2 from __future__
import unicode_literals
10 from .common
import InfoExtractor
11 from ..compat
import compat_urllib_parse_unquote
24 class TikTokBaseIE(InfoExtractor
):
25 _APP_VERSION
= '20.9.3'
26 _MANIFEST_APP_VERSION
= '291'
29 _API_HOSTNAME
= 'api-h2.tiktokv.com'
30 _UPLOADER_URL_FORMAT
= 'https://www.tiktok.com/@%s'
31 _WEBPAGE_HOST
= 'https://www.tiktok.com/'
32 QUALITIES
= ('360p', '540p', '720p')
34 def _call_api(self
, ep
, query
, video_id
, fatal
=True,
35 note
='Downloading API JSON', errnote
='Unable to download API page'):
38 'version_name': self
._APP
_VERSION
,
39 'version_code': self
._MANIFEST
_APP
_VERSION
,
40 'build_number': self
._APP
_VERSION
,
41 'manifest_version_code': self
._MANIFEST
_APP
_VERSION
,
42 'update_version_code': self
._MANIFEST
_APP
_VERSION
,
43 'openudid': ''.join(random
.choice('0123456789abcdef') for _
in range(16)),
44 'uuid': ''.join([random
.choice(string
.digits
) for _
in range(16)]),
45 '_rticket': int(time
.time() * 1000),
46 'ts': int(time
.time()),
47 'device_brand': 'Google',
48 'device_type': 'Pixel 4',
49 'device_platform': 'android',
50 'resolution': '1080*1920',
54 'carrier_region': 'US',
57 'app_name': self
._APP
_NAME
,
60 'timezone_name': 'America/New_York',
61 'timezone_offset': '-14400',
62 'channel': 'googleplay',
69 'cp': 'cbfhckdckkde1',
71 self
._set
_cookie
(self
._API
_HOSTNAME
, 'odin_tt', ''.join(random
.choice('0123456789abcdef') for _
in range(160)))
72 webpage_cookies
= self
._get
_cookies
(self
._WEBPAGE
_HOST
)
73 if webpage_cookies
.get('sid_tt'):
74 self
._set
_cookie
(self
._API
_HOSTNAME
, 'sid_tt', webpage_cookies
['sid_tt'].value
)
75 return self
._download
_json
(
76 'https://%s/aweme/v1/%s/' % (self
._API
_HOSTNAME
, ep
), video_id
=video_id
,
77 fatal
=fatal
, note
=note
, errnote
=errnote
, headers
={
78 'User-Agent': f
'com.ss.android.ugc.trill/{self._MANIFEST_APP_VERSION} (Linux; U; Android 10; en_US; Pixel 4; Build/QQ3A.200805.001; Cronet/58.0.2991.0)',
79 'Accept': 'application/json',
82 def _parse_aweme_video_app(self
, aweme_detail
):
83 aweme_id
= aweme_detail
['aweme_id']
84 video_info
= aweme_detail
['video']
86 def parse_url_key(url_key
):
87 format_id
, codec
, res
, bitrate
= self
._search
_regex
(
88 r
'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key
,
89 'url key', default
=(None, None, None, None), group
=('id', 'codec', 'res', 'bitrate'))
93 'format_id': format_id
,
94 'vcodec': 'h265' if codec
== 'bytevc1' else codec
,
95 'tbr': int_or_none(bitrate
, scale
=1000) or None,
96 'quality': qualities(self
.QUALITIES
)(res
),
99 known_resolutions
= {}
101 def extract_addr(addr
, add_meta
={}):
102 parsed_meta
, res
= parse_url_key(addr
.get('url_key', ''))
104 known_resolutions
.setdefault(res
, {}).setdefault('height', add_meta
.get('height'))
105 known_resolutions
[res
].setdefault('width', add_meta
.get('width'))
106 parsed_meta
.update(known_resolutions
.get(res
, {}))
107 add_meta
.setdefault('height', int_or_none(res
[:-1]))
110 'filesize': int_or_none(addr
.get('data_size')),
113 'source_preference': -2 if 'aweme/v1' in url
else -1, # Downloads from API might get blocked
114 **add_meta
, **parsed_meta
,
115 'format_note': join_nonempty(
116 add_meta
.get('format_note'), '(API)' if 'aweme/v1' in url
else None, delim
=' ')
117 } for url
in addr
.get('url_list') or []]
119 # Hack: Add direct video links first to prioritize them when removing duplicate formats
121 if video_info
.get('play_addr'):
122 formats
.extend(extract_addr(video_info
['play_addr'], {
123 'format_id': 'play_addr',
124 'format_note': 'Direct video',
125 'vcodec': 'h265' if traverse_obj(
126 video_info
, 'is_bytevc1', 'is_h265') else 'h264', # Always h264?
