]> jfr.im git - yt-dlp.git/blame - yt_dlp/extractor/tiktok.py
projects / yt-dlp.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
[cleanup] Mark some compat variables for removal (#2173)
[yt-dlp.git] / yt_dlp / extractor / tiktok.py
CommitLineData
1ead840d
KS		 1# coding: utf-8
2from __future__ import unicode_literals
f7f18f90
A		 3
4import itertools
bd9ff55b
M		 5import random
6import string
7import time
0fd6661e 8import json
1ead840d
KS		 9
10from .common import InfoExtractor
be1f331f
M		 11from ..compat import (
12 compat_urllib_parse_unquote,
13 compat_urllib_parse_urlparse
14)
1ead840d 15from ..utils import (
ce18a19b 16 ExtractorError,
88afe056 17 HEADRequest,
11e1c2e3 18 UnsupportedError,
ff91cf74 19 get_first,
1ead840d 20 int_or_none,
34921b43 21 join_nonempty,
b3187433 22 LazyList,
e0585e65 23 srt_subtitles_timecode,
1ead840d 24 str_or_none,
bd9ff55b
M		 25 traverse_obj,
26 try_get,
943d5ab1 27 url_or_none,
bd9ff55b 28 qualities,
1ead840d
KS		 29)
30
31
0fd6661e 32class TikTokBaseIE(InfoExtractor):
046cab39
M		 33 _APP_VERSIONS = [('20.9.3', '293'), ('20.4.3', '243'), ('20.2.1', '221'), ('20.1.2', '212'), ('20.0.4', '204')]
34 _WORKING_APP_VERSION = None
943d5ab1
M		 35 _APP_NAME = 'trill'
36 _AID = 1180
53dad39e 37 _API_HOSTNAME = 'api-h2.tiktokv.com'
943d5ab1 38 _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
53dad39e 39 _WEBPAGE_HOST = 'https://www.tiktok.com/'
be1f331f 40 QUALITIES = ('360p', '540p', '720p', '1080p')
ce18a19b 41
046cab39
M		 42 def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
43 note='Downloading API JSON', errnote='Unable to download API page'):
44 self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
45 webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
46 if webpage_cookies.get('sid_tt'):
47 self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
48 return self._download_json(
49 'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
50 fatal=fatal, note=note, errnote=errnote, headers={
51 'User-Agent': f'com.ss.android.ugc.trill/{manifest_app_version} (Linux; U; Android 10; en_US; Pixel 4; Build/QQ3A.200805.001; Cronet/58.0.2991.0)',
52 'Accept': 'application/json',
53 }, query=query)
54
55 def _build_api_query(self, query, app_version, manifest_app_version):
56 return {
0fd6661e 57 **query,
046cab39
M		 58 'version_name': app_version,
59 'version_code': manifest_app_version,
60 'build_number': app_version,
61 'manifest_version_code': manifest_app_version,
62 'update_version_code': manifest_app_version,
0930b11f 63 'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
64 'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
bd9ff55b
M		 65 '_rticket': int(time.time() * 1000),
66 'ts': int(time.time()),
67 'device_brand': 'Google',
68 'device_type': 'Pixel 4',
69 'device_platform': 'android',
70 'resolution': '1080*1920',
71 'dpi': 420,
72 'os_version': '10',
73 'os_api': '29',
74 'carrier_region': 'US',
75 'sys_region': 'US',
76 'region': 'US',
943d5ab1 77 'app_name': self._APP_NAME,
bd9ff55b
M		 78 'app_language': 'en',
79 'language': 'en',
80 'timezone_name': 'America/New_York',
81 'timezone_offset': '-14400',
82 'channel': 'googleplay',
83 'ac': 'wifi',
84 'mcc_mnc': '310260',
85 'is_my_cn': 0,
943d5ab1 86 'aid': self._AID,
bd9ff55b
M		 87 'ssmix': 'a',
88 'as': 'a1qwert123',
89 'cp': 'cbfhckdckkde1',
90 }
046cab39
M		 91
92 def _call_api(self, ep, query, video_id, fatal=True,
93 note='Downloading API JSON', errnote='Unable to download API page'):
94 if not self._WORKING_APP_VERSION:
95 app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0]
96 manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0]
97 if app_version and manifest_app_version:
98 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
99 self.write_debug('Imported app version combo from extractor arguments')
100 elif app_version or manifest_app_version:
101 self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True)
102
103 if self._WORKING_APP_VERSION:
