]> 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] Misc fixes (#2271)
[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,
1ead840d 17 int_or_none,
34921b43 18 join_nonempty,
1ead840d 19 str_or_none,
bd9ff55b
M		 20 traverse_obj,
21 try_get,
943d5ab1 22 url_or_none,
bd9ff55b 23 qualities,
1ead840d
KS		 24)
25
26
0fd6661e 27class TikTokBaseIE(InfoExtractor):
96f13f01 28 _APP_VERSION = '20.1.0'
be1f331f 29 _MANIFEST_APP_VERSION = '210'
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
0fd6661e
M		 37 def _call_api(self, ep, query, video_id, fatal=True,
38 note='Downloading API JSON', errnote='Unable to download API page'):
39 real_query = {
40 **query,
bd9ff55b
M		 41 'version_name': self._APP_VERSION,
42 'version_code': self._MANIFEST_APP_VERSION,
43 'build_number': self._APP_VERSION,
44 'manifest_version_code': self._MANIFEST_APP_VERSION,
45 'update_version_code': self._MANIFEST_APP_VERSION,
0930b11f 46 'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
47 'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
bd9ff55b
M		 48 '_rticket': int(time.time() * 1000),
49 'ts': int(time.time()),
50 'device_brand': 'Google',
51 'device_type': 'Pixel 4',
52 'device_platform': 'android',
53 'resolution': '1080*1920',
54 'dpi': 420,
55 'os_version': '10',
56 'os_api': '29',
57 'carrier_region': 'US',
58 'sys_region': 'US',
59 'region': 'US',
943d5ab1 60 'app_name': self._APP_NAME,
bd9ff55b
M		 61 'app_language': 'en',
62 'language': 'en',
63 'timezone_name': 'America/New_York',
64 'timezone_offset': '-14400',
65 'channel': 'googleplay',
66 'ac': 'wifi',
67 'mcc_mnc': '310260',
68 'is_my_cn': 0,
943d5ab1 69 'aid': self._AID,
bd9ff55b
M		 70 'ssmix': 'a',
71 'as': 'a1qwert123',
72 'cp': 'cbfhckdckkde1',
73 }
0930b11f 74 self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
53dad39e
M		 75 webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
76 if webpage_cookies.get('sid_tt'):
77 self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
0fd6661e 78 return self._download_json(
943d5ab1 79 'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
0fd6661e 80 fatal=fatal, note=note, errnote=errnote, headers={
bd9ff55b 81 '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)',
0fd6661e
M		 82 'Accept': 'application/json',
83 }, query=real_query)
84
943d5ab1 85 def _parse_aweme_video_app(self, aweme_detail):
0fd6661e 86 aweme_id = aweme_detail['aweme_id']
bd9ff55b
M		 87 video_info = aweme_detail['video']
88
89 def parse_url_key(url_key):
90 format_id, codec, res, bitrate = self._search_regex(
91 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
92 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
93 if not format_id:
94 return {}, None
95 return {
96 'format_id': format_id,
97 'vcodec': 'h265' if codec == 'bytevc1' else codec,
98 'tbr': int_or_none(bitrate, scale=1000) or None,
99 'quality': qualities(self.QUALITIES)(res),
100 }, res
101
102 known_resolutions = {}
103
104 def extract_addr(addr, add_meta={}):
105 parsed_meta, res = parse_url_key(addr.get('url_key', ''))
106 if res:
107 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
108 known_resolutions[res].setdefault('width', add_meta.get('width'))
109 parsed_meta.update(known_resolutions.get(res, {}))
110 add_meta.setdefault('height', int_or_none(res[:-1]))
111 return [{
112 'url': url,
113 'filesize': int_or_none(addr.get('data_size')),
114 'ext': 'mp4',
115 'acodec': 'aac',
0fd6661e
M		 116 'source_preference': -2 if 'aweme/v1' in url else -1, # Downloads from API might get blocked
117 **add_meta, **parsed_meta,
34921b43 118 'format_note': join_nonempty(
119 add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
bd9ff55b
M		 120 } for url in addr.get('url_list') or []]
121
122 # Hack: Add direct video links first to prioritize them when removing duplicate formats
123 formats = []
124 if video_info.get('play_addr'):
125 formats.extend(extract_addr(video_info['play_addr'], {
126 'format_id': 'play_addr',
127 'format_note': 'Direct video',
128 'vcodec': 'h265' if traverse_obj(
be1f331f 129 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		 130 'width': video_info.get('width'),
131 'height': video_info.get('height'),
132 }))
133 if video_info.get('download_addr'):
134 formats.extend(extract_addr(video_info['download_addr'], {
135 'format_id': 'download_addr',
136 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
