]> 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
[dplay] Add extractors for site changes (#2401)
[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,
b3187433 19 LazyList,
e0585e65 20 srt_subtitles_timecode,
1ead840d 21 str_or_none,
bd9ff55b
M		 22 traverse_obj,
23 try_get,
943d5ab1 24 url_or_none,
bd9ff55b 25 qualities,
1ead840d
KS		 26)
27
28
0fd6661e 29class TikTokBaseIE(InfoExtractor):
96f13f01 30 _APP_VERSION = '20.1.0'
be1f331f 31 _MANIFEST_APP_VERSION = '210'
943d5ab1
M		 32 _APP_NAME = 'trill'
33 _AID = 1180
53dad39e 34 _API_HOSTNAME = 'api-h2.tiktokv.com'
943d5ab1 35 _UPLOADER_URL_FORMAT = 'https://www.tiktok.com/@%s'
53dad39e 36 _WEBPAGE_HOST = 'https://www.tiktok.com/'
be1f331f 37 QUALITIES = ('360p', '540p', '720p', '1080p')
ce18a19b 38
0fd6661e
M		 39 def _call_api(self, ep, query, video_id, fatal=True,
40 note='Downloading API JSON', errnote='Unable to download API page'):
41 real_query = {
42 **query,
bd9ff55b
M		 43 'version_name': self._APP_VERSION,
44 'version_code': self._MANIFEST_APP_VERSION,
45 'build_number': self._APP_VERSION,
46 'manifest_version_code': self._MANIFEST_APP_VERSION,
47 'update_version_code': self._MANIFEST_APP_VERSION,
0930b11f 48 'openudid': ''.join(random.choice('0123456789abcdef') for _ in range(16)),
49 'uuid': ''.join([random.choice(string.digits) for _ in range(16)]),
bd9ff55b
M		 50 '_rticket': int(time.time() * 1000),
51 'ts': int(time.time()),
52 'device_brand': 'Google',
53 'device_type': 'Pixel 4',
54 'device_platform': 'android',
55 'resolution': '1080*1920',
56 'dpi': 420,
57 'os_version': '10',
58 'os_api': '29',
59 'carrier_region': 'US',
60 'sys_region': 'US',
61 'region': 'US',
943d5ab1 62 'app_name': self._APP_NAME,
bd9ff55b
M		 63 'app_language': 'en',
64 'language': 'en',
65 'timezone_name': 'America/New_York',
66 'timezone_offset': '-14400',
67 'channel': 'googleplay',
68 'ac': 'wifi',
69 'mcc_mnc': '310260',
70 'is_my_cn': 0,
943d5ab1 71 'aid': self._AID,
bd9ff55b
M		 72 'ssmix': 'a',
73 'as': 'a1qwert123',
74 'cp': 'cbfhckdckkde1',
75 }
0930b11f 76 self._set_cookie(self._API_HOSTNAME, 'odin_tt', ''.join(random.choice('0123456789abcdef') for _ in range(160)))
53dad39e
M		 77 webpage_cookies = self._get_cookies(self._WEBPAGE_HOST)
78 if webpage_cookies.get('sid_tt'):
79 self._set_cookie(self._API_HOSTNAME, 'sid_tt', webpage_cookies['sid_tt'].value)
0fd6661e 80 return self._download_json(
943d5ab1 81 'https://%s/aweme/v1/%s/' % (self._API_HOSTNAME, ep), video_id=video_id,
0fd6661e 82 fatal=fatal, note=note, errnote=errnote, headers={
bd9ff55b 83 '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		 84 'Accept': 'application/json',
85 }, query=real_query)
86
e0585e65
M		 87 def _get_subtitles(self, aweme_detail, aweme_id):
88 # TODO: Extract text positioning info
89 subtitles = {}
90 captions_info = traverse_obj(
91 aweme_detail, ('interaction_stickers', ..., 'auto_video_caption_info', 'auto_captions', ...), expected_type=dict, default=[])
92 for caption in captions_info:
93 caption_url = traverse_obj(caption, ('url', 'url_list', ...), expected_type=url_or_none, get_all=False)
94 if not caption_url:
95 continue
96 caption_json = self._download_json(
97 caption_url, aweme_id, note='Downloading captions', errnote='Unable to download captions', fatal=False)
98 if not caption_json:
99 continue
100 subtitles.setdefault(caption.get('language', 'en'), []).append({
101 'ext': 'srt',
102 'data': '\n\n'.join(
103 f'{i + 1}\n{srt_subtitles_timecode(line["start_time"] / 1000)} --> {srt_subtitles_timecode(line["end_time"] / 1000)}\n{line["text"]}'
104 for i, line in enumerate(caption_json['utterances']) if line.get('text'))
105 })
106 return subtitles
107
943d5ab1 108 def _parse_aweme_video_app(self, aweme_detail):
0fd6661e 109 aweme_id = aweme_detail['aweme_id']
bd9ff55b