104 app_version, manifest_app_version = self._WORKING_APP_VERSION
105 real_query = self._build_api_query(query, app_version, manifest_app_version)
106 return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
107
108 for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1):
109 real_query = self._build_api_query(query, app_version, manifest_app_version)
110 try:
111 res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
112 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
113 return res
114 except ExtractorError as e:
115 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
116 if count == len(self._APP_VERSIONS):
117 if fatal:
118 raise e
119 else:
120 self.report_warning(str(e.cause or e.msg))
121 return
122 self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS)))
123 continue
124 raise e
0fd6661e 125
e0585e65
M		 126 def _get_subtitles(self, aweme_detail, aweme_id):
127 # TODO: Extract text positioning info
128 subtitles = {}
129 captions_info = traverse_obj(
130 aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict, default=[])
131 for caption in captions_info:
132 caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
133 if not caption_url:
134 continue
135 caption_json = self._download_json(
136 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
137 if not caption_json:
138 continue
139 subtitles.setdefault(caption.get('language', 'en'), []).append({
140 'ext': 'srt',
141 'data': '\n\n'.join(
142 f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
143 for i, line in enumerate(caption_json['utterances']) if line.get('text'))
144 })
145 return subtitles
146
943d5ab1 147 def _parse_aweme_video_app(self, aweme_detail):
0fd6661e 148 aweme_id = aweme_detail['aweme_id']
bd9ff55b
M		 149 video_info = aweme_detail['video']
150
151 def parse_url_key(url_key):
152 format_id, codec, res, bitrate = self._search_regex(
153 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
154 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
155 if not format_id:
156 return {}, None
157 return {
158 'format_id': format_id,
159 'vcodec': 'h265' if codec == 'bytevc1' else codec,
160 'tbr': int_or_none(bitrate, scale=1000) or None,
161 'quality': qualities(self.QUALITIES)(res),
162 }, res
163
164 known_resolutions = {}
165
166 def extract_addr(addr, add_meta={}):
167 parsed_meta, res = parse_url_key(addr.get('url_key', ''))
168 if res:
169 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
170 known_resolutions[res].setdefault('width', add_meta.get('width'))
171 parsed_meta.update(known_resolutions.get(res, {}))
172 add_meta.setdefault('height', int_or_none(res[:-1]))
173 return [{
174 'url': url,
175 'filesize': int_or_none(addr.get('data_size')),
176 'ext': 'mp4',
177 'acodec': 'aac',
0fd6661e
M		 178 'source_preference': -2 if 'aweme/v1' in url else -1, # Downloads from API might get blocked
179 **add_meta, **parsed_meta,
34921b43 180 'format_note': join_nonempty(
181 add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
bd9ff55b
M		 182 } for url in addr.get('url_list') or []]
183
184 # Hack: Add direct video links first to prioritize them when removing duplicate formats
185 formats = []
186 if video_info.get('play_addr'):
187 formats.extend(extract_addr(video_info['play_addr'], {
188 'format_id': 'play_addr',
189 'format_note': 'Direct video',
190 'vcodec': 'h265' if traverse_obj(
be1f331f 191 video_info, 'is_bytevc1', 'is_h265') else 'h264', # TODO: Check for "direct iOS" videos, like https://www.tiktok.com/@cookierun_dev/video/7039716639834656002
bd9ff55b
M		 192 'width': video_info.get('width'),
193 'height': video_info.get('height'),
194 }))
195 if video_info.get('download_addr'):
196 formats.extend(extract_addr(video_info['download_addr'], {
197 'format_id': 'download_addr',
198 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
199 'vcodec': 'h264',
200 'width': video_info.get('width'),
201 'height': video_info.get('height'),
0fd6661e 202 'preference': -2 if video_info.get('has_watermark') else -1,