137 'vcodec': 'h264',
138 'width': video_info.get('width'),
139 'height': video_info.get('height'),
0fd6661e 140 'preference': -2 if video_info.get('has_watermark') else -1,
bd9ff55b
M		 141 }))
142 if video_info.get('play_addr_h264'):
143 formats.extend(extract_addr(video_info['play_addr_h264'], {
144 'format_id': 'play_addr_h264',
145 'format_note': 'Direct video',
146 'vcodec': 'h264',
147 }))
148 if video_info.get('play_addr_bytevc1'):
149 formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
150 'format_id': 'play_addr_bytevc1',
151 'format_note': 'Direct video',
152 'vcodec': 'h265',
153 }))
154
155 for bitrate in video_info.get('bit_rate', []):
156 if bitrate.get('play_addr'):
157 formats.extend(extract_addr(bitrate['play_addr'], {
158 'format_id': bitrate.get('gear_name'),
159 'format_note': 'Playback video',
160 'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
161 'vcodec': 'h265' if traverse_obj(
162 bitrate, 'is_bytevc1', 'is_h265') else 'h264',
943d5ab1 163 'fps': bitrate.get('FPS'),
bd9ff55b
M		 164 }))
165
166 self._remove_duplicate_formats(formats)
6134fbeb
M		 167 auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
168 if auth_cookie:
169 for f in formats:
be1f331f 170 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
0fd6661e 171 self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
bd9ff55b
M		 172
173 thumbnails = []
174 for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
175 'origin_cover', 'dynamic_cover'):
176 cover = video_info.get(cover_id)
177 if cover:
178 for cover_url in cover['url_list']:
179 thumbnails.append({
180 'id': cover_id,
181 'url': cover_url,
182 })
183
184 stats_info = aweme_detail.get('statistics', {})
185 author_info = aweme_detail.get('author', {})
186 music_info = aweme_detail.get('music', {})
943d5ab1
M		 187 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
188 'sec_uid', 'id', 'uid', 'unique_id',
189 expected_type=str_or_none, get_all=False))
a8549f19 190 labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
bd9ff55b
M		 191
192 contained_music_track = traverse_obj(
193 music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
194 contained_music_author = traverse_obj(
195 music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
196
197 is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
198 if is_generic_og_trackname:
199 music_track, music_author = contained_music_track or 'original sound', contained_music_author
200 else:
201 music_track, music_author = music_info.get('title'), music_info.get('author')
202
203 return {
204 'id': aweme_id,
205 'title': aweme_detail['desc'],
206 'description': aweme_detail['desc'],
207 'view_count': int_or_none(stats_info.get('play_count')),
208 'like_count': int_or_none(stats_info.get('digg_count')),
209 'repost_count': int_or_none(stats_info.get('share_count')),
210 'comment_count': int_or_none(stats_info.get('comment_count')),
211 'uploader': str_or_none(author_info.get('unique_id')),
943d5ab1 212 'creator': str_or_none(author_info.get('nickname')),
bd9ff55b 213 'uploader_id': str_or_none(author_info.get('uid')),
943d5ab1 214 'uploader_url': user_url,
bd9ff55b
M		 215 'track': music_track,
216 'album': str_or_none(music_info.get('album')) or None,
217 'artist': music_author,
218 'timestamp': int_or_none(aweme_detail.get('create_time')),
219 'formats': formats,
220 'thumbnails': thumbnails,
53dad39e
M		 221 'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
222 'availability': self._availability(
223 is_private='Private' in labels,
224 needs_subscription='Friends only' in labels,
225 is_unlisted='Followers only' in labels)
bd9ff55b
M		 226 }
227
0481e266 228 def _parse_aweme_video_web(self, aweme_detail, webpage_url):
943d5ab1 229 video_info = aweme_detail['video']
11aa91a1 230 author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
943d5ab1
M		 231 music_info = aweme_detail.get('music') or {}
232 stats_info = aweme_detail.get('stats') or {}
233 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
234 'secUid', 'id', 'uid', 'uniqueId',
11aa91a1
M		 235 expected_type=str_or_none, get_all=False)
236 or aweme_detail.get('authorSecId'))
943d5ab1
M		 237
238 formats = []
239 play_url = video_info.get('playAddr')
240 width = video_info.get('width')
241 height = video_info.get('height')
242 if isinstance(play_url, str):
243 formats = [{
244 'url': self._proto_relative_url(play_url),