M		 110 video_info = aweme_detail['video']
111
112 def parse_url_key(url_key):
113 format_id, codec, res, bitrate = self._search_regex(
114 r'v[^_]+_(?P<id>(?P<codec>[^_]+)_(?P<res>\d+p)_(?P<bitrate>\d+))', url_key,
115 'url key', default=(None, None, None, None), group=('id', 'codec', 'res', 'bitrate'))
116 if not format_id:
117 return {}, None
118 return {
119 'format_id': format_id,
120 'vcodec': 'h265' if codec == 'bytevc1' else codec,
121 'tbr': int_or_none(bitrate, scale=1000) or None,
122 'quality': qualities(self.QUALITIES)(res),
123 }, res
124
125 known_resolutions = {}
126
127 def extract_addr(addr, add_meta={}):
128 parsed_meta, res = parse_url_key(addr.get('url_key', ''))
129 if res:
130 known_resolutions.setdefault(res, {}).setdefault('height', add_meta.get('height'))
131 known_resolutions[res].setdefault('width', add_meta.get('width'))
132 parsed_meta.update(known_resolutions.get(res, {}))
133 add_meta.setdefault('height', int_or_none(res[:-1]))
134 return [{
135 'url': url,
136 'filesize': int_or_none(addr.get('data_size')),
137 'ext': 'mp4',
138 'acodec': 'aac',
0fd6661e
M		 139 'source_preference': -2 if 'aweme/v1' in url else -1, # Downloads from API might get blocked
140 **add_meta, **parsed_meta,
34921b43 141 'format_note': join_nonempty(
142 add_meta.get('format_note'), '(API)' if 'aweme/v1' in url else None, delim=' ')
bd9ff55b
M		 143 } for url in addr.get('url_list') or []]
144
145 # Hack: Add direct video links first to prioritize them when removing duplicate formats
146 formats = []
147 if video_info.get('play_addr'):
148 formats.extend(extract_addr(video_info['play_addr'], {
149 'format_id': 'play_addr',
150 'format_note': 'Direct video',
151 'vcodec': 'h265' if traverse_obj(
be1f331f 152 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		 153 'width': video_info.get('width'),
154 'height': video_info.get('height'),
155 }))
156 if video_info.get('download_addr'):
157 formats.extend(extract_addr(video_info['download_addr'], {
158 'format_id': 'download_addr',
159 'format_note': 'Download video%s' % (', watermarked' if video_info.get('has_watermark') else ''),
160 'vcodec': 'h264',
161 'width': video_info.get('width'),
162 'height': video_info.get('height'),
0fd6661e 163 'preference': -2 if video_info.get('has_watermark') else -1,
bd9ff55b
M		 164 }))
165 if video_info.get('play_addr_h264'):
166 formats.extend(extract_addr(video_info['play_addr_h264'], {
167 'format_id': 'play_addr_h264',
168 'format_note': 'Direct video',
169 'vcodec': 'h264',
170 }))
171 if video_info.get('play_addr_bytevc1'):
172 formats.extend(extract_addr(video_info['play_addr_bytevc1'], {
173 'format_id': 'play_addr_bytevc1',
174 'format_note': 'Direct video',
175 'vcodec': 'h265',
176 }))
177
178 for bitrate in video_info.get('bit_rate', []):
179 if bitrate.get('play_addr'):
180 formats.extend(extract_addr(bitrate['play_addr'], {
181 'format_id': bitrate.get('gear_name'),
182 'format_note': 'Playback video',
183 'tbr': try_get(bitrate, lambda x: x['bit_rate'] / 1000),
184 'vcodec': 'h265' if traverse_obj(
185 bitrate, 'is_bytevc1', 'is_h265') else 'h264',
943d5ab1 186 'fps': bitrate.get('FPS'),
bd9ff55b
M		 187 }))
188
189 self._remove_duplicate_formats(formats)
6134fbeb
M		 190 auth_cookie = self._get_cookies(self._WEBPAGE_HOST).get('sid_tt')
191 if auth_cookie:
192 for f in formats:
be1f331f 193 self._set_cookie(compat_urllib_parse_urlparse(f['url']).hostname, 'sid_tt', auth_cookie.value)
0fd6661e 194 self._sort_formats(formats, ('quality', 'codec', 'size', 'br'))
bd9ff55b
M		 195
196 thumbnails = []
197 for cover_id in ('cover', 'ai_dynamic_cover', 'animated_cover', 'ai_dynamic_cover_bak',
198 'origin_cover', 'dynamic_cover'):
199 cover = video_info.get(cover_id)
200 if cover:
201 for cover_url in cover['url_list']:
202 thumbnails.append({
203 'id': cover_id,
