diff --git a/youtube_dlc/extractor/extractors.py b/youtube_dlc/extractor/extractors.py index e70e779ffc..af1bc6e31d 100644 --- a/youtube_dlc/extractor/extractors.py +++ b/youtube_dlc/extractor/extractors.py @@ -1165,10 +1165,7 @@ from .thisamericanlife import ThisAmericanLifeIE from .thisav import ThisAVIE from .thisoldhouse import ThisOldHouseIE from .threeqsdn import ThreeQSDNIE -from .tiktok import ( - TikTokIE, - TikTokUserIE, -) +from .tiktok import TikTokIE from .tinypic import TinyPicIE from .tmz import ( TMZIE, diff --git a/youtube_dlc/extractor/tiktok.py b/youtube_dlc/extractor/tiktok.py index 66088b9abd..075a2cdf93 100644 --- a/youtube_dlc/extractor/tiktok.py +++ b/youtube_dlc/extractor/tiktok.py @@ -1,138 +1,136 @@ # coding: utf-8 from __future__ import unicode_literals +from datetime import datetime from .common import InfoExtractor from ..utils import ( - compat_str, ExtractorError, int_or_none, str_or_none, - try_get, - url_or_none, + try_get ) class TikTokBaseIE(InfoExtractor): - def _extract_aweme(self, data): - video = data['video'] - description = str_or_none(try_get(data, lambda x: x['desc'])) - width = int_or_none(try_get(data, lambda x: video['width'])) - height = int_or_none(try_get(data, lambda x: video['height'])) + def _extract_aweme(self, video_data, webpage): + video_info = try_get( + video_data, lambda x: x['videoData']['itemInfos'], dict) + author_info = try_get( + video_data, lambda x: x['videoData']['authorInfos'], dict) + share_info = try_get(video_data, lambda x: x['shareMeta'], dict) + + unique_id = str_or_none(author_info.get('uniqueId')) + timestamp = try_get(video_info, lambda x: int(x['createTime']), int) + date = datetime.fromtimestamp(timestamp).strftime('%Y%m%d') + + height = try_get(video_info, lambda x: x['video']['videoMeta']['height'], int) + width = try_get(video_info, lambda x: x['video']['videoMeta']['width'], int) + thumbnails = [] + thumbnails.append({ + 'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage), + 'width': width, + 'height': height + }) - format_urls = set() formats = [] - for format_id in ( - 'play_addr_lowbr', 'play_addr', 'play_addr_h264', - 'download_addr'): - for format in try_get( - video, lambda x: x[format_id]['url_list'], list) or []: - format_url = url_or_none(format) - if not format_url: - continue - if format_url in format_urls: - continue - format_urls.add(format_url) - formats.append({ - 'url': format_url, - 'ext': 'mp4', - 'height': height, - 'width': width, - }) - self._sort_formats(formats) - - thumbnail = url_or_none(try_get( - video, lambda x: x['cover']['url_list'][0], compat_str)) - uploader = try_get(data, lambda x: x['author']['nickname'], compat_str) - timestamp = int_or_none(data.get('create_time')) - comment_count = int_or_none(data.get('comment_count')) or int_or_none( - try_get(data, lambda x: x['statistics']['comment_count'])) - repost_count = int_or_none(try_get( - data, lambda x: x['statistics']['share_count'])) - - aweme_id = data['aweme_id'] + formats.append({ + 'url': try_get(video_info, lambda x: x['video']['urls'][0]), + 'ext': 'mp4', + 'height': height, + 'width': width + }) return { - 'id': aweme_id, - 'title': uploader or aweme_id, - 'description': description, - 'thumbnail': thumbnail, - 'uploader': uploader, + 'comment_count': int_or_none(video_info.get('commentCount')), + 'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int), + 'height': height, + 'id': str_or_none(video_info.get('id')), + 'like_count': int_or_none(video_info.get('diggCount')), + 'repost_count': int_or_none(video_info.get('shareCount')), + 'thumbnail': try_get(video_info, lambda x: x['covers'][0]), 'timestamp': timestamp, - 'comment_count': comment_count, - 'repost_count': repost_count, - 'formats': formats, + 'width': width, + 'title': str_or_none(share_info.get('title')) or self._og_search_title(webpage), + 'creator': str_or_none(author_info.get('nickName')), + 'uploader': unique_id, + 'uploader_id': str_or_none(author_info.get('userId')), + 'uploader_url': 'https://www.tiktok.com/@' + unique_id, + 'thumbnails': thumbnails, + 'upload_date': date, + 'webpage_url': self._og_search_url(webpage), + 'description': str_or_none(video_info.get('text')) or str_or_none(share_info.get('desc')), + 'ext': 'mp4', + 'formats': formats } class TikTokIE(TikTokBaseIE): - _VALID_URL = r'''(?x) - https?:// - (?: - (?:m\.)