '15212187',
+ 'uploader': 'Benny Johnson',
+ 'uploader_id': 'bennyjohnson',
+ 'uploader_url': 'https://twitter.com/bennyjohnson',
'age_limit': 0,
'tags': [],
- 'duration': 8.033,
- 'timestamp': 1675964711.0,
- 'upload_date': '20230209',
- 'thumbnail': r're:https://pbs\.twimg\.com/ext_tw_video_thumb/.+',
+ 'duration': 45.001,
+ 'timestamp': 1692962814.0,
+ 'upload_date': '20230825',
+ 'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
+ 'like_count': int,
+ 'repost_count': int,
+ 'comment_count': int,
+ '_old_archive_ids': ['twitter 1695424220702888009'],
+ },
+ }, {
+ # retweeted_status w/ legacy API
+ 'url': 'https://twitter.com/playstrumpcard/status/1695424220702888009',
+ 'info_dict': {
+ 'id': '1694928337846538240',
+ 'ext': 'mp4',
+ 'display_id': '1695424220702888009',
+ 'title': 'md5:e8daa9527bc2b947121395494f786d9d',
+ 'description': 'md5:004f2d37fd58737724ec75bc7e679938',
+ 'channel_id': '15212187',
+ 'uploader': 'Benny Johnson',
+ 'uploader_id': 'bennyjohnson',
+ 'uploader_url': 'https://twitter.com/bennyjohnson',
+ 'age_limit': 0,
+ 'tags': [],
+ 'duration': 45.001,
+ 'timestamp': 1692962814.0,
+ 'upload_date': '20230825',
+ 'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
'like_count': int,
+ 'repost_count': int,
+ '_old_archive_ids': ['twitter 1695424220702888009'],
+ },
+ 'params': {'extractor_args': {'twitter': {'api': ['legacy']}}},
+ }, {
+ # Broadcast embedded in tweet
+ 'url': 'https://twitter.com/JessicaDobsonWX/status/1731121063248175384',
+ 'info_dict': {
+ 'id': '1rmxPMjLzAXKN',
+ 'ext': 'mp4',
+ 'title': 'WAVE Weather Now - Saturday 12/2/23 Update',
+ 'uploader': 'Jessica Dobson',
+ 'uploader_id': 'JessicaDobsonWX',
+ 'uploader_url': 'https://twitter.com/JessicaDobsonWX',
+ 'timestamp': 1701566398,
+ 'upload_date': '20231203',
+ 'live_status': 'was_live',
+ 'thumbnail': r're:https://[^/]+pscp\.tv/.+\.jpg',
+ 'concurrent_view_count': int,
'view_count': int,
+ },
+ 'add_ie': ['TwitterBroadcast'],
+ }, {
+ # Animated gif and quote tweet video
+ 'url': 'https://twitter.com/BAKKOOONN/status/1696256659889565950',
+ 'playlist_mincount': 2,
+ 'info_dict': {
+ 'id': '1696256659889565950',
+ 'title': 'BAKOON - https://t.co/zom968d0a0',
+ 'description': 'https://t.co/zom968d0a0',
+ 'tags': [],
+ 'channel_id': '1263540390',
+ 'uploader': 'BAKOON',
+ 'uploader_id': 'BAKKOOONN',
+ 'uploader_url': 'https://twitter.com/BAKKOOONN',
+ 'age_limit': 18,
+ 'timestamp': 1693254077.0,
+ 'upload_date': '20230828',
+ 'like_count': int,
+ 'comment_count': int,
+ 'repost_count': int,
+ },
+ 'skip': 'Requires authentication',
+ }, {
+ # "stale tweet" with typename "TweetWithVisibilityResults"
+ 'url': 'https://twitter.com/RobertKennedyJr/status/1724884212803834154',
+ 'md5': '511377ff8dfa7545307084dca4dce319',
+ 'info_dict': {
+ 'id': '1724883339285544960',
+ 'ext': 'mp4',
+ 'title': 'md5:cc56716f9ed0b368de2ba54c478e493c',
+ 'description': 'md5:9dc14f5b0f1311fc7caf591ae253a164',
+ 'display_id': '1724884212803834154',
+ 'channel_id': '337808606',
+ 'uploader': 'Robert F. Kennedy Jr',
+ 'uploader_id': 'RobertKennedyJr',
+ 'uploader_url': 'https://twitter.com/RobertKennedyJr',
+ 'upload_date': '20231115',
+ 'timestamp': 1700079417.0,
+ 'duration': 341.048,
+ 'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
+ 'tags': ['Kennedy24'],
'repost_count': int,
+ 'like_count': int,
+ 'comment_count': int,
+ 'age_limit': 0,
+ '_old_archive_ids': ['twitter 1724884212803834154'],
+ },
+ }, {
+ # x.com
