]> 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
[tiktok] Detect embeds
[yt-dlp.git] / yt_dlp / extractor / tiktok.py
CommitLineData
f7f18f90 1import itertools
b801cd71 2import json
bd9ff55b 3import random
b801cd71 4import re
bd9ff55b
M		 5import string
6import time
1ead840d
KS		 7
8from .common import InfoExtractor
b801cd71 9from ..compat import compat_urllib_parse_unquote, compat_urllib_parse_urlparse
1ead840d 10from ..utils import (
ce18a19b 11 ExtractorError,
88afe056 12 HEADRequest,
b801cd71 13 LazyList,
11e1c2e3 14 UnsupportedError,
ff91cf74 15 get_first,
1ead840d 16 int_or_none,
34921b43 17 join_nonempty,
b801cd71 18 qualities,
e0585e65 19 srt_subtitles_timecode,
1ead840d 20 str_or_none,
bd9ff55b
M		 21 traverse_obj,
22 try_get,
943d5ab1 23 url_or_none,
1ead840d
KS		 24)
25
26
0fd6661e 27class TikTokBaseIE(InfoExtractor):
046cab39
M		 28 _APP_VERSIONS = [('20.9.3', '293'), ('20.4.3', '243'), ('20.2.1', '221'), ('20.1.2', '212'), ('20.0.4', '204')]
29 _WORKING_APP_VERSION = None
943d5ab1
M		 30 _APP_NAME = 'trill'
31 _AID = 1180
53dad39e 32 _API_HOSTNAME = 'api-h2.tiktokv.com'
943d5ab1 33 _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
53dad39e 34 _WEBPAGE_HOST = 'https://www.tiktok.com/'
be1f331f 35 QUALITIES = ('360p', '540p', '720p', '1080p')
ce18a19b 36
b801cd71 37 @staticmethod
38 def _create_url(user_id, video_id):
39 return f'https://www.tiktok.com/@{user_id or "_"}/video/{video_id}'
40
046cab39
M		 41 def _call_api_impl(self, ep, query, manifest_app_version, video_id, fatal=True,
42 note='Downloading API JSON', errnote='Unable to download API page'):
43 self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
44 webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
45 if webpage_cookies.get('sid_tt'):
46 self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
47 return self._download_json(
48 'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
49 fatal=fatal, note=note, errnote=errnote, headers={
50 '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)',
51 'Accept': 'application/json',
52 }, query=query)
53
54 def _build_api_query(self, query, app_version, manifest_app_version):
55 return {
0fd6661e 56 **query,
046cab39
M		 57 'version_name': app_version,
58 'version_code': manifest_app_version,
59 'build_number': app_version,
60 'manifest_version_code': manifest_app_version,
61 'update_version_code': manifest_app_version,
0930b11f 62 'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
63 'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
bd9ff55b
M		 64 '_rticket': int(time.time() * 1000),
65 'ts': int(time.time()),
66 'device_brand': 'Google',
67 'device_type': 'Pixel 4',
68 'device_platform': 'android',
69 'resolution': '1080*1920',
70 'dpi': 420,
71 'os_version': '10',
72 'os_api': '29',
73 'carrier_region': 'US',
74 'sys_region': 'US',
75 'region': 'US',
943d5ab1 76 'app_name': self._APP_NAME,
bd9ff55b
M		 77 'app_language': 'en',
78 'language': 'en',
79 'timezone_name': 'America/New_York',
80 'timezone_offset': '-14400',
81 'channel': 'googleplay',
82 'ac': 'wifi',
83 'mcc_mnc': '310260',
84 'is_my_cn': 0,
943d5ab1 85 'aid': self._AID,
bd9ff55b
M		 86 'ssmix': 'a',
87 'as': 'a1qwert123',
88 'cp': 'cbfhckdckkde1',
89 }
046cab39
M		 90
91 def _call_api(self, ep, query, video_id, fatal=True,
92 note='Downloading API JSON', errnote='Unable to download API page'):
93 if not self._WORKING_APP_VERSION:
94 app_version = self._configuration_arg('app_version', [''], ie_key=TikTokIE.ie_key())[0]
95 manifest_app_version = self._configuration_arg('manifest_app_version', [''], ie_key=TikTokIE.ie_key())[0]
96 if app_version and manifest_app_version:
97 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
98 self.write_debug('Imported app version combo from extractor arguments')
99 elif app_version or manifest_app_version:
100 self.report_warning('Only one of the two required version params are passed as extractor arguments', only_once=True)
101
102 if self._WORKING_APP_VERSION:
103 app_version, manifest_app_version = self._WORKING_APP_VERSION