127 'width': video_info
.get('width'),
128 'height': video_info
.get('height'),
130 if video_info
.get('download_addr'):
131 formats
.extend(extract_addr(video_info
['download_addr'], {
132 'format_id': 'download_addr',
133 'format_note': 'Download video%s' % (', watermarked' if video_info
.get('has_watermark') else ''),
135 'width': video_info
.get('width'),
136 'height': video_info
.get('height'),
137 'preference': -2 if video_info
.get('has_watermark') else -1,
139 if video_info
.get('play_addr_h264'):
140 formats
.extend(extract_addr(video_info
['play_addr_h264'], {
141 'format_id': 'play_addr_h264',
142 'format_note': 'Direct video',
145 if video_info
.get('play_addr_bytevc1'):
146 formats
.extend(extract_addr(video_info
['play_addr_bytevc1'], {
147 'format_id': 'play_addr_bytevc1',
148 'format_note': 'Direct video',
152 for bitrate
in video_info
.get('bit_rate', []):
153 if bitrate
.get('play_addr'):
154 formats
.extend(extract_addr(bitrate
['play_addr'], {
155 'format_id': bitrate
.get('gear_name'),
156 'format_note': 'Playback video',
157 'tbr': try_get(bitrate
, lambda x
: x
['bit_rate'] / 1000),
158 'vcodec': 'h265' if traverse_obj(
159 bitrate
, 'is_bytevc1', 'is_h265') else 'h264',
160 'fps': bitrate
.get('FPS'),
163 self
._remove
_duplicate
_formats
(formats
)
164 self
._sort
_formats
(formats
, ('quality', 'codec', 'size', 'br'))
167 for cover_id
in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
168 'origin_cover', 'dynamic_cover'):
169 cover
= video_info
.get(cover_id
)
171 for cover_url
in cover
['url_list']:
177 stats_info
= aweme_detail
.get('statistics', {})
178 author_info
= aweme_detail
.get('author', {})
179 music_info
= aweme_detail
.get('music', {})
180 user_url
= self
._UPLOADER
_URL
_FORMAT
% (traverse_obj(author_info
,
181 'sec_uid', 'id', 'uid', 'unique_id',
182 expected_type
=str_or_none
, get_all
=False))
183 labels
= traverse_obj(aweme_detail
, ('hybrid_label', ..., 'text'), expected_type
=str, default
=[])
185 contained_music_track
= traverse_obj(
186 music_info
, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type
=str)
187 contained_music_author
= traverse_obj(
188 music_info
, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type
=str)
190 is_generic_og_trackname
= music_info
.get('is_original_sound') and music_info
.get('title') == 'original sound - %s' % music_info
.get('owner_handle')
191 if is_generic_og_trackname
:
192 music_track
, music_author
= contained_music_track
or 'original sound', contained_music_author
194 music_track
, music_author
= music_info
.get('title'), music_info
.get('author')
198 'title': aweme_detail
['desc'],
199 'description': aweme_detail
['desc'],
200 'view_count': int_or_none(stats_info
.get('play_count')),
201 'like_count': int_or_none(stats_info
.get('digg_count')),
202 'repost_count': int_or_none(stats_info
.get('share_count')),
203 'comment_count': int_or_none(stats_info
.get('comment_count')),
204 'uploader': str_or_none(author_info
.get('unique_id')),
205 'creator': str_or_none(author_info
.get('nickname')),
206 'uploader_id': str_or_none(author_info
.get('uid')),
207 'uploader_url': user_url
,
208 'track': music_track
,
209 'album': str_or_none(music_info
.get('album')) or None,
210 'artist': music_author
,
211 'timestamp': int_or_none(aweme_detail
.get('create_time')),
213 'thumbnails': thumbnails
,
214 'duration': int_or_none(traverse_obj(video_info
, 'duration', ('download_addr', 'duration')), scale
=1000),
215 'availability': self