245 'ext': 'mp4',
246 'width': width,
247 'height': height,
248 }]
249 elif isinstance(play_url, list):
250 formats = [{
251 'url': self._proto_relative_url(url),
252 'ext': 'mp4',
253 'width': width,
254 'height': height,
255 } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
256
257 download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
258 if download_url:
259 formats.append({
260 'format_id': 'download',
261 'url': self._proto_relative_url(download_url),
262 'ext': 'mp4',
263 'width': width,
264 'height': height,
265 })
266 self._remove_duplicate_formats(formats)
267 self._sort_formats(formats)
268
269 thumbnails = []
270 for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
271 if aweme_detail.get(thumbnail_name):
272 thumbnails = [{
273 'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
274 'width': width,
275 'height': height
276 }]
277
278 return {
279 'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
280 'title': aweme_detail.get('desc'),
281 'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
282 'view_count': int_or_none(stats_info.get('playCount')),
283 'like_count': int_or_none(stats_info.get('diggCount')),
284 'repost_count': int_or_none(stats_info.get('shareCount')),
285 'comment_count': int_or_none(stats_info.get('commentCount')),
286 'timestamp': int_or_none(aweme_detail.get('createTime')),
287 'creator': str_or_none(author_info.get('nickname')),
11aa91a1
M		 288 'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
289 'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
943d5ab1
M		 290 'uploader_url': user_url,
291 'track': str_or_none(music_info.get('title')),
292 'album': str_or_none(music_info.get('album')) or None,
293 'artist': str_or_none(music_info.get('authorName')),
294 'formats': formats,
295 'thumbnails': thumbnails,
296 'description': str_or_none(aweme_detail.get('desc')),
297 'http_headers': {
0481e266 298 'Referer': webpage_url
943d5ab1
M		 299 }
300 }
301
0fd6661e
M		 302
303class TikTokIE(TikTokBaseIE):
304 _VALID_URL = r'https?://www\.tiktok\.com/@[\w\.-]+/video/(?P<id>\d+)'
305
306 _TESTS = [{
307 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
0481e266 308 'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
0fd6661e
M		 309 'info_dict': {
310 'id': '6748451240264420610',
311 'ext': 'mp4',
312 'title': '#jassmanak #lehanga #leenabhushan',
313 'description': '#jassmanak #lehanga #leenabhushan',
314 'duration': 13,
0481e266 315 'height': 1024,
316 'width': 576,
0fd6661e
M		 317 'uploader': 'leenabhushan',
318 'uploader_id': '6691488002098119685',
0481e266 319 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
0fd6661e
M		 320 'creator': 'facestoriesbyleenabh',
321 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
322 'upload_date': '20191016',
323 'timestamp': 1571246252,
324 'view_count': int,
325 'like_count': int,
326 'repost_count': int,
327 'comment_count': int,
328 }
329 }, {
330 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
0481e266 331 'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
0fd6661e
M		 332 'info_dict': {
333 'id': '6742501081818877190',
334 'ext': 'mp4',
335 'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
336 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
337 'duration': 27,
338 'height': 960,
339 'width': 540,
340 'uploader': 'patrox',
341 'uploader_id': '18702747',
0481e266 342 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
0fd6661e
M		 343 'creator': 'patroX',
344 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
345 'upload_date': '20190930',
346 'timestamp': 1569860870,
347 'view_count': int,
348 'like_count': int,
349 'repost_count': int,
350 'comment_count': int,
351 }
352 }, {
96f13f01
M		 353 # Banned audio, only available on the app
354 'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
355 'info_dict': {
356 'id': '6984138651336838402',
357 'ext': 'mp4',
358 'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
359 'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
360 'uploader': 'barudakhb_',
361 'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
362 'uploader_id': '6974687867511718913',
363 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
364 'track': 'Boka Dance',
365 'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
366 'timestamp': 1626121503,
367 'duration': 18,
368 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
369 'upload_date': '20210712',