104 real_query = self._build_api_query(query, app_version, manifest_app_version)
105 return self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
106
107 for count, (app_version, manifest_app_version) in enumerate(self._APP_VERSIONS, start=1):
108 real_query = self._build_api_query(query, app_version, manifest_app_version)
109 try:
110 res = self._call_api_impl(ep, real_query, manifest_app_version, video_id, fatal, note, errnote)
111 self._WORKING_APP_VERSION = (app_version, manifest_app_version)
112 return res
113 except ExtractorError as e:
114 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0:
115 if count == len(self._APP_VERSIONS):
116 if fatal:
117 raise e
118 else:
119 self.report_warning(str(e.cause or e.msg))
120 return
121 self.report_warning('%s. Retrying... (attempt %s of %s)' % (str(e.cause or e.msg), count, len(self._APP_VERSIONS)))
122 continue
123 raise e
0fd6661e 124
e0585e65
M		 125 def _get_subtitles(self, aweme_detail, aweme_id):
126 # TODO: Extract text positioning info
127 subtitles = {}
128 captions_info = traverse_obj(
129 aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict, default=[])
130 for caption in captions_info:
131 caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
132 if not caption_url:
133 continue
134 caption_json = self._download_json(
135 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
136 if not caption_json:
137 continue
138 subtitles.setdefault(caption.get('language', 'en'), []).append({
139 'ext': 'srt',
140 'data': '\n\n'.join(
141 f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
142 for i, line in enumerate(caption_json['utterances']) if line.get('text'))
143 })
144 return subtitles
145
943d5ab1 146 def _parse_aweme_video_app(self, aweme_detail):
0fd6661e 147 aweme_id = aweme_detail['aweme_id']
bd9ff55b
M		 148 video_info = aweme_detail['video']
149
150 def parse_url_key(url_key):
151 format_id, codec, res, bitrate = self._search_regex(
152 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
153 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
154 if not format_id:
155 return {}, None
156 return {
157 'format_id': format_id,
158 'vcodec': 'h265' if codec == 'bytevc1' else codec,
159 'tbr': int_or_none(bitrate, scale=1000) or None,
160 'quality': qualities(self.QUALITIES)(res),
161 }, res
162
163 known_resolutions = {}
164
165 def extract_addr(addr, add_meta={}):
166 parsed_meta, res = parse_url_key(addr.get('url_key', ''))
167 if res:
168 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
169 known_resolutions[res].setdefault('width', add_meta.get('width'))
170 parsed_meta.update(known_resolutions.get(res, {}))
171 add_meta.setdefault('height', int_or_none(res[:-1]))
172 return [{
173 'url': url,
174 'filesize': int_or_none(addr.get('data_size')),
175 'ext': 'mp4',
176 'acodec': 'aac',
0fd6661e
M		 177 'source_preference': -2 if 'aweme/v1' in url else -1, # Downloads from API might get blocked
178 **add_meta, **parsed_meta,
34921b43 179 'format_note': join_nonempty(
180 add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
bd9ff55b
M		 181 } for url in addr.get('url_list') or []]
182
183 # Hack: Add direct video links first to prioritize them when removing duplicate formats
184 formats = []
185 if video_info.get('play_addr'):
186 formats.extend(extract_addr(video_info['play_addr'], {
187 'format_id': 'play_addr',
188 'format_note': 'Direct video',
189 'vcodec': 'h265' if traverse_obj(
be1f331f 190 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		 191 'width': video_info.get('width'),
192 'height': video_info.get('height'),
193 }))
194 if video_info.get('download_addr'):
195 formats.extend(extract_addr(video_info['download_addr'], {
196 'format_id': 'download_addr',
197 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
198 'vcodec': 'h264',
199 'width': video_info.get('width'),
200 'height': video_info.get('height'),
0fd6661e 201 'preference': -2 if video_info.get('has_watermark') else -1,
bd9ff55b
M		 202 }))
203 if video_info.get('play_addr_h264'):
204 formats.extend(extract_addr(video_info['play_addr_h264'], {