bd9ff55b
M		 203 }))
204 if video_info.get('play_addr_h264'):
205 formats.extend(extract_addr(video_info['play_addr_h264'], {
206 'format_id': 'play_addr_h264',
207 'format_note': 'Direct video',
208 'vcodec': 'h264',
209 }))
210 if video_info.get('play_addr_bytevc1'):
211 formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
212 'format_id': 'play_addr_bytevc1',
213 'format_note': 'Direct video',
214 'vcodec': 'h265',
215 }))
216
217 for bitrate in video_info.get('bit_rate', []):
218 if bitrate.get('play_addr'):
219 formats.extend(extract_addr(bitrate['play_addr'], {
220 'format_id': bitrate.get('gear_name'),
221 'format_note': 'Playback video',
222 'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
223 'vcodec': 'h265' if traverse_obj(
224 bitrate, 'is_bytevc1', 'is_h265') else 'h264',
943d5ab1 225 'fps': bitrate.get('FPS'),
bd9ff55b
M		 226 }))
227
228 self._remove_duplicate_formats(formats)
6134fbeb
M		 229 auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
230 if auth_cookie:
231 for f in formats:
be1f331f 232 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
0fd6661e 233 self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
bd9ff55b
M		 234
235 thumbnails = []
236 for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
237 'origin_cover', 'dynamic_cover'):
238 cover = video_info.get(cover_id)
239 if cover:
240 for cover_url in cover['url_list']:
241 thumbnails.append({
242 'id': cover_id,
243 'url': cover_url,
244 })
245
246 stats_info = aweme_detail.get('statistics', {})
247 author_info = aweme_detail.get('author', {})
248 music_info = aweme_detail.get('music', {})
943d5ab1
M		 249 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
250 'sec_uid', 'id', 'uid', 'unique_id',
251 expected_type=str_or_none, get_all=False))
a8549f19 252 labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
bd9ff55b
M		 253
254 contained_music_track = traverse_obj(
255 music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
256 contained_music_author = traverse_obj(
257 music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
258
259 is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
260 if is_generic_og_trackname:
261 music_track, music_author = contained_music_track or 'original sound', contained_music_author
262 else:
263 music_track, music_author = music_info.get('title'), music_info.get('author')
264
265 return {
266 'id': aweme_id,
5fa3c9a8 267 'title': aweme_detail.get('desc'),
a44ca5a4 268 'description': aweme_detail.get('desc'),
bd9ff55b
M		 269 'view_count': int_or_none(stats_info.get('play_count')),
270 'like_count': int_or_none(stats_info.get('digg_count')),
271 'repost_count': int_or_none(stats_info.get('share_count')),
272 'comment_count': int_or_none(stats_info.get('comment_count')),
273 'uploader': str_or_none(author_info.get('unique_id')),
943d5ab1 274 'creator': str_or_none(author_info.get('nickname')),
bd9ff55b 275 'uploader_id': str_or_none(author_info.get('uid')),
943d5ab1 276 'uploader_url': user_url,
bd9ff55b
M		 277 'track': music_track,
278 'album': str_or_none(music_info.get('album')) or None,
279 'artist': music_author,
280 'timestamp': int_or_none(aweme_detail.get('create_time')),
281 'formats': formats,
e0585e65 282 'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
bd9ff55b 283 'thumbnails': thumbnails,
53dad39e
M		 284 'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
285 'availability': self._availability(
286 is_private='Private' in labels,
287 needs_subscription='Friends only' in labels,
288 is_unlisted='Followers only' in labels)
bd9ff55b
M		 289 }
290
0481e266 291 def _parse_aweme_video_web(self, aweme_detail, webpage_url):
943d5ab1 292 video_info = aweme_detail['video']
11aa91a1 293 author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
943d5ab1
M		 294 music_info = aweme_detail.get('music') or {}
295 stats_info = aweme_detail.get('stats') or {}
296 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