._availability
(
216 is_private
='Private' in labels
,
217 needs_subscription
='Friends only' in labels
,
218 is_unlisted
='Followers only' in labels
)
221 def _parse_aweme_video_web(self
, aweme_detail
, webpage_url
):
222 video_info
= aweme_detail
['video']
223 author_info
= traverse_obj(aweme_detail
, 'author', 'authorInfo', default
={})
224 music_info
= aweme_detail
.get('music') or {}
225 stats_info
= aweme_detail
.get('stats') or {}
226 user_url
= self
._UPLOADER
_URL
_FORMAT
% (traverse_obj(author_info
,
227 'secUid', 'id', 'uid', 'uniqueId',
228 expected_type
=str_or_none
, get_all
=False))
231 play_url
= video_info
.get('playAddr')
232 width
= video_info
.get('width')
233 height
= video_info
.get('height')
234 if isinstance(play_url
, str):
236 'url': self
._proto
_relative
_url
(play_url
),
241 elif isinstance(play_url
, list):
243 'url': self
._proto
_relative
_url
(url
),
247 } for url
in traverse_obj(play_url
, (..., 'src'), expected_type
=url_or_none
, default
=[]) if url
]
249 download_url
= url_or_none(video_info
.get('downloadAddr')) or traverse_obj(video_info
, ('download', 'url'), expected_type
=url_or_none
)
252 'format_id': 'download',
253 'url': self
._proto
_relative
_url
(download_url
),
258 self
._remove
_duplicate
_formats
(formats
)
259 self
._sort
_formats
(formats
)
262 for thumbnail_name
in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
263 if aweme_detail
.get(thumbnail_name
):
265 'url': self
._proto
_relative
_url
(aweme_detail
[thumbnail_name
]),
271 'id': traverse_obj(aweme_detail
, 'id', 'awemeId', expected_type
=str_or_none
),
272 'title': aweme_detail
.get('desc'),
273 'duration': try_get(aweme_detail
, lambda x
: x
['video']['duration'], int),
274 'view_count': int_or_none(stats_info
.get('playCount')),
275 'like_count': int_or_none(stats_info
.get('diggCount')),
276 'repost_count': int_or_none(stats_info
.get('shareCount')),
277 'comment_count': int_or_none(stats_info
.get('commentCount')),
278 'timestamp': int_or_none(aweme_detail
.get('createTime')),
279 'creator': str_or_none(author_info
.get('nickname')),
280 'uploader': str_or_none(author_info
.get('uniqueId')),
281 'uploader_id': str_or_none(author_info
.get('id')),
282 'uploader_url': user_url
,
283 'track': str_or_none(music_info
.get('title')),
284 'album': str_or_none(music_info
.get('album')) or None,
285 'artist': str_or_none(music_info
.get('authorName')),
287 'thumbnails': thumbnails
,
288 'description': str_or_none(aweme_detail
.get('desc')),
290 'Referer': webpage_url
295 class TikTokIE(TikTokBaseIE
):
296 _VALID_URL
= r
'https?://www\.tiktok\.com/@[\w\.-]+/video/(?P<id>\d+)'
299 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
300 'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
302 'id': '6748451240264420610',
304 'title': '#jassmanak #lehanga #leenabhushan',
305 'description': '#jassmanak #lehanga #leenabhushan',
309 'uploader': 'leenabhushan',
310 'uploader_id': '6691488002098119685',
311 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
312 'creator': 'facestoriesbyleenabh',
313 'thumbnail': r
're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
314 'upload_date': '20191016',
315 'timestamp': 1571246252,
319 'comment_count': int,
322 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
323 'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
325 'id': '6742501081818877190',
327 'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
328 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