205 'format_id': 'play_addr_h264',
206 'format_note': 'Direct video',
207 'vcodec': 'h264',
208 }))
209 if video_info.get('play_addr_bytevc1'):
210 formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
211 'format_id': 'play_addr_bytevc1',
212 'format_note': 'Direct video',
213 'vcodec': 'h265',
214 }))
215
216 for bitrate in video_info.get('bit_rate', []):
217 if bitrate.get('play_addr'):
218 formats.extend(extract_addr(bitrate['play_addr'], {
219 'format_id': bitrate.get('gear_name'),
220 'format_note': 'Playback video',
221 'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
222 'vcodec': 'h265' if traverse_obj(
223 bitrate, 'is_bytevc1', 'is_h265') else 'h264',
943d5ab1 224 'fps': bitrate.get('FPS'),
bd9ff55b
M		 225 }))
226
227 self._remove_duplicate_formats(formats)
6134fbeb
M		 228 auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
229 if auth_cookie:
230 for f in formats:
be1f331f 231 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
0fd6661e 232 self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
bd9ff55b
M		 233
234 thumbnails = []
235 for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
236 'origin_cover', 'dynamic_cover'):
237 cover = video_info.get(cover_id)
238 if cover:
239 for cover_url in cover['url_list']:
240 thumbnails.append({
241 'id': cover_id,
242 'url': cover_url,
243 })
244
245 stats_info = aweme_detail.get('statistics', {})
246 author_info = aweme_detail.get('author', {})
247 music_info = aweme_detail.get('music', {})
943d5ab1
M		 248 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
249 'sec_uid', 'id', 'uid', 'unique_id',
250 expected_type=str_or_none, get_all=False))
a8549f19 251 labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
bd9ff55b
M		 252
253 contained_music_track = traverse_obj(
254 music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
255 contained_music_author = traverse_obj(
256 music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
257
258 is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
259 if is_generic_og_trackname:
260 music_track, music_author = contained_music_track or 'original sound', contained_music_author
261 else:
262 music_track, music_author = music_info.get('title'), music_info.get('author')
263
264 return {
265 'id': aweme_id,
5fa3c9a8 266 'title': aweme_detail.get('desc'),
a44ca5a4 267 'description': aweme_detail.get('desc'),
bd9ff55b
M		 268 'view_count': int_or_none(stats_info.get('play_count')),
269 'like_count': int_or_none(stats_info.get('digg_count')),
270 'repost_count': int_or_none(stats_info.get('share_count')),
271 'comment_count': int_or_none(stats_info.get('comment_count')),
272 'uploader': str_or_none(author_info.get('unique_id')),
943d5ab1 273 'creator': str_or_none(author_info.get('nickname')),
bd9ff55b 274 'uploader_id': str_or_none(author_info.get('uid')),
943d5ab1 275 'uploader_url': user_url,
bd9ff55b
M		 276 'track': music_track,
277 'album': str_or_none(music_info.get('album')) or None,
278 'artist': music_author,
279 'timestamp': int_or_none(aweme_detail.get('create_time')),
280 'formats': formats,
e0585e65 281 'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
bd9ff55b 282 'thumbnails': thumbnails,
53dad39e
M		 283 'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
284 'availability': self._availability(
285 is_private='Private' in labels,
286 needs_subscription='Friends only' in labels,
287 is_unlisted='Followers only' in labels)
bd9ff55b
M		 288 }
289
0481e266 290 def _parse_aweme_video_web(self, aweme_detail, webpage_url):
943d5ab1 291 video_info = aweme_detail['video']
11aa91a1 292 author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
943d5ab1
M		 293 music_info = aweme_detail.get('music') or {}
294 stats_info = aweme_detail.get('stats') or {}
295 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
296 'secUid', 'id', 'uid', 'uniqueId',
11aa91a1
M		 297 expected_type=str_or_none, get_all=False)
298 or aweme_detail.get('authorSecId'))
943d5ab1
M		 299
300 formats = []
301 play_url = video_info.get('playAddr')
302 width = video_info.get('width')