204 'url': cover_url,
205 })
206
207 stats_info = aweme_detail.get('statistics', {})
208 author_info = aweme_detail.get('author', {})
209 music_info = aweme_detail.get('music', {})
943d5ab1
M		 210 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
211 'sec_uid', 'id', 'uid', 'unique_id',
212 expected_type=str_or_none, get_all=False))
a8549f19 213 labels = traverse_obj(aweme_detail, ('hybrid_label', ..., 'text'), expected_type=str, default=[])
bd9ff55b
M		 214
215 contained_music_track = traverse_obj(
216 music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
217 contained_music_author = traverse_obj(
218 music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
219
220 is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
221 if is_generic_og_trackname:
222 music_track, music_author = contained_music_track or 'original sound', contained_music_author
223 else:
224 music_track, music_author = music_info.get('title'), music_info.get('author')
225
226 return {
227 'id': aweme_id,
228 'title': aweme_detail['desc'],
229 'description': aweme_detail['desc'],
230 'view_count': int_or_none(stats_info.get('play_count')),
231 'like_count': int_or_none(stats_info.get('digg_count')),
232 'repost_count': int_or_none(stats_info.get('share_count')),
233 'comment_count': int_or_none(stats_info.get('comment_count')),
234 'uploader': str_or_none(author_info.get('unique_id')),
943d5ab1 235 'creator': str_or_none(author_info.get('nickname')),
bd9ff55b 236 'uploader_id': str_or_none(author_info.get('uid')),
943d5ab1 237 'uploader_url': user_url,
bd9ff55b
M		 238 'track': music_track,
239 'album': str_or_none(music_info.get('album')) or None,
240 'artist': music_author,
241 'timestamp': int_or_none(aweme_detail.get('create_time')),
242 'formats': formats,
e0585e65 243 'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
bd9ff55b 244 'thumbnails': thumbnails,
53dad39e
M		 245 'duration': int_or_none(traverse_obj(video_info, 'duration', ('download_addr', 'duration')), scale=1000),
246 'availability': self._availability(
247 is_private='Private' in labels,
248 needs_subscription='Friends only' in labels,
249 is_unlisted='Followers only' in labels)
bd9ff55b
M		 250 }
251
0481e266 252 def _parse_aweme_video_web(self, aweme_detail, webpage_url):
943d5ab1 253 video_info = aweme_detail['video']
11aa91a1 254 author_info = traverse_obj(aweme_detail, 'authorInfo', 'author', expected_type=dict, default={})
943d5ab1
M		 255 music_info = aweme_detail.get('music') or {}
256 stats_info = aweme_detail.get('stats') or {}
257 user_url = self._UPLOADER_URL_FORMAT % (traverse_obj(author_info,
258 'secUid', 'id', 'uid', 'uniqueId',
11aa91a1
M		 259 expected_type=str_or_none, get_all=False)
260 or aweme_detail.get('authorSecId'))
943d5ab1
M		 261
262 formats = []
263 play_url = video_info.get('playAddr')
264 width = video_info.get('width')
265 height = video_info.get('height')
266 if isinstance(play_url, str):
267 formats = [{
268 'url': self._proto_relative_url(play_url),
269 'ext': 'mp4',
270 'width': width,
271 'height': height,
272 }]
273 elif isinstance(play_url, list):
274 formats = [{
275 'url': self._proto_relative_url(url),
276 'ext': 'mp4',
277 'width': width,
278 'height': height,
279 } for url in traverse_obj(play_url, (..., 'src'), expected_type=url_or_none, default=[]) if url]
280
281 download_url = url_or_none(video_info.get('downloadAddr')) or traverse_obj(video_info, ('download', 'url'), expected_type=url_or_none)
282 if download_url:
283 formats.append({
284 'format_id': 'download',
285 'url': self._proto_relative_url(download_url),
286 'ext': 'mp4',
287 'width': width,
288 'height': height,
289 })
290 self._remove_duplicate_formats(formats)
291 self._sort_formats(formats)
292
293 thumbnails = []
294 for thumbnail_name in ('thumbnail', 'cover', 'dynamicCover', 'originCover'):