+ 'url': 'https://x.com/historyinmemes/status/1790637656616943991',
+ 'md5': 'daca3952ba0defe2cfafb1276d4c1ea5',
+ 'info_dict': {
+ 'id': '1790637589910654976',
+ 'ext': 'mp4',
+ 'title': 'Historic Vids - One of the most intense moments in history',
+ 'description': 'One of the most intense moments in history https://t.co/Zgzhvix8ES',
+ 'display_id': '1790637656616943991',
+ 'uploader': 'Historic Vids',
+ 'uploader_id': 'historyinmemes',
+ 'uploader_url': 'https://twitter.com/historyinmemes',
+ 'channel_id': '855481986290524160',
+ 'upload_date': '20240515',
+ 'timestamp': 1715756260.0,
+ 'duration': 15.488,
+ 'tags': [],
'comment_count': int,
+ 'repost_count': int,
+ 'like_count': int,
+ 'thumbnail': r're:https://pbs\.twimg\.com/amplify_video_thumb/.+',
+ 'age_limit': 0,
+ '_old_archive_ids': ['twitter 1790637656616943991'],
},
}, {
# onion route
'only_matching': True,
}]
+ _MEDIA_ID_RE = re.compile(r'_video/(\d+)/')
+
+ @property
+ def _GRAPHQL_ENDPOINT(self):
+ if self.is_logged_in:
+ return 'zZXycP0V6H7m-2r0mOnFcA/TweetDetail'
+ return '2ICDjqPd81tulZcYrtpTuQ/TweetResultByRestId'
+
def _graphql_to_legacy(self, data, twid):
result = traverse_obj(data, (
'threaded_conversation_with_injections_v2', 'instructions', 0, 'entries',
), default={}, get_all=False) if self.is_logged_in else traverse_obj(
data, ('tweetResult', 'result', {dict}), default={})
- if result.get('__typename') not in ('Tweet', 'TweetTombstone', 'TweetUnavailable', None):
- self.report_warning(f'Unknown typename: {result.get("__typename")}', twid, only_once=True)
+ typename = result.get('__typename')
+ if typename not in ('Tweet', 'TweetWithVisibilityResults', 'TweetTombstone', 'TweetUnavailable', None):
+ self.report_warning(f'Unknown typename: {typename}', twid, only_once=True)
if 'tombstone' in result:
cause = remove_end(traverse_obj(result, ('tombstone', 'text', 'text', {str})), '. Learn more')
raise ExtractorError(f'Twitter API says: {cause or "Unknown error"}', expected=True)
- elif result.get('__typename') == 'TweetUnavailable':
+ elif typename == 'TweetUnavailable':
reason = result.get('reason')
if reason == 'NsfwLoggedOut':
self.raise_login_required('NSFW tweet requires authentication')
+ elif reason == 'Protected':
+ self.raise_login_required('You are not authorized to view this protected tweet')
raise ExtractorError(reason or 'Requested tweet is unavailable', expected=True)
+ # Result for "stale tweet" needs additional transformation
+ elif typename == 'TweetWithVisibilityResults':
+ result = traverse_obj(result, ('tweet', {dict})) or {}
status = result.get('legacy', {})
status.update(traverse_obj(result, {
'user': ('core', 'user_results', 'result', 'legacy'),
'card': ('card', 'legacy'),
'quoted_status': ('quoted_status_result', 'result', 'legacy'),
+ 'retweeted_status': ('legacy', 'retweeted_status_result', 'result', 'legacy'),
}, expected_type=dict, default={}))
- # extra transformation is needed since result does not match legacy format
+ # extra transformations needed since result does not match legacy format
+ if status.get('retweeted_status'):
+ status['retweeted_status']['user'] = traverse_obj(status, (
+ 'retweeted_status_result', 'result', 'core', 'user_results', 'result', 'legacy', {dict})) or {}
+
binding_values = {
binding_value.get('key'): binding_value.get('value')
for binding_value in traverse_obj(status, ('card', 'binding_values', ..., {dict}))
'responsive_web_media_download_video_enabled': False,