303 height = video_info.get('height')
304 if isinstance(play_url, str):
305 formats = [{
306 'url': self._proto_relative_url(play_url),
307 'ext': 'mp4',
308 'width': width,
309 'height': height,
310 }]
311 elif isinstance(play_url, list):
312 formats = [{
313 'url': self._proto_relative_url(url),
314 'ext': 'mp4',
315 'width': width,
316 'height': height,
317 } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
318
319 download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
320 if download_url:
321 formats.append({
322 'format_id': 'download',
323 'url': self._proto_relative_url(download_url),
324 'ext': 'mp4',
325 'width': width,
326 'height': height,
327 })
328 self._remove_duplicate_formats(formats)
329 self._sort_formats(formats)
330
331 thumbnails = []
332 for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
333 if aweme_detail.get(thumbnail_name):
334 thumbnails = [{
335 'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
336 'width': width,
337 'height': height
338 }]
339
340 return {
341 'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
342 'title': aweme_detail.get('desc'),
343 'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
344 'view_count': int_or_none(stats_info.get('playCount')),
345 'like_count': int_or_none(stats_info.get('diggCount')),
346 'repost_count': int_or_none(stats_info.get('shareCount')),
347 'comment_count': int_or_none(stats_info.get('commentCount')),
348 'timestamp': int_or_none(aweme_detail.get('createTime')),
349 'creator': str_or_none(author_info.get('nickname')),
11aa91a1
M		 350 'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
351 'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
943d5ab1
M		 352 'uploader_url': user_url,
353 'track': str_or_none(music_info.get('title')),
354 'album': str_or_none(music_info.get('album')) or None,
355 'artist': str_or_none(music_info.get('authorName')),
356 'formats': formats,
357 'thumbnails': thumbnails,
358 'description': str_or_none(aweme_detail.get('desc')),
359 'http_headers': {
0481e266 360 'Referer': webpage_url
943d5ab1
M		 361 }
362 }
363
0fd6661e
M		 364
365class TikTokIE(TikTokBaseIE):
b801cd71 366 _VALID_URL = r'https?://www\.tiktok\.com/(?:embed|@(?P<user_id>[\w\.-]+)/video)/(?P<id>\d+)'
0fd6661e
M		 367
368 _TESTS = [{
369 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
0481e266 370 'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
0fd6661e
M		 371 'info_dict': {
372 'id': '6748451240264420610',
373 'ext': 'mp4',
374 'title': '#jassmanak #lehanga #leenabhushan',
375 'description': '#jassmanak #lehanga #leenabhushan',
376 'duration': 13,
0481e266 377 'height': 1024,
378 'width': 576,
0fd6661e
M		 379 'uploader': 'leenabhushan',
380 'uploader_id': '6691488002098119685',
0481e266 381 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
0fd6661e
M		 382 'creator': 'facestoriesbyleenabh',
383 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
384 'upload_date': '20191016',
385 'timestamp': 1571246252,
386 'view_count': int,
387 'like_count': int,
388 'repost_count': int,
389 'comment_count': int,
a44ca5a4 390 'artist': 'Ysrbeats',
391 'album': 'Lehanga',
392 'track': 'Lehanga',
0fd6661e
M		 393 }
394 }, {
395 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
0481e266 396 'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
0fd6661e
M		 397 'info_dict': {
398 'id': '6742501081818877190',
399 'ext': 'mp4',
400 'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
401 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
402 'duration': 27,
403 'height': 960,
404 'width': 540,
405 'uploader': 'patrox',
406 'uploader_id': '18702747',
0481e266 407 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
0fd6661e
M		 408 'creator': 'patroX',
409 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
410 'upload_date': '20190930',
411 'timestamp': 1569860870,
412 'view_count': int,
413 'like_count': int,
414 'repost_count': int,
415 'comment_count': int,
a44ca5a4 416 'artist': 'Evan Todd, Jessica Keenan Wynn, Alice Lee, Barrett Wilbert Weed & Jon Eidson',