332 'uploader': 'patrox',
333 'uploader_id': '18702747',
334 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
336 'thumbnail': r
're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
337 'upload_date': '20190930',
338 'timestamp': 1569860870,
342 'comment_count': int,
345 # Promoted content/ad
346 'url': 'https://www.tiktok.com/@MS4wLjABAAAAAR29F6J2Ktu0Daw03BJyXPNoRQ-W7U5a0Mn3lVCq2rQhjOd_WNLclHUoFgwX8Eno/video/6932675057474981122',
347 'only_matching': True,
350 def _extract_aweme_app(self
, aweme_id
):
351 aweme_detail
= self
._call
_api
('aweme/detail', {'aweme_id': aweme_id}
, aweme_id
,
352 note
='Downloading video details', errnote
='Unable to download video details').get('aweme_detail')
354 raise ExtractorError('Video not available', video_id
=aweme_id
)
355 return self
._parse
_aweme
_video
_app
(aweme_detail
)
357 def _real_extract(self
, url
):
358 video_id
= self
._match
_id
(url
)
361 return self
._extract
_aweme
_app
(video_id
)
362 except ExtractorError
as e
:
363 self
.report_warning(f
'{e}; Retrying with webpage')
365 # If we only call once, we get a 403 when downlaoding the video.
366 self
._download
_webpage
(url
, video_id
)
367 webpage
= self
._download
_webpage
(url
, video_id
, note
='Downloading video webpage')
368 json_string
= self
._search
_regex
(
369 r
'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)',
370 webpage
, 'json_string', group
='json_string_ld')
371 json_data
= self
._parse
_json
(json_string
, video_id
)
372 props_data
= try_get(json_data
, lambda x
: x
['props'], expected_type
=dict)
374 # Chech statusCode for success
375 status
= props_data
.get('pageProps').get('statusCode')
377 return self
._parse
_aweme
_video
_web
(props_data
['pageProps']['itemInfo']['itemStruct'], url
)
378 elif status
== 10216:
379 raise ExtractorError('This video is private', expected
=True)
381 raise ExtractorError('Video not available', video_id
=video_id
)
384 class TikTokUserIE(TikTokBaseIE
):
385 IE_NAME
= 'tiktok:user'
386 _VALID_URL
= r
'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
388 'url': 'https://tiktok.com/@corgibobaa?lang=en',
389 'playlist_mincount': 45,
391 'id': '6935371178089399301',
392 'title': 'corgibobaa',
394 'expected_warnings': ['Retrying']
396 'url': 'https://www.tiktok.com/@meme',
397 'playlist_mincount': 593,
399 'id': '79005827461758976',
402 'expected_warnings': ['Retrying']
405 r
''' # TODO: Fix by adding _signature to api_url
406 def _entries(self, webpage, user_id, username):
407 secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
408 verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
409 if not verifyfp_cookie:
410 raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
411 api_url = f'https://m.tiktok.com/api/post/item_list/?aid=1988&cookie_enabled=true&count=30&verifyFp={verifyfp_cookie.value}&secUid={secuid}&cursor='
413 for page in itertools.count():
414 data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
415 for video in data_json.get('itemList', []):
416 video_id = video['id']
417 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
418 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
419 if not data_json.get('hasMore'):
421 cursor = data_json['cursor']
424 def _entries_api(self
, webpage
, user_id
, username
):
430 'retry_type': 'no_retry',
431 'device_id': ''.join(random
.choice(string
.digits
) for _
in range(19)), # Some endpoints don't like randomized device_id, so it isn't directly set in _call_api.