370 'view_count': int,
371 'like_count': int,
372 'repost_count': int,
373 'comment_count': int,
374 }
375 }, {
376 # Sponsored video, only available with feed workaround
377 'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
378 'info_dict': {
379 'id': '7042692929109986561',
380 'ext': 'mp4',
381 'title': 'Slap and Run!',
382 'description': 'Slap and Run!',
383 'uploader': 'user440922249',
384 'creator': 'Slap And Run',
385 'uploader_id': '7036055384943690754',
386 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
387 'track': 'Promoted Music',
388 'timestamp': 1639754738,
389 'duration': 30,
390 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
391 'upload_date': '20211217',
392 'view_count': int,
393 'like_count': int,
394 'repost_count': int,
395 'comment_count': int,
396 },
397 'expected_warnings': ['Video not available']
0fd6661e
M		 398 }]
399
0fd6661e 400 def _extract_aweme_app(self, aweme_id):
e540c56f
M		 401 try:
402 aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id,
403 note='Downloading video details', errnote='Unable to download video details').get('aweme_detail')
404 if not aweme_detail:
405 raise ExtractorError('Video not available', video_id=aweme_id)
406 except ExtractorError as e:
407 self.report_warning(f'{e}; Retrying with feed workaround')
408 feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
409 note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
be1f331f 410 aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
e540c56f
M		 411 if not aweme_detail:
412 raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
943d5ab1 413 return self._parse_aweme_video_app(aweme_detail)
0fd6661e 414
ce18a19b
S		 415 def _real_extract(self, url):
416 video_id = self._match_id(url)
ce18a19b 417
bd9ff55b
M		 418 try:
419 return self._extract_aweme_app(video_id)
420 except ExtractorError as e:
421 self.report_warning(f'{e}; Retrying with webpage')
422
7bbc0bbc 423 # If we only call once, we get a 403 when downlaoding the video.
61e76c1e 424 self._download_webpage(url, video_id)
6fb11ca8 425 webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
11aa91a1
M		 426 next_json = self._search_regex(
427 r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<next_data>[^<]+)',
428 webpage, 'next data', group='next_data', default=None)
429
430 if next_json:
431 next_data = self._parse_json(next_json, video_id)
432 status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
433 video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
434 else:
435 sigi_json = self._search_regex(
436 r'>\s*window\[[\'"]SIGI_STATE[\'"]\]\s*=\s*(?P<sigi_state>{.+});',
437 webpage, 'sigi data', group='sigi_state')
438 sigi_data = self._parse_json(sigi_json, video_id)
439 status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
440 video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
441
1418a043 442 if status == 0:
11aa91a1 443 return self._parse_aweme_video_web(video_data, url)
1418a043 444 elif status == 10216:
445 raise ExtractorError('This video is private', expected=True)
6fb11ca8 446 raise ExtractorError('Video not available', video_id=video_id)
f7f18f90
A		 447
448
0fd6661e 449class TikTokUserIE(TikTokBaseIE):
f7f18f90 450 IE_NAME = 'tiktok:user'
0fd6661e 451 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
f7f18f90 452 _TESTS = [{
526d74ec 453 'url': 'https://tiktok.com/@corgibobaa?lang=en',
f7f18f90
A		 454 'playlist_mincount': 45,
455 'info_dict': {
456 'id': '6935371178089399301',
0481e266 457 'title': 'corgibobaa',
f7f18f90 458 },
0481e266 459 'expected_warnings': ['Retrying']
f7f18f90
A		 460 }, {
461 'url': 'https://www.tiktok.com/@meme',
462 'playlist_mincount': 593,
463 'info_dict': {
464 'id': '79005827461758976',
0481e266 465 'title': 'meme',
f7f18f90 466 },
0481e266 467 'expected_warnings': ['Retrying']
f7f18f90
A		 468 }]
469
0fd6661e
M		 470 r''' # TODO: Fix by adding _signature to api_url
471 def _entries(self, webpage, user_id, username):
472 secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
f7f18f90
A		 473 verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
474 if not verifyfp_cookie:
475 raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
476 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='
477 cursor = '0'
478 for page in itertools.count():
0fd6661e 479 data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
f7f18f90