?tiktok\.com/v| - (?:www\.)?tiktok\.com/share/video - ) - /(?P\d+) - ''' + _VALID_URL = r'https?://www\.tiktok\.com/@[\w\._]+/video/(?P\d+)' + _TESTS = [{ - 'url': 'https://m.tiktok.com/v/6606727368545406213.html', - 'md5': 'd584b572e92fcd48888051f238022420', + 'url': 'https://www.tiktok.com/@leenabhushan/video/6748451240264420610', + 'md5': '34a7543afd5a151b0840ba6736fb633b', 'info_dict': { - 'id': '6606727368545406213', - 'ext': 'mp4', - 'title': 'Zureeal', - 'description': '#bowsette#mario#cosplay#uk#lgbt#gaming#asian#bowsettecosplay', - 'thumbnail': r're:^https?://.*~noop.image', - 'uploader': 'Zureeal', - 'timestamp': 1538248586, - 'upload_date': '20180929', 'comment_count': int, + 'creator': 'facestoriesbyleenabh', + 'description': 'md5:a9f6c0c44a1ff2249cae610372d0ae95', + 'duration': 13, + 'ext': 'mp4', + 'formats': list, + 'height': 1280, + 'id': '6748451240264420610', + 'like_count': int, 'repost_count': int, + 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', + 'thumbnails': list, + 'timestamp': 1571246252, + 'title': 'facestoriesbyleenabh on TikTok', + 'upload_date': '20191016', + 'uploader': 'leenabhushan', + 'uploader_id': '6691488002098119685', + 'uploader_url': r're:https://www.tiktok.com/@leenabhushan', + 'webpage_url': r're:https://www.tiktok.com/@leenabhushan/(video/)?6748451240264420610', + 'width': 720, } }, { - 'url': 'https://www.tiktok.com/share/video/6606727368545406213', - 'only_matching': True, + 'url': 'https://www.tiktok.com/@patroxofficial/video/6742501081818877190?langCountry=en', + 'md5': '06b9800d47d5fe51a19e322dd86e61c9', + 'info_dict': { + 'comment_count': int, + 'creator': 'patroX', + 'description': 'md5:5e2a23877420bb85ce6521dbee39ba94', + 'duration': 27, + 'ext': 'mp4', + 'formats': list, + 'height': 960, + 'id': '6742501081818877190', + 'like_count': int, + 'repost_count': int, + 'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?', + 'thumbnails': list, + 'timestamp': 1569860870, + 'title': 'patroX on TikTok', + 'upload_date': '20190930', + 'uploader': 'patroxofficial', + 'uploader_id': '18702747', + 'uploader_url': r're:https://www.tiktok.com/@patroxofficial', + 'webpage_url': r're:https://www.tiktok.com/@patroxofficial/(video/)?6742501081818877190', + 'width': 540, + } }] def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage( - 'https://m.tiktok.com/v/%s.html' % video_id, video_id) - data = self._parse_json(self._search_regex( - r'\bdata\s*=\s*({.+?})\s*;', webpage, 'data'), video_id) - return self._extract_aweme(data) + webpage = self._download_webpage(url, video_id, note='Downloading video webpage') + json_string = self._search_regex( + r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P[^<]+)', + webpage, 'json_string', group='json_string_ld') + json_data = self._parse_json(json_string, video_id) + video_data = try_get(json_data, lambda x: x['props']['pageProps'], expected_type=dict) -class TikTokUserIE(TikTokBaseIE): - _VALID_URL = r'''(?x) - https?:// - (?: - (?:m\.)?tiktok\.com/h5/share/usr| - (?:www\.)?tiktok\.com/share/user - ) - /(?P\d+) - ''' - _TESTS = [{ - 'url': 'https://m.tiktok.com/h5/share/usr/188294915489964032.html', - 'info_dict': { - 'id': '188294915489964032', - }, - 'playlist_mincount': 24, - }, { - 'url': 'https://www.tiktok.com/share/user/188294915489964032', - 'only_matching': True, - }] + # Chech statusCode for success + if video_data.get('statusCode') == 0: + return self._extract_aweme(video_data, webpage) - def _real_extract(self, url): - user_id = self._match_id(url) - data = self._download_json( - 'https://m.tiktok.com/h5/share/usr/list/%s/' % user_id, user_id, - query={'_signature': '_'}) - entries = [] - for aweme in data['aweme_list']: - try: - entry = self._extract_aweme(aweme) - except ExtractorError: - continue - entry['extractor_key'] = TikTokIE.ie_key() - entries.append(entry) - return self.playlist_result(entries, user_id) + raise ExtractorError('Video not available', video_id=video_id)