'responsive_web_graphql_skip_user_profile_image_extensions_enabled': False,
'responsive_web_graphql_timeline_navigation_enabled': True,
- 'responsive_web_enhance_cards_enabled': False
+ 'responsive_web_enhance_cards_enabled': False,
},
'fieldToggles': {
- 'withArticleRichContentState': False
- }
+ 'withArticleRichContentState': False,
+ },
}
+ def _call_syndication_api(self, twid):
+ self.report_warning(
+ 'Not all metadata or media is available via syndication endpoint', twid, only_once=True)
+ status = self._download_json(
+ 'https://cdn.syndication.twimg.com/tweet-result', twid, 'Downloading syndication JSON',
+ headers={'User-Agent': 'Googlebot'}, query={
+ 'id': twid,
+ # TODO: token = ((Number(twid) / 1e15) * Math.PI).toString(36).replace(/(0+|\.)/g, '')
+ 'token': ''.join(random.choices('123456789abcdefghijklmnopqrstuvwxyz', k=10)),
+ })
+ if not status:
+ raise ExtractorError('Syndication endpoint returned empty JSON response')
+ # Transform the result so its structure matches that of legacy/graphql
+ media = []
+ for detail in traverse_obj(status, ((None, 'quoted_tweet'), 'mediaDetails', ..., {dict})):
+ detail['id_str'] = traverse_obj(detail, (
+ 'video_info', 'variants', ..., 'url', {self._MEDIA_ID_RE.search}, 1), get_all=False) or twid
+ media.append(detail)
+ status['extended_entities'] = {'media': media}
+
+ return status
+
+ def _extract_status(self, twid):
+ if self._selected_api not in ('graphql', 'legacy', 'syndication'):
+ raise ExtractorError(f'{self._selected_api!r} is not a valid API selection', expected=True)
+
+ try:
+ if self.is_logged_in or self._selected_api == 'graphql':
+ status = self._graphql_to_legacy(self._call_graphql_api(self._GRAPHQL_ENDPOINT, twid), twid)
+ elif self._selected_api == 'legacy':
+ status = self._call_api(f'statuses/show/{twid}.json', twid, {
+ 'cards_platform': 'Web-12',
+ 'include_cards': 1,
+ 'include_reply_count': 1,
+ 'include_user_entities': 0,
+ 'tweet_mode': 'extended',
+ })
+ except ExtractorError as e:
+ if not isinstance(e.cause, HTTPError) or not e.cause.status == 429:
+ raise
+ self.report_warning('Rate-limit exceeded; falling back to syndication endpoint')
+ status = self._call_syndication_api(twid)
+
+ if self._selected_api == 'syndication':
+ status = self._call_syndication_api(twid)
+
+ return traverse_obj(status, 'retweeted_status', None, expected_type=dict) or {}
+
def _real_extract(self, url):
twid, selected_index = self._match_valid_url(url).group('id', 'index')
- if not self.is_logged_in and self._configuration_arg('legacy_api'):
- status = traverse_obj(self._call_api(f'statuses/show/{twid}.json', twid, {
- 'cards_platform': 'Web-12',
- 'include_cards': 1,
- 'include_reply_count': 1,
- 'include_user_entities': 0,
- 'tweet_mode': 'extended',
- }), 'retweeted_status', None)
- elif not self.is_logged_in:
- status = self._graphql_to_legacy(
- self._call_graphql_api('2ICDjqPd81tulZcYrtpTuQ/TweetResultByRestId', twid), twid)
- else:
- status = self._graphql_to_legacy(
- self._call_graphql_api('zZXycP0V6H7m-2r0mOnFcA/TweetDetail', twid), twid)
+ status = self._extract_status(twid)
title = description = traverse_obj(
status, (('full_text', 'text'), {lambda x: x.replace('\n', ' ')}), get_all=False) or ''
'description': description,
'uploader': uploader,
'timestamp': unified_timestamp(status.get('created_at')),
+ 'channel_id': str_or_none(status.get('user_id_str')) or str_or_none(user.get('id_str')),
'uploader_id': uploader_id,