417 'track': 'Big Fun',
0fd6661e
M		 418 }
419 }, {
96f13f01
M		 420 # Banned audio, only available on the app
421 'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
422 'info_dict': {
423 'id': '6984138651336838402',
424 'ext': 'mp4',
425 'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
426 'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
427 'uploader': 'barudakhb_',
428 'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
429 'uploader_id': '6974687867511718913',
430 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
431 'track': 'Boka Dance',
432 'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
433 'timestamp': 1626121503,
434 'duration': 18,
435 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
436 'upload_date': '20210712',
437 'view_count': int,
438 'like_count': int,
439 'repost_count': int,
440 'comment_count': int,
441 }
442 }, {
443 # Sponsored video, only available with feed workaround
444 'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
445 'info_dict': {
446 'id': '7042692929109986561',
447 'ext': 'mp4',
448 'title': 'Slap and Run!',
449 'description': 'Slap and Run!',
450 'uploader': 'user440922249',
451 'creator': 'Slap And Run',
452 'uploader_id': '7036055384943690754',
453 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
454 'track': 'Promoted Music',
455 'timestamp': 1639754738,
456 'duration': 30,
457 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
458 'upload_date': '20211217',
459 'view_count': int,
460 'like_count': int,
461 'repost_count': int,
462 'comment_count': int,
463 },
464 'expected_warnings': ['Video not available']
5fa3c9a8
HTL		 465 }, {
466 # Video without title and description
467 'url': 'https://www.tiktok.com/@pokemonlife22/video/7059698374567611694',
468 'info_dict': {
469 'id': '7059698374567611694',
470 'ext': 'mp4',
b801cd71 471 'title': 'TikTok video #7059698374567611694',
5fa3c9a8
HTL		 472 'description': '',
473 'uploader': 'pokemonlife22',
474 'creator': 'Pokemon',
475 'uploader_id': '6820838815978423302',
476 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
477 'track': 'original sound',
478 'timestamp': 1643714123,
479 'duration': 6,
480 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
481 'upload_date': '20220201',
482 'artist': 'Pokemon',
483 'view_count': int,
484 'like_count': int,
485 'repost_count': int,
486 'comment_count': int,
487 },
a44ca5a4 488 'expected_warnings': ['Video not available', 'Creating a generic title']
e0585e65
M		 489 }, {
490 # Auto-captions available
491 'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
492 'only_matching': True
0fd6661e
M		 493 }]
494
b801cd71 495 @classmethod
496 def _extract_urls(cls, webpage):
497 return [mobj.group('url') for mobj in re.finditer(
498 rf'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{cls._VALID_URL})', webpage)]
499
0fd6661e 500 def _extract_aweme_app(self, aweme_id):
e540c56f
M		 501 try:
502 aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id,
503 note='Downloading video details', errnote='Unable to download video details').get('aweme_detail')
504 if not aweme_detail:
505 raise ExtractorError('Video not available', video_id=aweme_id)
506 except ExtractorError as e:
507 self.report_warning(f'{e}; Retrying with feed workaround')
508 feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
509 note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
be1f331f 510 aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
e540c56f
M		 511 if not aweme_detail:
512 raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
943d5ab1 513 return self._parse_aweme_video_app(aweme_detail)
0fd6661e 514
ce18a19b 515 def _real_extract(self, url):
b801cd71 516 video_id, user_id = self._match_valid_url(url).group('id', 'user_id')
517 url = self._create_url(user_id, video_id)
ce18a19b 518
bd9ff55b
M		 519 try:
520 return self._extract_aweme_app(video_id)
521 except ExtractorError as e:
522 self.report_warning(f'{e}; Retrying with webpage')
523
7bbc0bbc 524 # If we only call once, we get a 403 when downlaoding the video.