297 'secUid', 'id', 'uid', 'uniqueId',
11aa91a1
M		 298 expected_type=str_or_none, get_all=False)
299 or aweme_detail.get('authorSecId'))
943d5ab1
M		 300
301 formats = []
302 play_url = video_info.get('playAddr')
303 width = video_info.get('width')
304 height = video_info.get('height')
305 if isinstance(play_url, str):
306 formats = [{
307 'url': self._proto_relative_url(play_url),
308 'ext': 'mp4',
309 'width': width,
310 'height': height,
311 }]
312 elif isinstance(play_url, list):
313 formats = [{
314 'url': self._proto_relative_url(url),
315 'ext': 'mp4',
316 'width': width,
317 'height': height,
318 } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
319
320 download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
321 if download_url:
322 formats.append({
323 'format_id': 'download',
324 'url': self._proto_relative_url(download_url),
325 'ext': 'mp4',
326 'width': width,
327 'height': height,
328 })
329 self._remove_duplicate_formats(formats)
330 self._sort_formats(formats)
331
332 thumbnails = []
333 for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
334 if aweme_detail.get(thumbnail_name):
335 thumbnails = [{
336 'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
337 'width': width,
338 'height': height
339 }]
340
341 return {
342 'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
343 'title': aweme_detail.get('desc'),
344 'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
345 'view_count': int_or_none(stats_info.get('playCount')),
346 'like_count': int_or_none(stats_info.get('diggCount')),
347 'repost_count': int_or_none(stats_info.get('shareCount')),
348 'comment_count': int_or_none(stats_info.get('commentCount')),
349 'timestamp': int_or_none(aweme_detail.get('createTime')),
350 'creator': str_or_none(author_info.get('nickname')),
11aa91a1
M		 351 'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
352 'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
943d5ab1
M		 353 'uploader_url': user_url,
354 'track': str_or_none(music_info.get('title')),
355 'album': str_or_none(music_info.get('album')) or None,
356 'artist': str_or_none(music_info.get('authorName')),
357 'formats': formats,
358 'thumbnails': thumbnails,
359 'description': str_or_none(aweme_detail.get('desc')),
360 'http_headers': {
0481e266 361 'Referer': webpage_url
943d5ab1
M		 362 }
363 }
364
0fd6661e
M		 365
366class TikTokIE(TikTokBaseIE):
367 _VALID_URL = r'https?://www\.tiktok\.com/@[\w\.-]+/video/(?P<id>\d+)'
368
369 _TESTS = [{
370 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
0481e266 371 'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
0fd6661e
M		 372 'info_dict': {
373 'id': '6748451240264420610',
374 'ext': 'mp4',
375 'title': '#jassmanak #lehanga #leenabhushan',
376 'description': '#jassmanak #lehanga #leenabhushan',
377 'duration': 13,
0481e266 378 'height': 1024,
379 'width': 576,
0fd6661e
M		 380 'uploader': 'leenabhushan',
381 'uploader_id': '6691488002098119685',
0481e266 382 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
0fd6661e
M		 383 'creator': 'facestoriesbyleenabh',
384 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
385 'upload_date': '20191016',
386 'timestamp': 1571246252,
387 'view_count': int,
388 'like_count': int,
389 'repost_count': int,
390 'comment_count': int,
a44ca5a4 391 'artist': 'Ysrbeats',
392 'album': 'Lehanga',
393 'track': 'Lehanga',
0fd6661e
M		 394 }
395 }, {
396 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
0481e266 397 'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
0fd6661e
M		 398 'info_dict': {
399 'id': '6742501081818877190',
400 'ext': 'mp4',
401 'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
402 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
403 'duration': 27,
404 'height': 960,
405 'width': 540,
406 'uploader': 'patrox',
407 'uploader_id': '18702747',
0481e266 408 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
0fd6661e
M		 409 'creator': 'patroX',
410 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