A		 480 for video in data_json.get('itemList', []):
481 video_id = video['id']
482 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
bd9ff55b 483 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
526d74ec 484 if not data_json.get('hasMore'):
f7f18f90
A		 485 break
486 cursor = data_json['cursor']
0fd6661e
M		 487 '''
488
489 def _entries_api(self, webpage, user_id, username):
490 query = {
491 'user_id': user_id,
492 'count': 21,
493 'max_cursor': 0,
494 'min_cursor': 0,
495 'retry_type': 'no_retry',
0930b11f 496 '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		 497 }
498
499 max_retries = self.get_param('extractor_retries', 3)
500 for page in itertools.count(1):
501 for retries in itertools.count():
502 try:
503 post_list = self._call_api('aweme/post', query, username,
504 note='Downloading user video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
505 errnote='Unable to download user video list')
506 except ExtractorError as e:
507 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
508 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
509 continue
510 raise
511 break
512 for video in post_list.get('aweme_list', []):
513 yield {
0481e266 514 **self._parse_aweme_video_app(video),
0b77924a 515 'extractor_key': TikTokIE.ie_key(),
0fd6661e 516 'extractor': 'TikTok',
b28cdcc0 517 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
0fd6661e
M		 518 }
519 if not post_list.get('has_more'):
520 break
521 query['max_cursor'] = post_list['max_cursor']
f7f18f90
A		 522
523 def _real_extract(self, url):
0481e266 524 user_name = self._match_id(url)
525 webpage = self._download_webpage(url, user_name, headers={
0fd6661e
M		 526 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
527 })
0481e266 528 user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID')
529 return self.playlist_result(self._entries_api(webpage, user_id, user_name), user_id, user_name)
943d5ab1
M		 530
531
8126298c
M		 532class TikTokBaseListIE(TikTokBaseIE):
533 def _entries(self, list_id, display_id):
534 query = {
535 self._QUERY_NAME: list_id,
536 'cursor': 0,
537 'count': 20,
538 'type': 5,
539 'device_id': ''.join(random.choice(string.digits) for i in range(19))
540 }
541
542 max_retries = self.get_param('extractor_retries', 3)
543 for page in itertools.count(1):
544 for retries in itertools.count():
545 try:
546 post_list = self._call_api(self._API_ENDPOINT, query, display_id,
547 note='Downloading video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
548 errnote='Unable to download video list')
549 except ExtractorError as e:
550 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
551 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
552 continue
553 raise
554 break
555 for video in post_list.get('aweme_list', []):
556 yield {
557 **self._parse_aweme_video_app(video),
0b77924a 558 'extractor_key': TikTokIE.ie_key(),
8126298c
M		 559 'extractor': 'TikTok',
560 'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
561 }
562 if not post_list.get('has_more'):
563 break
564 query['cursor'] = post_list['cursor']
565
566 def _real_extract(self, url):
567 list_id = self._match_id(url)
568 return self.playlist_result(self._entries(list_id, list_id), list_id)
569
570
571class TikTokSoundIE(TikTokBaseListIE):
572 IE_NAME = 'tiktok:sound'
573 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
574 _QUERY_NAME = 'music_id'
575 _API_ENDPOINT = 'music/aweme'
576 _TESTS = [{
577 'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
578 'playlist_mincount': 100,
579 'info_dict': {
580 'id': '6956990112127585029'
581 },
582 'expected_warnings': ['Retrying']
583 }, {
584 # Actual entries are less than listed video count
585 'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
586 'playlist_mincount': 2182,
587 'info_dict': {
588 'id': '7036843036118469381'
589 },
590 'expected_warnings': ['Retrying']
591 }]
592
593
594class TikTokEffectIE(TikTokBaseListIE):
595 IE_NAME = 'tiktok:effect'
596 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
597 _QUERY_NAME = 'sticker_id'
598 _API_ENDPOINT = 'sticker/aweme'
599 _TESTS = [{
600 'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
601 'playlist_mincount': 100,
602 'info_dict': {
603 'id': '1258156',
604 },
605 'expected_warnings': ['Retrying']
606 }, {
607 # Different entries between mobile and web, depending on region
608 'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