295 if aweme_detail.get(thumbnail_name):
296 thumbnails = [{
297 'url': self._proto_relative_url(aweme_detail[thumbnail_name]),
298 'width': width,
299 'height': height
300 }]
301
302 return {
303 'id': traverse_obj(aweme_detail, 'id', 'awemeId', expected_type=str_or_none),
304 'title': aweme_detail.get('desc'),
305 'duration': try_get(aweme_detail, lambda x: x['video']['duration'], int),
306 'view_count': int_or_none(stats_info.get('playCount')),
307 'like_count': int_or_none(stats_info.get('diggCount')),
308 'repost_count': int_or_none(stats_info.get('shareCount')),
309 'comment_count': int_or_none(stats_info.get('commentCount')),
310 'timestamp': int_or_none(aweme_detail.get('createTime')),
311 'creator': str_or_none(author_info.get('nickname')),
11aa91a1
M		 312 'uploader': str_or_none(author_info.get('uniqueId') or aweme_detail.get('author')),
313 'uploader_id': str_or_none(author_info.get('id') or aweme_detail.get('authorId')),
943d5ab1
M		 314 'uploader_url': user_url,
315 'track': str_or_none(music_info.get('title')),
316 'album': str_or_none(music_info.get('album')) or None,
317 'artist': str_or_none(music_info.get('authorName')),
318 'formats': formats,
319 'thumbnails': thumbnails,
320 'description': str_or_none(aweme_detail.get('desc')),
321 'http_headers': {
0481e266 322 'Referer': webpage_url
943d5ab1
M		 323 }
324 }
325
0fd6661e
M		 326
327class TikTokIE(TikTokBaseIE):
328 _VALID_URL = r'https?://www\.tiktok\.com/@[\w\.-]+/video/(?P<id>\d+)'
329
330 _TESTS = [{
331 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610',
0481e266 332 'md5': '736bb7a466c6f0a6afeb597da1e6f5b7',
0fd6661e
M		 333 'info_dict': {
334 'id': '6748451240264420610',
335 'ext': 'mp4',
336 'title': '#jassmanak #lehanga #leenabhushan',
337 'description': '#jassmanak #lehanga #leenabhushan',
338 'duration': 13,
0481e266 339 'height': 1024,
340 'width': 576,
0fd6661e
M		 341 'uploader': 'leenabhushan',
342 'uploader_id': '6691488002098119685',
0481e266 343 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA_Eb4t1vodM1IuTy_cvp9CY22RAb59xqrO0Xtz9CYQJvgXaDvZxYnZYRzDWhhgJmy',
0fd6661e
M		 344 'creator': 'facestoriesbyleenabh',
345 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
346 'upload_date': '20191016',
347 'timestamp': 1571246252,
348 'view_count': int,
349 'like_count': int,
350 'repost_count': int,
351 'comment_count': int,
352 }
353 }, {
354 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en',
0481e266 355 'md5': '6f3cf8cdd9b28cb8363fe0a9a160695b',
0fd6661e
M		 356 'info_dict': {
357 'id': '6742501081818877190',
358 'ext': 'mp4',
359 'title': 'md5:5e2a23877420bb85ce6521dbee39ba94',
360 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94',
361 'duration': 27,
362 'height': 960,
363 'width': 540,
364 'uploader': 'patrox',
365 'uploader_id': '18702747',
0481e266 366 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAiFnldaILebi5heDoVU6bn4jBWWycX6-9U3xuNPqZ8Ws',
0fd6661e
M		 367 'creator': 'patroX',
368 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
369 'upload_date': '20190930',
370 'timestamp': 1569860870,
371 'view_count': int,
372 'like_count': int,
373 'repost_count': int,
374 'comment_count': int,
375 }
376 }, {
96f13f01
M		 377 # Banned audio, only available on the app
378 'url': 'https://www.tiktok.com/@barudakhb_/video/6984138651336838402',
379 'info_dict': {
380 'id': '6984138651336838402',
381 'ext': 'mp4',
382 'title': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
383 'description': 'Balas @yolaaftwsr hayu yu ? #SquadRandom_ 🔥',
384 'uploader': 'barudakhb_',
385 'creator': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
386 'uploader_id': '6974687867511718913',
387 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
388 'track': 'Boka Dance',
389 'artist': 'md5:29f238c49bc0c176cb3cef1a9cea9fa6',