'uploader_url': format_field(uploader_id, None, 'https://twitter.com/%s'),
'like_count': int_or_none(status.get('favorite_count')),
'formats': formats,
'subtitles': subtitles,
'thumbnails': thumbnails,
- 'view_count': traverse_obj(media, ('mediaStats', 'viewCount', {int_or_none})),
+ 'view_count': traverse_obj(media, ('mediaStats', 'viewCount', {int_or_none})), # No longer available
'duration': float_or_none(traverse_obj(media, ('video_info', 'duration_millis')), 1000),
- # The codec of http formats are unknown
- '_format_sort_fields': ('res', 'br', 'size', 'proto'),
+ # Prioritize m3u8 formats for compat, see https://github.com/yt-dlp/yt-dlp/issues/8117
+ '_format_sort_fields': ('res', 'proto:m3u8', 'br', 'size'), # http format codec is unknown
}
def extract_from_card_info(card):
def _find_dimension(target):
w = int_or_none(self._html_search_meta(
- 'twitter:%s:width' % target, webpage, fatal=False))
+ f'twitter:{target}:width', webpage, fatal=False))
h = int_or_none(self._html_search_meta(
- 'twitter:%s:height' % target, webpage, fatal=False))
+ f'twitter:{target}:height', webpage, fatal=False))
return w, h
if thumbnail:
IE_NAME = 'twitter:broadcast'
_VALID_URL = TwitterBaseIE._BASE_REGEX + r'i/broadcasts/(?P<id>[0-9a-zA-Z]{13})'
- _TEST = {
+ _TESTS = [{
# untitled Periscope video
'url': 'https://twitter.com/i/broadcasts/1yNGaQLWpejGj',
'info_dict': {
'ext': 'mp4',
'title': 'Andrea May Sahouri - Periscope Broadcast',
'uploader': 'Andrea May Sahouri',
- 'uploader_id': '1PXEdBZWpGwKe',
+ 'uploader_id': 'andreamsahouri',
+ 'uploader_url': 'https://twitter.com/andreamsahouri',
+ 'timestamp': 1590973638,
+ 'upload_date': '20200601',
'thumbnail': r're:^https?://[^?#]+\.jpg\?token=',
'view_count': int,
},
- }
+ }, {
+ 'url': 'https://twitter.com/i/broadcasts/1ZkKzeyrPbaxv',
+ 'info_dict': {
+ 'id': '1ZkKzeyrPbaxv',
+ 'ext': 'mp4',
+ 'title': 'Starship | SN10 | High-Altitude Flight Test',
+ 'uploader': 'SpaceX',
+ 'uploader_id': 'SpaceX',
+ 'uploader_url': 'https://twitter.com/SpaceX',
+ 'timestamp': 1614812942,
+ 'upload_date': '20210303',
+ 'thumbnail': r're:^https?://[^?#]+\.jpg\?token=',
+ 'view_count': int,
+ },
+ }, {
+ 'url': 'https://twitter.com/i/broadcasts/1OyKAVQrgzwGb',
+ 'info_dict': {
+ 'id': '1OyKAVQrgzwGb',
+ 'ext': 'mp4',
+ 'title': 'Starship Flight Test',
+ 'uploader': 'SpaceX',
+ 'uploader_id': 'SpaceX',
+ 'uploader_url': 'https://twitter.com/SpaceX',
+ 'timestamp': 1681993964,
+ 'upload_date': '20230420',
+ 'thumbnail': r're:^https?://[^?#]+\.jpg\?token=',
+ 'view_count': int,
+ },
+ }]
def _real_extract(self, url):
broadcast_id = self._match_id(url)
broadcast = self._call_api(
'broadcasts/show.json', broadcast_id,
{'ids': broadcast_id})['broadcasts'][broadcast_id]
+ if not broadcast:
+ raise ExtractorError('Broadcast no longer exists', expected=True)
info = self._parse_broadcast_data(broadcast, broadcast_id)
+ info['title'] = broadcast.get('status') or info.get('title')
+ info['uploader_id'] = broadcast.get('twitter_username') or info.get('uploader_id')
+ info['uploader_url'] = format_field(broadcast, 'twitter_username', 'https://twitter.com/%s', default=None)
+ if info['live_status'] == 'is_upcoming':
+ return info
+
media_key = broadcast['media_key']
source = self._call_api(
f'live_video_stream/status/{media_key}', media_key)['source']
m3u8_url = source.get('noRedirectPlaybackUrl') or source['location']
if '/live_video_stream/geoblocked/' in m3u8_url:
self.raise_geo_restricted()