61e76c1e 525 self._download_webpage(url, video_id)
6fb11ca8 526 webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
135dfa2c 527 next_data = self._search_nextjs_data(webpage, video_id, default='{}')
11aa91a1 528
135dfa2c 529 if next_data:
11aa91a1
M		 530 status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
531 video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
532 else:
533 sigi_json = self._search_regex(
534 r'>\s*window\[[\'"]SIGI_STATE[\'"]\]\s*=\s*(?P<sigi_state>{.+});',
535 webpage, 'sigi data', group='sigi_state')
536 sigi_data = self._parse_json(sigi_json, video_id)
537 status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
538 video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
539
1418a043 540 if status == 0:
11aa91a1 541 return self._parse_aweme_video_web(video_data, url)
1418a043 542 elif status == 10216:
543 raise ExtractorError('This video is private', expected=True)
6fb11ca8 544 raise ExtractorError('Video not available', video_id=video_id)
f7f18f90
A		 545
546
0fd6661e 547class TikTokUserIE(TikTokBaseIE):
f7f18f90 548 IE_NAME = 'tiktok:user'
0fd6661e 549 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
f7f18f90 550 _TESTS = [{
526d74ec 551 'url': 'https://tiktok.com/@corgibobaa?lang=en',
f7f18f90
A		 552 'playlist_mincount': 45,
553 'info_dict': {
554 'id': '6935371178089399301',
0481e266 555 'title': 'corgibobaa',
b3187433 556 'thumbnail': r're:https://.+_1080x1080\.webp'
f7f18f90 557 },
0481e266 558 'expected_warnings': ['Retrying']
5fa3c9a8
HTL		 559 }, {
560 'url': 'https://www.tiktok.com/@6820838815978423302',
561 'playlist_mincount': 5,
562 'info_dict': {
563 'id': '6820838815978423302',
564 'title': '6820838815978423302',
565 'thumbnail': r're:https://.+_1080x1080\.webp'
566 },
567 'expected_warnings': ['Retrying']
f7f18f90
A		 568 }, {
569 'url': 'https://www.tiktok.com/@meme',
570 'playlist_mincount': 593,
571 'info_dict': {
572 'id': '79005827461758976',
0481e266 573 'title': 'meme',
b3187433 574 'thumbnail': r're:https://.+_1080x1080\.webp'
f7f18f90 575 },
0481e266 576 'expected_warnings': ['Retrying']
f7f18f90
A		 577 }]
578
0fd6661e
M		 579 r''' # TODO: Fix by adding _signature to api_url
580 def _entries(self, webpage, user_id, username):
581 secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
f7f18f90
A		 582 verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
583 if not verifyfp_cookie:
584 raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
585 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='
586 cursor = '0'
587 for page in itertools.count():
0fd6661e 588 data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
f7f18f90
A		 589 for video in data_json.get('itemList', []):
590 video_id = video['id']
591 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
bd9ff55b 592 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
526d74ec 593 if not data_json.get('hasMore'):
f7f18f90
A		 594 break
595 cursor = data_json['cursor']
0fd6661e
M		 596 '''
597
b3187433 598 def _video_entries_api(self, webpage, user_id, username):
0fd6661e
M		 599 query = {
600 'user_id': user_id,
601 'count': 21,
602 'max_cursor': 0,
603 'min_cursor': 0,
604 'retry_type': 'no_retry',
0930b11f 605 '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		 606 }
607
608 max_retries = self.get_param('extractor_retries', 3)
609 for page in itertools.count(1):
610 for retries in itertools.count():
611 try:
612 post_list = self._call_api('aweme/post', query, username,
613 note='Downloading user video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
614 errnote='Unable to download user video list')
615 except ExtractorError as e:
616 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
617 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
618 continue
619 raise
620 break
b3187433 621 yield from post_list.get('aweme_list', [])
0fd6661e
M		 622 if not post_list.get('has_more'):
623 break
624 query['max_cursor'] = post_list['max_cursor']
f7f18f90 625
b3187433 626 def _entries_api(self, user_id, videos):
627 for video in videos:
628 yield {
629 **self._parse_aweme_video_app(video),
630 'extractor_key': TikTokIE.ie_key(),
631 'extractor': 'TikTok',
632 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
633 }
634
f7f18f90 635 def _real_extract(self, url):
0481e266 636 user_name = self._match_id(url)