411 'upload_date': '20190930',
412 'timestamp': 1569860870,
413 'view_count': int,
414 'like_count': int,
415 'repost_count': int,
416 'comment_count': int,
a44ca5a4 417 'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson',
418 'track': 'Big Fun',
0fd6661e
M		 419 }
420 }, {
96f13f01
M		 421 # Banned audio, only available on the app
422 'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
423 'info_dict': {
424 'id': '6984138651336838402',
425 'ext': 'mp4',
426 'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
427 'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
428 'uploader': 'barudakhb_',
429 'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
430 'uploader_id': '6974687867511718913',
431 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
432 'track': 'Boka Dance',
433 'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
434 'timestamp': 1626121503,
435 'duration': 18,
436 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
437 'upload_date': '20210712',
438 'view_count': int,
439 'like_count': int,
440 'repost_count': int,
441 'comment_count': int,
442 }
443 }, {
444 # Sponsored video, only available with feed workaround
445 'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
446 'info_dict': {
447 'id': '7042692929109986561',
448 'ext': 'mp4',
449 'title': 'Slap and Run!',
450 'description': 'Slap and Run!',
451 'uploader': 'user440922249',
452 'creator': 'Slap And Run',
453 'uploader_id': '7036055384943690754',
454 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
455 'track': 'Promoted Music',
456 'timestamp': 1639754738,
457 'duration': 30,
458 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
459 'upload_date': '20211217',
460 'view_count': int,
461 'like_count': int,
462 'repost_count': int,
463 'comment_count': int,
464 },
465 'expected_warnings': ['Video not available']
5fa3c9a8
HTL		 466 }, {
467 # Video without title and description
468 'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
469 'info_dict': {
470 'id': '7059698374567611694',
471 'ext': 'mp4',
a44ca5a4 472 'title': 'tiktok video #7059698374567611694',
5fa3c9a8
HTL		 473 'description': '',
474 'uploader': 'pokemonlife22',
475 'creator': 'Pokemon',
476 'uploader_id': '6820838815978423302',
477 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
478 'track': 'original sound',
479 'timestamp': 1643714123,
480 'duration': 6,
481 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
482 'upload_date': '20220201',
483 'artist': 'Pokemon',
484 'view_count': int,
485 'like_count': int,
486 'repost_count': int,
487 'comment_count': int,
488 },
a44ca5a4 489 'expected_warnings': ['Video not available', 'Creating a generic title']
e0585e65
M		 490 }, {
491 # Auto-captions available
492 'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
493 'only_matching': True
0fd6661e
M		 494 }]
495
0fd6661e 496 def _extract_aweme_app(self, aweme_id):
e540c56f
M		 497 try:
498 aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id,
499 note='Downloading video details', errnote='Unable to download video details').get('aweme_detail')
500 if not aweme_detail:
501 raise ExtractorError('Video not available', video_id=aweme_id)
502 except ExtractorError as e:
503 self.report_warning(f'{e}; Retrying with feed workaround')
504 feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
505 note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
be1f331f 506 aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
e540c56f
M		 507 if not aweme_detail:
508 raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
943d5ab1 509 return self._parse_aweme_video_app(aweme_detail)
0fd6661e 510
ce18a19b
S		 511 def _real_extract(self, url):
512 video_id = self._match_id(url)
ce18a19b 513
bd9ff55b
M		 514 try:
515 return self._extract_aweme_app(video_id)
516 except ExtractorError as e:
517 self.report_warning(f'{e}; Retrying with webpage')
518
7bbc0bbc 519 # If we only call once, we get a 403 when downlaoding the video.