390 'timestamp': 1626121503,
391 'duration': 18,
392 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
393 'upload_date': '20210712',
394 'view_count': int,
395 'like_count': int,
396 'repost_count': int,
397 'comment_count': int,
398 }
399 }, {
400 # Sponsored video, only available with feed workaround
401 'url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_/video/7042692929109986561',
402 'info_dict': {
403 'id': '7042692929109986561',
404 'ext': 'mp4',
405 'title': 'Slap and Run!',
406 'description': 'Slap and Run!',
407 'uploader': 'user440922249',
408 'creator': 'Slap And Run',
409 'uploader_id': '7036055384943690754',
410 'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
411 'track': 'Promoted Music',
412 'timestamp': 1639754738,
413 'duration': 30,
414 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
415 'upload_date': '20211217',
416 'view_count': int,
417 'like_count': int,
418 'repost_count': int,
419 'comment_count': int,
420 },
421 'expected_warnings': ['Video not available']
e0585e65
M		 422 }, {
423 # Auto-captions available
424 'url': 'https://www.tiktok.com/@hankgreen1/video/7047596209028074758',
425 'only_matching': True
0fd6661e
M		 426 }]
427
0fd6661e 428 def _extract_aweme_app(self, aweme_id):
e540c56f
M		 429 try:
430 aweme_detail = self._call_api('aweme/detail', {'aweme_id': aweme_id}, aweme_id,
431 note='Downloading video details', errnote='Unable to download video details').get('aweme_detail')
432 if not aweme_detail:
433 raise ExtractorError('Video not available', video_id=aweme_id)
434 except ExtractorError as e:
435 self.report_warning(f'{e}; Retrying with feed workaround')
436 feed_list = self._call_api('feed', {'aweme_id': aweme_id}, aweme_id,
437 note='Downloading video feed', errnote='Unable to download video feed').get('aweme_list') or []
be1f331f 438 aweme_detail = next((aweme for aweme in feed_list if str(aweme.get('aweme_id')) == aweme_id), None)
e540c56f
M		 439 if not aweme_detail:
440 raise ExtractorError('Unable to find video in feed', video_id=aweme_id)
943d5ab1 441 return self._parse_aweme_video_app(aweme_detail)
0fd6661e 442
ce18a19b
S		 443 def _real_extract(self, url):
444 video_id = self._match_id(url)
ce18a19b 445
bd9ff55b
M		 446 try:
447 return self._extract_aweme_app(video_id)
448 except ExtractorError as e:
449 self.report_warning(f'{e}; Retrying with webpage')
450
7bbc0bbc 451 # If we only call once, we get a 403 when downlaoding the video.
61e76c1e 452 self._download_webpage(url, video_id)
6fb11ca8 453 webpage = self._download_webpage(url, video_id, note='Downloading video webpage')
135dfa2c 454 next_data = self._search_nextjs_data(webpage, video_id, default='{}')
11aa91a1 455
135dfa2c 456 if next_data:
11aa91a1
M		 457 status = traverse_obj(next_data, ('props', 'pageProps', 'statusCode'), expected_type=int) or 0
458 video_data = traverse_obj(next_data, ('props', 'pageProps', 'itemInfo', 'itemStruct'), expected_type=dict)
459 else:
460 sigi_json = self._search_regex(
461 r'>\s*window\[[\'"]SIGI_STATE[\'"]\]\s*=\s*(?P<sigi_state>{.+});',
462 webpage, 'sigi data', group='sigi_state')
463 sigi_data = self._parse_json(sigi_json, video_id)
464 status = traverse_obj(sigi_data, ('VideoPage', 'statusCode'), expected_type=int) or 0
465 video_data = traverse_obj(sigi_data, ('ItemModule', video_id), expected_type=dict)
466
1418a043 467 if status == 0:
11aa91a1 468 return self._parse_aweme_video_web(video_data, url)
1418a043 469 elif status == 10216:
470 raise ExtractorError('This video is private', expected=True)
6fb11ca8 471 raise ExtractorError('Video not available', video_id=video_id)
f7f18f90
A		 472
473
0fd6661e 474class TikTokUserIE(TikTokBaseIE):
f7f18f90 475 IE_NAME = 'tiktok:user'
0fd6661e 476 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/@(?P<id>[\w\.-]+)/?(?:$|[#?])'