637 webpage = self._download_webpage(url, user_name, headers={
0fd6661e
M		 638 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
639 })
5fa3c9a8 640 user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID', default=None) or user_name
b3187433 641
642 videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
643 thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
644
645 return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
943d5ab1
M		 646
647
8126298c
M		 648class TikTokBaseListIE(TikTokBaseIE):
649 def _entries(self, list_id, display_id):
650 query = {
651 self._QUERY_NAME: list_id,
652 'cursor': 0,
653 'count': 20,
654 'type': 5,
655 'device_id': ''.join(random.choice(string.digits) for i in range(19))
656 }
657
658 max_retries = self.get_param('extractor_retries', 3)
659 for page in itertools.count(1):
660 for retries in itertools.count():
661 try:
662 post_list = self._call_api(self._API_ENDPOINT, query, display_id,
663 note='Downloading video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
664 errnote='Unable to download video list')
665 except ExtractorError as e:
666 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
667 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
668 continue
669 raise
670 break
671 for video in post_list.get('aweme_list', []):
672 yield {
673 **self._parse_aweme_video_app(video),
0b77924a 674 'extractor_key': TikTokIE.ie_key(),
8126298c
M		 675 'extractor': 'TikTok',
676 'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
677 }
678 if not post_list.get('has_more'):
679 break
680 query['cursor'] = post_list['cursor']
681
682 def _real_extract(self, url):
683 list_id = self._match_id(url)
684 return self.playlist_result(self._entries(list_id, list_id), list_id)
685
686
687class TikTokSoundIE(TikTokBaseListIE):
688 IE_NAME = 'tiktok:sound'
689 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
690 _QUERY_NAME = 'music_id'
691 _API_ENDPOINT = 'music/aweme'
692 _TESTS = [{
693 'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
694 'playlist_mincount': 100,
695 'info_dict': {
696 'id': '6956990112127585029'
697 },
698 'expected_warnings': ['Retrying']
699 }, {
700 # Actual entries are less than listed video count
701 'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
702 'playlist_mincount': 2182,
703 'info_dict': {
704 'id': '7036843036118469381'
705 },
706 'expected_warnings': ['Retrying']
707 }]
708
709
710class TikTokEffectIE(TikTokBaseListIE):
711 IE_NAME = 'tiktok:effect'
712 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
713 _QUERY_NAME = 'sticker_id'
714 _API_ENDPOINT = 'sticker/aweme'
715 _TESTS = [{
716 'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
717 'playlist_mincount': 100,
718 'info_dict': {
719 'id': '1258156',
720 },
721 'expected_warnings': ['Retrying']
722 }, {
723 # Different entries between mobile and web, depending on region
724 'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
725 'only_matching': True
726 }]
727
728
729class TikTokTagIE(TikTokBaseListIE):
730 IE_NAME = 'tiktok:tag'
731 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
732 _QUERY_NAME = 'ch_id'
733 _API_ENDPOINT = 'challenge/aweme'
734 _TESTS = [{
735 'url': 'https://tiktok.com/tag/hello2018',
736 'playlist_mincount': 39,
737 'info_dict': {
738 'id': '46294678',
739 'title': 'hello2018',
740 },
741 'expected_warnings': ['Retrying']
742 }, {
743 'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
744 'only_matching': True
745 }]
746
747 def _real_extract(self, url):
748 display_id = self._match_id(url)
749 webpage = self._download_webpage(url, display_id, headers={
750 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
751 })
752 tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
753 return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
754
755
943d5ab1
M		 756class DouyinIE(TikTokIE):
757 _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
758 _TESTS = [{
759 'url': 'https://www.douyin.com/video/6961737553342991651',
760 'md5': '10523312c8b8100f353620ac9dc8f067',
761 'info_dict': {
762 'id': '6961737553342991651',
763 'ext': 'mp4',
764 'title': '#杨超越 小小水手带你去远航❤️',
765 'uploader': '杨超越',
766 'upload_date': '20210513',
767 'timestamp': 1620905839,
768 'uploader_id': '110403406559',
769 'view_count': int,
770 'like_count': int,