61e76c1e 520 self._download_webpage(url, video_id)
6fb11ca8 521 webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
135dfa2c 522 next_data = self._search_nextjs_data(webpage, video_id, default='{}')
11aa91a1 523
135dfa2c 524 if next_data:
11aa91a1
M		 525 status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
526 video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
527 else:
528 sigi_json = self._search_regex(
529 r'>\s*window\[[\'"]SIGI_STATE[\'"]\]\s*=\s*(?P<sigi_state>{.+});',
530 webpage, 'sigi data', group='sigi_state')
531 sigi_data = self._parse_json(sigi_json, video_id)
532 status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
533 video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
534
1418a043 535 if status == 0:
11aa91a1 536 return self._parse_aweme_video_web(video_data, url)
1418a043 537 elif status == 10216:
538 raise ExtractorError('This video is private', expected=True)
6fb11ca8 539 raise ExtractorError('Video not available', video_id=video_id)
f7f18f90
A		 540
541
0fd6661e 542class TikTokUserIE(TikTokBaseIE):
f7f18f90 543 IE_NAME = 'tiktok:user'
0fd6661e 544 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
f7f18f90 545 _TESTS = [{
526d74ec 546 'url': 'https://tiktok.com/@corgibobaa?lang=en',
f7f18f90
A		 547 'playlist_mincount': 45,
548 'info_dict': {
549 'id': '6935371178089399301',
0481e266 550 'title': 'corgibobaa',
b3187433 551 'thumbnail': r're:https://.+_1080x1080\.webp'
f7f18f90 552 },
0481e266 553 'expected_warnings': ['Retrying']
5fa3c9a8
HTL		 554 }, {
555 'url': 'https://www.tiktok.com/@6820838815978423302',
556 'playlist_mincount': 5,
557 'info_dict': {
558 'id': '6820838815978423302',
559 'title': '6820838815978423302',
560 'thumbnail': r're:https://.+_1080x1080\.webp'
561 },
562 'expected_warnings': ['Retrying']
f7f18f90
A		 563 }, {
564 'url': 'https://www.tiktok.com/@meme',
565 'playlist_mincount': 593,
566 'info_dict': {
567 'id': '79005827461758976',
0481e266 568 'title': 'meme',
b3187433 569 'thumbnail': r're:https://.+_1080x1080\.webp'
f7f18f90 570 },
0481e266 571 'expected_warnings': ['Retrying']
f7f18f90
A		 572 }]
573
0fd6661e
M		 574 r''' # TODO: Fix by adding _signature to api_url
575 def _entries(self, webpage, user_id, username):
576 secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
f7f18f90
A		 577 verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
578 if not verifyfp_cookie:
579 raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
580 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='
581 cursor = '0'
582 for page in itertools.count():
0fd6661e 583 data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
f7f18f90
A		 584 for video in data_json.get('itemList', []):
585 video_id = video['id']
586 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
bd9ff55b 587 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
526d74ec 588 if not data_json.get('hasMore'):
f7f18f90
A		 589 break
590 cursor = data_json['cursor']
0fd6661e
M		 591 '''
592
b3187433 593 def _video_entries_api(self, webpage, user_id, username):
0fd6661e
M		 594 query = {
595 'user_id': user_id,
596 'count': 21,
597 'max_cursor': 0,
598 'min_cursor': 0,
599 'retry_type': 'no_retry',
0930b11f 600 '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.
0fd6661e
M		 601 }
602
603 max_retries = self.get_param('extractor_retries', 3)
604 for page in itertools.count(1):
605 for retries in itertools.count():
606 try:
607 post_list = self._call_api('aweme/post', query, username,
608 note='Downloading user video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
609 errnote='Unable to download user video list')
610 except ExtractorError as e:
611 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
612 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
613 continue
614 raise
615 break
b3187433 616 yield from post_list.get('aweme_list', [])
0fd6661e
M		 617 if not post_list.get('has_more'):
618 break
619 query['max_cursor'] = post_list['max_cursor']
f7f18f90 620
b3187433 621 def _entries_api(self, user_id, videos):
622 for video in videos:
623 yield {
624 **self._parse_aweme_video_app(video),
625 'extractor_key': TikTokIE.ie_key(),
626 'extractor': 'TikTok',
627 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
628 }
629
f7f18f90 630 def _real_extract(self, url):