f7f18f90 477 _TESTS = [{
526d74ec 478 'url': 'https://tiktok.com/@corgibobaa?lang=en',
f7f18f90
A		 479 'playlist_mincount': 45,
480 'info_dict': {
481 'id': '6935371178089399301',
0481e266 482 'title': 'corgibobaa',
b3187433 483 'thumbnail': r're:https://.+_1080x1080\.webp'
f7f18f90 484 },
0481e266 485 'expected_warnings': ['Retrying']
f7f18f90
A		 486 }, {
487 'url': 'https://www.tiktok.com/@meme',
488 'playlist_mincount': 593,
489 'info_dict': {
490 'id': '79005827461758976',
0481e266 491 'title': 'meme',
b3187433 492 'thumbnail': r're:https://.+_1080x1080\.webp'
f7f18f90 493 },
0481e266 494 'expected_warnings': ['Retrying']
f7f18f90
A		 495 }]
496
0fd6661e
M		 497 r''' # TODO: Fix by adding _signature to api_url
498 def _entries(self, webpage, user_id, username):
499 secuid = self._search_regex(r'\"secUid\":\"(?P<secUid>[^\"]+)', webpage, username)
f7f18f90
A		 500 verifyfp_cookie = self._get_cookies('https://www.tiktok.com').get('s_v_web_id')
501 if not verifyfp_cookie:
502 raise ExtractorError('Improper cookies (missing s_v_web_id).', expected=True)
503 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='
504 cursor = '0'
505 for page in itertools.count():
0fd6661e 506 data_json = self._download_json(api_url + cursor, username, note='Downloading Page %d' % page)
f7f18f90
A		 507 for video in data_json.get('itemList', []):
508 video_id = video['id']
509 video_url = f'https://www.tiktok.com/@{user_id}/video/{video_id}'
bd9ff55b 510 yield self._url_result(video_url, 'TikTok', video_id, str_or_none(video.get('desc')))
526d74ec 511 if not data_json.get('hasMore'):
f7f18f90
A		 512 break
513 cursor = data_json['cursor']
0fd6661e
M		 514 '''
515
b3187433 516 def _video_entries_api(self, webpage, user_id, username):
0fd6661e
M		 517 query = {
518 'user_id': user_id,
519 'count': 21,
520 'max_cursor': 0,
521 'min_cursor': 0,
522 'retry_type': 'no_retry',
0930b11f 523 '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		 524 }
525
526 max_retries = self.get_param('extractor_retries', 3)
527 for page in itertools.count(1):
528 for retries in itertools.count():
529 try:
530 post_list = self._call_api('aweme/post', query, username,
531 note='Downloading user video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
532 errnote='Unable to download user video list')
533 except ExtractorError as e:
534 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
535 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
536 continue
537 raise
538 break
b3187433 539 yield from post_list.get('aweme_list', [])
0fd6661e
M		 540 if not post_list.get('has_more'):
541 break
542 query['max_cursor'] = post_list['max_cursor']
f7f18f90 543
b3187433 544 def _entries_api(self, user_id, videos):
545 for video in videos:
546 yield {
547 **self._parse_aweme_video_app(video),
548 'extractor_key': TikTokIE.ie_key(),
549 'extractor': 'TikTok',
550 'webpage_url': f'https://tiktok.com/@{user_id}/video/{video["aweme_id"]}',
551 }
552
f7f18f90 553 def _real_extract(self, url):
0481e266 554 user_name = self._match_id(url)
555 webpage = self._download_webpage(url, user_name, headers={
0fd6661e
M		 556 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
557 })
0481e266 558 user_id = self._html_search_regex(r'snssdk\d*://user/profile/(\d+)', webpage, 'user ID')
b3187433 559
560 videos = LazyList(self._video_entries_api(webpage, user_id, user_name))
561 thumbnail = traverse_obj(videos, (0, 'author', 'avatar_larger', 'url_list', 0))
562
563 return self.playlist_result(self._entries_api(user_id, videos), user_id, user_name, thumbnail=thumbnail)
943d5ab1
M		 564
565
8126298c