434 max_retries
= self
.get_param('extractor_retries', 3)
435 for page
in itertools
.count(1):
436 for retries
in itertools
.count():
438 post_list
= self
._call
_api
('aweme/post', query
, username
,
439 note
='Downloading user video list page %d%s' % (page
, f
' (attempt {retries})' if retries
!= 0 else ''),
440 errnote
='Unable to download user video list')
441 except ExtractorError
as e
:
442 if isinstance(e
.cause
, json
.JSONDecodeError
) and e
.cause
.pos
== 0 and retries
!= max_retries
:
443 self
.report_warning('%s. Retrying...' % str(e
.cause
or e
.msg
))
447 for video
in post_list
.get('aweme_list', []):
449 **self
._parse
_aweme
_video
_app
(video
),
450 'ie_key': TikTokIE
.ie_key(),
451 'extractor': 'TikTok',
452 'webpage_url': f
'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
454 if not post_list
.get('has_more'):
456 query
['max_cursor'] = post_list
['max_cursor']
458 def _real_extract(self
, url
):
459 user_name
= self
._match
_id
(url
)
460 webpage
= self
._download
_webpage
(url
, user_name
, headers
={
461 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
463 user_id
= self
._html
_search
_regex
(r
'snssdk\d*://user/profile/(\d+)', webpage
, 'user ID')
464 return self
.playlist_result(self
._entries
_api
(webpage
, user_id
, user_name
), user_id
, user_name
)
467 class DouyinIE(TikTokIE
):
468 _VALID_URL
= r
'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
470 'url': 'https://www.douyin.com/video/6961737553342991651',
471 'md5': '10523312c8b8100f353620ac9dc8f067',
473 'id': '6961737553342991651',
475 'title': '#杨超越 小小水手带你去远航❤️',
477 'upload_date': '20210513',
478 'timestamp': 1620905839,
479 'uploader_id': '110403406559',
483 'comment_count': int,
486 'url': 'https://www.douyin.com/video/6982497745948921092',
487 'md5': 'd78408c984b9b5102904cf6b6bc2d712',
489 'id': '6982497745948921092',
491 'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
492 'uploader': '杨超越工作室',
493 'upload_date': '20210708',
494 'timestamp': 1625739481,
495 'uploader_id': '408654318141572',
499 'comment_count': int,
502 'url': 'https://www.douyin.com/video/6953975910773099811',
503 'md5': '72e882e24f75064c218b76c8b713c185',
505 'id': '6953975910773099811',
507 'title': '#一起看海 出现在你的夏日里',
509 'upload_date': '20210422',
510 'timestamp': 1619098692,
511 'uploader_id': '110403406559',
515 'comment_count': int,
518 'url': 'https://www.douyin.com/video/6950251282489675042',
519 'md5': 'b4db86aec367ef810ddd38b1737d2fed',
521 'id': '6950251282489675042',
523 'title': '哈哈哈,成功了哈哈哈哈哈哈',
525 'upload_date': '20210412',
526 'timestamp': 1618231483,
527 'uploader_id': '110403406559',
531 'comment_count': int,
534 'url': 'https://www.douyin.com/video/6963263655114722595',
535 'md5': '1abe1c477d05ee62efb40bf2329957cf',
537 'id': '6963263655114722595',
539 'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
541 'upload_date': '20210517',
542 'timestamp': 1621261163,
543 'uploader_id': '110403406559',
547 'comment_count': int,
550 _APP_VERSION
= '9.6.0'
551 _MANIFEST_APP_VERSION
= '960'
554 _API_HOSTNAME
= 'aweme.snssdk.com'
555 _UPLOADER_URL_FORMAT
= 'https://www.douyin.com/user/%s'
556 _WEBPAGE_HOST
= 'https://www.douyin.com/'
558 def _real_extract(self
, url
):
559 video_id
= self
._match
_id
(url
)
562 return self
._extract
_aweme
_app
(video_id
)
563 except ExtractorError
as e
:
564 self
.report_warning(f
'{e}; Retrying with webpage')
566 webpage
= self
._download
_webpage
(url
, video_id
)
567 render_data_json
= self
._search
_regex
(
568 r
'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^
>]*>(%7B
.+%7D
)</script
>',
569 webpage, 'render data
', default=None)
570 if not render_data_json:
571 # TODO: Run verification challenge code to generate signature cookies
572 raise ExtractorError('Fresh
cookies (not necessarily logged
in) are needed
')
574 render_data = self._parse_json(
575 render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
576 return self._parse_aweme_video_web(
577 traverse_obj(render_data, (..., 'aweme
', 'detail
'), get_all=False), url)