0481e266 631 user_name = self._match_id(url)
632 webpage = self._download_webpage(url, user_name, headers={
0fd6661e
M		 633 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
634 })
5fa3c9a8 635 user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
b3187433 636
637 videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
638 thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
639
640 return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
943d5ab1
M		 641
642
8126298c
M		 643class TikTokBaseListIE(TikTokBaseIE):
644 def _entries(self, list_id, display_id):
645 query = {
646 self._QUERY_NAME: list_id,
647 'cursor': 0,
648 'count': 20,
649 'type': 5,
650 'device_id': ''.join(random.choice(string.digits) for i in range(19))
651 }
652
653 max_retries = self.get_param('extractor_retries', 3)
654 for page in itertools.count(1):
655 for retries in itertools.count():
656 try:
657 post_list = self._call_api(self._API_ENDPOINT, query, display_id,
658 note='Downloading video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
659 errnote='Unable to download video list')
660 except ExtractorError as e:
661 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
662 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
663 continue
664 raise
665 break
666 for video in post_list.get('aweme_list', []):
667 yield {
668 **self._parse_aweme_video_app(video),
0b77924a 669 'extractor_key': TikTokIE.ie_key(),
8126298c
M		 670 'extractor': 'TikTok',
671 'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
672 }
673 if not post_list.get('has_more'):
674 break
675 query['cursor'] = post_list['cursor']
676
677 def _real_extract(self, url):
678 list_id = self._match_id(url)
679 return self.playlist_result(self._entries(list_id, list_id), list_id)
680
681
682class TikTokSoundIE(TikTokBaseListIE):
683 IE_NAME = 'tiktok:sound'
684 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
685 _QUERY_NAME = 'music_id'
686 _API_ENDPOINT = 'music/aweme'
687 _TESTS = [{
688 'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
689 'playlist_mincount': 100,
690 'info_dict': {
691 'id': '6956990112127585029'
692 },
693 'expected_warnings': ['Retrying']
694 }, {
695 # Actual entries are less than listed video count
696 'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
697 'playlist_mincount': 2182,
698 'info_dict': {
699 'id': '7036843036118469381'
700 },
701 'expected_warnings': ['Retrying']
702 }]
703
704
705class TikTokEffectIE(TikTokBaseListIE):
706 IE_NAME = 'tiktok:effect'
707 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
708 _QUERY_NAME = 'sticker_id'
709 _API_ENDPOINT = 'sticker/aweme'
710 _TESTS = [{
711 'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
712 'playlist_mincount': 100,
713 'info_dict': {
714 'id': '1258156',
715 },
716 'expected_warnings': ['Retrying']
717 }, {
718 # Different entries between mobile and web, depending on region
719 'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
720 'only_matching': True
721 }]
722
723
724class TikTokTagIE(TikTokBaseListIE):
725 IE_NAME = 'tiktok:tag'
726 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
727 _QUERY_NAME = 'ch_id'
728 _API_ENDPOINT = 'challenge/aweme'
729 _TESTS = [{
730 'url': 'https://tiktok.com/tag/hello2018',
731 'playlist_mincount': 39,
732 'info_dict': {
733 'id': '46294678',
734 'title': 'hello2018',
735 },
736 'expected_warnings': ['Retrying']
737 }, {
738 'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
739 'only_matching': True
740 }]
741
742 def _real_extract(self, url):
743 display_id = self._match_id(url)
744 webpage = self._download_webpage(url, display_id, headers={
745 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
746 })
747 tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
748 return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
749
750
943d5ab1
M		 751class DouyinIE(TikTokIE):
752 _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
753 _TESTS = [{
754 'url': 'https://www.douyin.com/video/6961737553342991651',
755 'md5': '10523312c8b8100f353620ac9dc8f067',
756 'info_dict': {
757 'id': '6961737553342991651',
758 'ext': 'mp4',
759 'title': '#杨超越 小小水手带你去远航❤️',
760 'uploader': '杨超越',
761 'upload_date': '20210513',
762 'timestamp': 1620905839,
763 'uploader_id': '110403406559',
764 'view_count': int,