M		 566class TikTokBaseListIE(TikTokBaseIE):
567 def _entries(self, list_id, display_id):
568 query = {
569 self._QUERY_NAME: list_id,
570 'cursor': 0,
571 'count': 20,
572 'type': 5,
573 'device_id': ''.join(random.choice(string.digits) for i in range(19))
574 }
575
576 max_retries = self.get_param('extractor_retries', 3)
577 for page in itertools.count(1):
578 for retries in itertools.count():
579 try:
580 post_list = self._call_api(self._API_ENDPOINT, query, display_id,
581 note='Downloading video list page %d%s' % (page, f' (attempt {retries})' if retries != 0 else ''),
582 errnote='Unable to download video list')
583 except ExtractorError as e:
584 if isinstance(e.cause, json.JSONDecodeError) and e.cause.pos == 0 and retries != max_retries:
585 self.report_warning('%s. Retrying...' % str(e.cause or e.msg))
586 continue
587 raise
588 break
589 for video in post_list.get('aweme_list', []):
590 yield {
591 **self._parse_aweme_video_app(video),
0b77924a 592 'extractor_key': TikTokIE.ie_key(),
8126298c
M		 593 'extractor': 'TikTok',
594 'webpage_url': f'https://tiktok.com/@_/video/{video["aweme_id"]}',
595 }
596 if not post_list.get('has_more'):
597 break
598 query['cursor'] = post_list['cursor']
599
600 def _real_extract(self, url):
601 list_id = self._match_id(url)
602 return self.playlist_result(self._entries(list_id, list_id), list_id)
603
604
605class TikTokSoundIE(TikTokBaseListIE):
606 IE_NAME = 'tiktok:sound'
607 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/music/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
608 _QUERY_NAME = 'music_id'
609 _API_ENDPOINT = 'music/aweme'
610 _TESTS = [{
611 'url': 'https://www.tiktok.com/music/Build-a-Btch-6956990112127585029?lang=en',
612 'playlist_mincount': 100,
613 'info_dict': {
614 'id': '6956990112127585029'
615 },
616 'expected_warnings': ['Retrying']
617 }, {
618 # Actual entries are less than listed video count
619 'url': 'https://www.tiktok.com/music/jiefei-soap-remix-7036843036118469381',
620 'playlist_mincount': 2182,
621 'info_dict': {
622 'id': '7036843036118469381'
623 },
624 'expected_warnings': ['Retrying']
625 }]
626
627
628class TikTokEffectIE(TikTokBaseListIE):
629 IE_NAME = 'tiktok:effect'
630 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/sticker/[\w\.-]+-(?P<id>[\d]+)[/?#&]?'
631 _QUERY_NAME = 'sticker_id'
632 _API_ENDPOINT = 'sticker/aweme'
633 _TESTS = [{
634 'url': 'https://www.tiktok.com/sticker/MATERIAL-GWOOORL-1258156',
635 'playlist_mincount': 100,
636 'info_dict': {
637 'id': '1258156',
638 },
639 'expected_warnings': ['Retrying']
640 }, {
641 # Different entries between mobile and web, depending on region
642 'url': 'https://www.tiktok.com/sticker/Elf-Friend-479565',
643 'only_matching': True
644 }]
645
646
647class TikTokTagIE(TikTokBaseListIE):
648 IE_NAME = 'tiktok:tag'
649 _VALID_URL = r'https?://(?:www\.)?tiktok\.com/tag/(?P<id>[^/?#&]+)'
650 _QUERY_NAME = 'ch_id'
651 _API_ENDPOINT = 'challenge/aweme'
652 _TESTS = [{
653 'url': 'https://tiktok.com/tag/hello2018',
654 'playlist_mincount': 39,
655 'info_dict': {
656 'id': '46294678',
657 'title': 'hello2018',
658 },
659 'expected_warnings': ['Retrying']
660 }, {
661 'url': 'https://tiktok.com/tag/fypシ?is_copy_url=0&is_from_webapp=v1',
662 'only_matching': True
663 }]
664
665 def _real_extract(self, url):
666 display_id = self._match_id(url)
667 webpage = self._download_webpage(url, display_id, headers={
668 'User-Agent': 'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)'
669 })
670 tag_id = self._html_search_regex(r'snssdk\d*://challenge/detail/(\d+)', webpage, 'tag ID')