- m3u8_id = compat_parse_qs(compat_urllib_parse_urlparse(
+ m3u8_id = urllib.parse.parse_qs(urllib.parse.urlparse(
m3u8_url).query).get('type', [None])[0]
state, width, height = self._extract_common_format_info(broadcast)
info['formats'] = self._extract_pscp_m3u8_formats(
'release_date': '20220807',
},
'params': {'skip_download': 'm3u8'},
+ }, {
+ # post_live/TimedOut but downloadable
+ 'url': 'https://twitter.com/i/spaces/1vAxRAVQWONJl',
+ 'info_dict': {
+ 'id': '1vAxRAVQWONJl',
+ 'ext': 'm4a',
+ 'title': 'Framing Up FinOps: Billing Tools',
+ 'description': 'Twitter Space participated by rupa, Alfonso Hernandez',
+ 'uploader': 'Google Cloud',
+ 'uploader_id': 'googlecloud',
+ 'live_status': 'post_live',
+ 'timestamp': 1681409554,
+ 'upload_date': '20230413',
+ 'release_timestamp': 1681839000,
+ 'release_date': '20230418',
+ },
+ 'params': {'skip_download': 'm3u8'},
+ }, {
+ # Needs ffmpeg as downloader, see: https://github.com/yt-dlp/yt-dlp/issues/7536
+ 'url': 'https://twitter.com/i/spaces/1eaKbrQbjoRKX',
+ 'info_dict': {
+ 'id': '1eaKbrQbjoRKX',
+ 'ext': 'm4a',
+ 'title': 'あ',
+ 'description': 'Twitter Space participated by nobody yet',
+ 'uploader': '息根とめる🔪Twitchで復活',
+ 'uploader_id': 'tomeru_ikinone',
+ 'live_status': 'was_live',
+ 'timestamp': 1685617198,
+ 'upload_date': '20230601',
+ },
+ 'params': {'skip_download': 'm3u8'},
}]
SPACE_STATUS = {
is_live = live_status == 'is_live'
formats = []
+ headers = {'Referer': 'https://twitter.com/'}
if live_status == 'is_upcoming':
self.raise_no_formats('Twitter Space not started yet', expected=True)
elif not is_live and not metadata.get('is_space_available_for_replay'):
source = traverse_obj(
self._call_api(f'live_video_stream/status/{metadata["media_key"]}', metadata['media_key']),
('source', ('noRedirectPlaybackUrl', 'location'), {url_or_none}), get_all=False)
- formats = self._extract_m3u8_formats(
- source, metadata['media_key'], 'm4a', live=is_live, fatal=False,
- headers={'Referer': 'https://twitter.com/'}) if source else []
+ formats = self._extract_m3u8_formats( # XXX: Some Spaces need ffmpeg as downloader
+ source, metadata['media_key'], 'm4a', entry_protocol='m3u8', live=is_live,
+ headers=headers, fatal=False) if source else []
for fmt in formats:
fmt.update({'vcodec': 'none', 'acodec': 'aac'})
if not is_live:
lambda: int_or_none(metadata['scheduled_start'], scale=1000)),
'timestamp': int_or_none(metadata.get('created_at'), scale=1000),
'formats': formats,
+ 'http_headers': headers,
}
class TwitterShortenerIE(TwitterBaseIE):
IE_NAME = 'twitter:shortener'
- _VALID_URL = r'https?://t.co/(?P<id>[^?]+)|tco:(?P<eid>[^?]+)'
+ _VALID_URL = r'https?://t\.co/(?P<id>[^?#]+)|tco:(?P<eid>[^?#]+)'
_BASE_URL = 'https://t.co/'
def _real_extract(self, url):
mobj = self._match_valid_url(url)
- eid, id = mobj.group('eid', 'id')
+ eid, shortcode = mobj.group('eid', 'id')
if eid:
- id = eid
- url = self._BASE_URL + id
- new_url = self._request_webpage(url, id, headers={'User-Agent': 'curl'}).url
- __UNSAFE_LINK = "https://twitter.com/safety/unsafe_link_warning?unsafe_link="
+ shortcode = eid
+ url = self._BASE_URL + shortcode
+ new_url = self._request_webpage(url, shortcode, headers={'User-Agent': 'curl'}).url
+ __UNSAFE_LINK = 'https://twitter.com/safety/unsafe_link_warning?unsafe_link='
if new_url.startswith(__UNSAFE_LINK):
- new_url = new_url.replace(__UNSAFE_LINK, "")
+ new_url = new_url.replace(__UNSAFE_LINK, '')
return self.url_result(new_url)
projects / yt-dlp.git / blobdiff