765 'like_count': int,
766 'repost_count': int,
767 'comment_count': int,
768 }
769 }, {
770 'url': 'https://www.douyin.com/video/6982497745948921092',
771 'md5': 'd78408c984b9b5102904cf6b6bc2d712',
772 'info_dict': {
773 'id': '6982497745948921092',
774 'ext': 'mp4',
775 'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
776 'uploader': '杨超越工作室',
777 'upload_date': '20210708',
778 'timestamp': 1625739481,
779 'uploader_id': '408654318141572',
780 'view_count': int,
781 'like_count': int,
782 'repost_count': int,
783 'comment_count': int,
784 }
785 }, {
786 'url': 'https://www.douyin.com/video/6953975910773099811',
787 'md5': '72e882e24f75064c218b76c8b713c185',
788 'info_dict': {
789 'id': '6953975910773099811',
790 'ext': 'mp4',
791 'title': '#一起看海 出现在你的夏日里',
792 'uploader': '杨超越',
793 'upload_date': '20210422',
794 'timestamp': 1619098692,
795 'uploader_id': '110403406559',
796 'view_count': int,
797 'like_count': int,
798 'repost_count': int,
799 'comment_count': int,
800 }
801 }, {
802 'url': 'https://www.douyin.com/video/6950251282489675042',
803 'md5': 'b4db86aec367ef810ddd38b1737d2fed',
804 'info_dict': {
805 'id': '6950251282489675042',
806 'ext': 'mp4',
807 'title': '哈哈哈，成功了哈哈哈哈哈哈',
808 'uploader': '杨超越',
809 'upload_date': '20210412',
810 'timestamp': 1618231483,
811 'uploader_id': '110403406559',
812 'view_count': int,
813 'like_count': int,
814 'repost_count': int,
815 'comment_count': int,
816 }
817 }, {
818 'url': 'https://www.douyin.com/video/6963263655114722595',
819 'md5': '1abe1c477d05ee62efb40bf2329957cf',
820 'info_dict': {
821 'id': '6963263655114722595',
822 'ext': 'mp4',
823 'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
824 'uploader': '杨超越',
825 'upload_date': '20210517',
826 'timestamp': 1621261163,
827 'uploader_id': '110403406559',
828 'view_count': int,
829 'like_count': int,
830 'repost_count': int,
831 'comment_count': int,
832 }
833 }]
046cab39 834 _APP_VERSIONS = [('9.6.0', '960')]
943d5ab1
M		 835 _APP_NAME = 'aweme'
836 _AID = 1128
837 _API_HOSTNAME = 'aweme.snssdk.com'
838 _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
53dad39e 839 _WEBPAGE_HOST = 'https://www.douyin.com/'
943d5ab1
M		 840
841 def _real_extract(self, url):
842 video_id = self._match_id(url)
843
844 try:
845 return self._extract_aweme_app(video_id)
846 except ExtractorError as e:
847 self.report_warning(f'{e}; Retrying with webpage')
848
849 webpage = self._download_webpage(url, video_id)
850 render_data_json = self._search_regex(
851 r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
852 webpage, 'render data', default=None)
853 if not render_data_json:
854 # TODO: Run verification challenge code to generate signature cookies
855 raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
856
857 render_data = self._parse_json(
858 render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
ff91cf74 859 return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url)
88afe056 860
861
49895f06 862class TikTokVMIE(InfoExtractor):
863 _VALID_URL = r'https?://(?:vm|vt)\.tiktok\.com/(?P<id>\w+)'
88afe056 864 IE_NAME = 'vm.tiktok'
865
49895f06 866 _TESTS = [{
867 'url': 'https://vm.tiktok.com/ZSe4FqkKd',
868 'info_dict': {
869 'id': '7023491746608712966',
870 'ext': 'mp4',
871 'title': 'md5:5607564db90271abbbf8294cca77eddd',
872 'description': 'md5:5607564db90271abbbf8294cca77eddd',
873 'duration': 11,
874 'upload_date': '20211026',
875 'uploader_id': '7007385080558846981',
876 'creator': 'Memes',
877 'artist': 'Memes',
878 'track': 'original sound',
879 'uploader': 'susmandem',
880 'timestamp': 1635284105,
881 'thumbnail': r're:https://.+\.webp.*',
882 'like_count': int,
883 'view_count': int,
884 'comment_count': int,
885 'repost_count': int,
886 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAXcNoOEOxVyBzuII_E--T0MeCrLP0ay1Sm6x_n3dluiWEoWZD0VlQOytwad4W0i0n',
887 }
888 }, {
889 'url': 'https://vt.tiktok.com/ZSe4FqkKd',
890 'only_matching': True,
891 }]
892
88afe056 893 def _real_extract(self, url):
11e1c2e3 894 new_url = self._request_webpage(
895 HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl()
896 if self.suitable(new_url): # Prevent infinite loop in case redirect fails
897 raise UnsupportedError(new_url)
898 return self.url_result(new_url)
fork of https://github.com/yt-dlp/yt-dlp/