671 return self.playlist_result(self._entries(tag_id, display_id), tag_id, display_id)
672
673
943d5ab1
M		 674class DouyinIE(TikTokIE):
675 _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
676 _TESTS = [{
677 'url': 'https://www.douyin.com/video/6961737553342991651',
678 'md5': '10523312c8b8100f353620ac9dc8f067',
679 'info_dict': {
680 'id': '6961737553342991651',
681 'ext': 'mp4',
682 'title': '#杨超越 小小水手带你去远航❤️',
683 'uploader': '杨超越',
684 'upload_date': '20210513',
685 'timestamp': 1620905839,
686 'uploader_id': '110403406559',
687 'view_count': int,
688 'like_count': int,
689 'repost_count': int,
690 'comment_count': int,
691 }
692 }, {
693 'url': 'https://www.douyin.com/video/6982497745948921092',
694 'md5': 'd78408c984b9b5102904cf6b6bc2d712',
695 'info_dict': {
696 'id': '6982497745948921092',
697 'ext': 'mp4',
698 'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
699 'uploader': '杨超越工作室',
700 'upload_date': '20210708',
701 'timestamp': 1625739481,
702 'uploader_id': '408654318141572',
703 'view_count': int,
704 'like_count': int,
705 'repost_count': int,
706 'comment_count': int,
707 }
708 }, {
709 'url': 'https://www.douyin.com/video/6953975910773099811',
710 'md5': '72e882e24f75064c218b76c8b713c185',
711 'info_dict': {
712 'id': '6953975910773099811',
713 'ext': 'mp4',
714 'title': '#一起看海 出现在你的夏日里',
715 'uploader': '杨超越',
716 'upload_date': '20210422',
717 'timestamp': 1619098692,
718 'uploader_id': '110403406559',
719 'view_count': int,
720 'like_count': int,
721 'repost_count': int,
722 'comment_count': int,
723 }
724 }, {
725 'url': 'https://www.douyin.com/video/6950251282489675042',
726 'md5': 'b4db86aec367ef810ddd38b1737d2fed',
727 'info_dict': {
728 'id': '6950251282489675042',
729 'ext': 'mp4',
730 'title': '哈哈哈，成功了哈哈哈哈哈哈',
731 'uploader': '杨超越',
732 'upload_date': '20210412',
733 'timestamp': 1618231483,
734 'uploader_id': '110403406559',
735 'view_count': int,
736 'like_count': int,
737 'repost_count': int,
738 'comment_count': int,
739 }
740 }, {
741 'url': 'https://www.douyin.com/video/6963263655114722595',
742 'md5': '1abe1c477d05ee62efb40bf2329957cf',
743 'info_dict': {
744 'id': '6963263655114722595',
745 'ext': 'mp4',
746 'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
747 'uploader': '杨超越',
748 'upload_date': '20210517',
749 'timestamp': 1621261163,
750 'uploader_id': '110403406559',
751 'view_count': int,
752 'like_count': int,
753 'repost_count': int,
754 'comment_count': int,
755 }
756 }]
757 _APP_VERSION = '9.6.0'
758 _MANIFEST_APP_VERSION = '960'
759 _APP_NAME = 'aweme'
760 _AID = 1128
761 _API_HOSTNAME = 'aweme.snssdk.com'
762 _UPLOADER_URL_FORMAT = 'https://www.douyin.com/user/%s'
53dad39e 763 _WEBPAGE_HOST = 'https://www.douyin.com/'
943d5ab1
M		 764
765 def _real_extract(self, url):
766 video_id = self._match_id(url)
767
768 try:
769 return self._extract_aweme_app(video_id)
770 except ExtractorError as e:
771 self.report_warning(f'{e}; Retrying with webpage')
772
773 webpage = self._download_webpage(url, video_id)
774 render_data_json = self._search_regex(
775 r'<script [^>]*\bid=[\'"]RENDER_DATA[\'"][^>]*>(%7B.+%7D)</script>',
776 webpage, 'render data', default=None)
777 if not render_data_json:
778 # TODO: Run verification challenge code to generate signature cookies
779 raise ExtractorError('Fresh cookies (not necessarily logged in) are needed')
780
781 render_data = self._parse_json(
782 render_data_json, video_id, transform_source=compat_urllib_parse_unquote)
783 return self._parse_aweme_video_web(
0481e266 784 traverse_obj(render_data, (..., 'aweme', 'detail'), get_all=False), url)
fork of https://github.com/yt-dlp/yt-dlp/