771 'repost_count': int,
772 'comment_count': int,
773 }
774 }, {
775 'url': 'https://www.douyin.com/video/6982497745948921092',
776 'md5': 'd78408c984b9b5102904cf6b6bc2d712',
777 'info_dict': {
778 'id': '6982497745948921092',
779 'ext': 'mp4',
780 'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
781 'uploader': '杨超越工作室',
782 'upload_date': '20210708',
783 'timestamp': 1625739481,
784 'uploader_id': '408654318141572',
785 'view_count': int,
786 'like_count': int,
787 'repost_count': int,
788 'comment_count': int,
789 }
790 }, {
791 'url': 'https://www.douyin.com/video/6953975910773099811',
792 'md5': '72e882e24f75064c218b76c8b713c185',
793 'info_dict': {
794 'id': '6953975910773099811',
795 'ext': 'mp4',
796 'title': '#一起看海 出现在你的夏日里',
797 'uploader': '杨超越',
798 'upload_date': '20210422',
799 'timestamp': 1619098692,
800 'uploader_id': '110403406559',
801 'view_count': int,
802 'like_count': int,
803 'repost_count': int,
804 'comment_count': int,
805 }
806 }, {
807 'url': 'https://www.douyin.com/video/6950251282489675042',
808 'md5': 'b4db86aec367ef810ddd38b1737d2fed',
809 'info_dict': {
810 'id': '6950251282489675042',
811 'ext': 'mp4',
812 'title': '哈哈哈，成功了哈哈哈哈哈哈',
813 'uploader': '杨超越',
814 'upload_date': '20210412',
815 'timestamp': 1618231483,
816 'uploader_id': '110403406559',
817 'view_count': int,
818 'like_count': int,
819 'repost_count': int,
820 'comment_count': int,
821 }
822 }, {
823 'url': 'https://www.douyin.com/video/6963263655114722595',
824 'md5': '1abe1c477d05ee62efb40bf2329957cf',
825 'info_dict': {
826 'id': '6963263655114722595',
827 'ext': 'mp4',
828 'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
829 'uploader': '杨超越',
830 'upload_date': '20210517',
831 'timestamp': 1621261163,
832 'uploader_id': '110403406559',
833 'view_count': int,
834 'like_count': int,
835 'repost_count': int,
836 'comment_count': int,
837 }
838 }]
046cab39 839 _APP_VERSIONS = [('9.6.0', '960')]
943d5ab1
M		 840 _APP_NAME = 'aweme'
841 _AID = 1128
842 _API_HOSTNAME = 'aweme.snssdk.com'
843 _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
53dad39e 844 _WEBPAGE_HOST = 'https://www.douyin.com/'
943d5ab1
M		 845
846 def _real_extract(self, url):
847 video_id = self._match_id(url)
848
849 try:
850 return self._extract_aweme_app(video_id)
851 except ExtractorError as e:
852 self.report_warning(f'{e}; Retrying with webpage')
853
854 webpage = self._download_webpage(url, video_id)
855 render_data_json = self._search_regex(
856 r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
857 webpage, 'render data', default=None)
858 if not render_data_json:
859 # TODO: Run verification challenge code to generate signature cookies
860 raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
861
862 render_data = self._parse_json(
863 render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
ff91cf74 864 return self._parse_aweme_video_web(get_first(render_data, ('aweme', 'detail')), url)
88afe056 865
866
49895f06 867class TikTokVMIE(InfoExtractor):
868 _VALID_URL = r'https?://(?:vm|vt)\.tiktok\.com/(?P<id>\w+)'
88afe056 869 IE_NAME = 'vm.tiktok'
870
49895f06 871 _TESTS = [{
872 'url': 'https://vm.tiktok.com/ZSe4FqkKd',
873 'info_dict': {
874 'id': '7023491746608712966',
875 'ext': 'mp4',
876 'title': 'md5:5607564db90271abbbf8294cca77eddd',
877 'description': 'md5:5607564db90271abbbf8294cca77eddd',
878 'duration': 11,
879 'upload_date': '20211026',
880 'uploader_id': '7007385080558846981',
881 'creator': 'Memes',
882 'artist': 'Memes',
883 'track': 'original sound',
884 'uploader': 'susmandem',
885 'timestamp': 1635284105,
886 'thumbnail': r're:https://.+\.webp.*',
887 'like_count': int,
888 'view_count': int,
889 'comment_count': int,
890 'repost_count': int,
891 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAXcNoOEOxVyBzuII_E--T0MeCrLP0ay1Sm6x_n3dluiWEoWZD0VlQOytwad4W0i0n',
892 }
893 }, {
894 'url': 'https://vt.tiktok.com/ZSe4FqkKd',
895 'only_matching': True,
896 }]
897
88afe056 898 def _real_extract(self, url):
11e1c2e3 899 new_url = self._request_webpage(
900 HEADRequest(url), self._match_id(url), headers={'User-Agent': 'facebookexternalhit/1.1'}).geturl()
901 if self.suitable(new_url): # Prevent infinite loop in case redirect fails
902 raise UnsupportedError(new_url)
903 return self.url_result(new_url)
fork of https://github.com/yt-dlp/yt-dlp/