609 'only_matching': True
610 }]
611
612
613class TikTokTagIE(TikTokBaseListIE):
614 IE_NAME = 'tiktok:tag'
615 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
616 _QUERY_NAME = 'ch_id'
617 _API_ENDPOINT = 'challenge/aweme'
618 _TESTS = [{
619 'url': 'https://tiktok.com/tag/hello2018',
620 'playlist_mincount': 39,
621 'info_dict': {
622 'id': '46294678',
623 'title': 'hello2018',
624 },
625 'expected_warnings': ['Retrying']
626 }, {
627 'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
628 'only_matching': True
629 }]
630
631 def _real_extract(self, url):
632 display_id = self._match_id(url)
633 webpage = self._download_webpage(url, display_id, headers={
634 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
635 })
636 tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
637 return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
638
639
943d5ab1
M		 640class DouyinIE(TikTokIE):
641 _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
642 _TESTS = [{
643 'url': 'https://www.douyin.com/video/6961737553342991651',
644 'md5': '10523312c8b8100f353620ac9dc8f067',
645 'info_dict': {
646 'id': '6961737553342991651',
647 'ext': 'mp4',
648 'title': '#杨超越 小小水手带你去远航❤️',
649 'uploader': '杨超越',
650 'upload_date': '20210513',
651 'timestamp': 1620905839,
652 'uploader_id': '110403406559',
653 'view_count': int,
654 'like_count': int,
655 'repost_count': int,
656 'comment_count': int,
657 }
658 }, {
659 'url': 'https://www.douyin.com/video/6982497745948921092',
660 'md5': 'd78408c984b9b5102904cf6b6bc2d712',
661 'info_dict': {
662 'id': '6982497745948921092',
663 'ext': 'mp4',
664 'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
665 'uploader': '杨超越工作室',
666 'upload_date': '20210708',
667 'timestamp': 1625739481,
668 'uploader_id': '408654318141572',
669 'view_count': int,
670 'like_count': int,
671 'repost_count': int,
672 'comment_count': int,
673 }
674 }, {
675 'url': 'https://www.douyin.com/video/6953975910773099811',
676 'md5': '72e882e24f75064c218b76c8b713c185',
677 'info_dict': {
678 'id': '6953975910773099811',
679 'ext': 'mp4',
680 'title': '#一起看海 出现在你的夏日里',
681 'uploader': '杨超越',
682 'upload_date': '20210422',
683 'timestamp': 1619098692,
684 'uploader_id': '110403406559',
685 'view_count': int,
686 'like_count': int,
687 'repost_count': int,
688 'comment_count': int,
689 }
690 }, {
691 'url': 'https://www.douyin.com/video/6950251282489675042',
692 'md5': 'b4db86aec367ef810ddd38b1737d2fed',
693 'info_dict': {
694 'id': '6950251282489675042',
695 'ext': 'mp4',
696 'title': '哈哈哈，成功了哈哈哈哈哈哈',
697 'uploader': '杨超越',
698 'upload_date': '20210412',
699 'timestamp': 1618231483,
700 'uploader_id': '110403406559',
701 'view_count': int,
702 'like_count': int,
703 'repost_count': int,
704 'comment_count': int,
705 }
706 }, {
707 'url': 'https://www.douyin.com/video/6963263655114722595',
708 'md5': '1abe1c477d05ee62efb40bf2329957cf',
709 'info_dict': {
710 'id': '6963263655114722595',
711 'ext': 'mp4',
712 'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
713 'uploader': '杨超越',
714 'upload_date': '20210517',
715 'timestamp': 1621261163,
716 'uploader_id': '110403406559',
717 'view_count': int,
718 'like_count': int,
719 'repost_count': int,
720 'comment_count': int,
721 }
722 }]
723 _APP_VERSION = '9.6.0'
724 _MANIFEST_APP_VERSION = '960'
725 _APP_NAME = 'aweme'
726 _AID = 1128
727 _API_HOSTNAME = 'aweme.snssdk.com'
728 _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
53dad39e 729 _WEBPAGE_HOST = 'https://www.douyin.com/'
943d5ab1
M		 730
731 def _real_extract(self, url):
732 video_id = self._match_id(url)
733
734 try:
735 return self._extract_aweme_app(video_id)
736 except ExtractorError as e:
737 self.report_warning(f'{e}; Retrying with webpage')
738
739 webpage = self._download_webpage(url, video_id)
740 render_data_json = self._search_regex(
741 r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
742 webpage, 'render data', default=None)
743 if not render_data_json:
744 # TODO: Run verification challenge code to generate signature cookies
745 raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
746
747 render_data = self._parse_json(
748 render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
749 return self._parse_aweme_video_web(
0481e266 750 traverse_obj(render_data, (..., 'aweme', 'detail'), get_all=False), url)
fork of https://github.com/yt-dlp/yt-dlp/