[tiktok] rewrite for new API and website

youtube_dl/extractor/extractors.py

@@ -1143,10 +1143,7 @@ from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
 from .thisoldhouse import ThisOldHouseIE
 from .threeqsdn import ThreeQSDNIE
-from .tiktok import (
+from .tiktok import TikTokIE
-    TikTokIE,
-    TikTokUserIE,
-)
 from .tinypic import TinyPicIE
 from .tmz import (
     TMZIE,

youtube_dl/extractor/tiktok.py

@@ -1,138 +1,47 @@
 # coding: utf-8
-from __future__ import unicode_literals
-
 from .common import InfoExtractor
 from ..utils import (
-    compat_str,
-    ExtractorError,
     int_or_none,
-    str_or_none,
-    try_get,
-    url_or_none,
 )
 
 
-class TikTokBaseIE(InfoExtractor):
+class TikTokIE(InfoExtractor):
-    def _extract_aweme(self, data):
+    _VALID_URL = r'https?://(?:www\.)?tiktok\.com/.*/video/(?P<id>[^/?#&]+)'
-        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']))
-
-        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']
-
-        return {
-            'id': aweme_id,
-            'title': uploader or aweme_id,
-            'description': description,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
-            'timestamp': timestamp,
-            'comment_count': comment_count,
-            'repost_count': repost_count,
-            'formats': formats,
-        }
-
-
-class TikTokIE(TikTokBaseIE):
-    _VALID_URL = r'''(?x)
-                        https?://
-                            (?:
-                                (?:m\.)?tiktok\.com/v|
-                                (?:www\.)?tiktok\.com/share/video
-                            )
-                            /(?P<id>\d+)
-                    '''
     _TESTS = [{
-        'url': 'https://m.tiktok.com/v/6606727368545406213.html',
+        'url': 'https://www.tiktok.com/@ben_brainard/video/6858321293117312262',
-        'md5': 'd584b572e92fcd48888051f238022420',
+        # 'md5': 'd584b572e92fcd48888051f238022420',
         'info_dict': {
-            'id': '6606727368545406213',
+            'id': '6858321293117312262',
             'ext': 'mp4',
-            'title': 'Zureeal',
+            'title': 'ben_brainard',
-            'description': '#bowsette#mario#cosplay#uk#lgbt#gaming#asian#bowsettecosplay',
+            'description': 'Hurricane... hurricane... hurricane... I got nothing.',
-            'thumbnail': r're:^https?://.*~noop.image',
+            'uploader': 'ben_brainard',
-            'uploader': 'Zureeal',
+            'timestamp': 1596827366,
-            'timestamp': 1538248586,
+            'upload_date': '20200807',
-            'upload_date': '20180929',
+            'comment_count': int
-            'comment_count': int,
-            'repost_count': int,
         }
-    }, {
-        'url': 'https://www.tiktok.com/share/video/6606727368545406213',
-        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(
+        webpage = self._download_webpage(url, video_id)
-            'https://m.tiktok.com/v/%s.html' % video_id, video_id)
+        # print webpage
-        data = self._parse_json(self._search_regex(
+        description = self._search_regex(r'<title[^>]*>([^<]+)</title>', webpage, 'description')
-            r'\bdata\s*=\s*({.+?})\s*;', webpage, 'data'), video_id)
+        mediaUrl = self._search_regex(r'"video":{"urls":\["([^"]+)"', webpage, 'url')
-        return self._extract_aweme(data)
+        # data = self._parse_html5_media_entries(url,webpage,video_id)
+        uploader = self._search_regex(r'"authorName":"([^"]+)"', webpage, 'uploader')
+        commentCount = self._search_regex(r'"commentCount":([^,]+)', webpage, 'comment_count')
+        viewCount = self._search_regex(r'"playCount":([^,]+)', webpage, 'view_count')
+        timestamp = self._search_regex(r'"createTime":"([^"]+)"', webpage, 'timestamp')
 
-
+        return {
-class TikTokUserIE(TikTokBaseIE):
+            'id': video_id,
-    _VALID_URL = r'''(?x)
+            'url': mediaUrl,
-                        https?://
+            'title': uploader,
-                            (?:
+            'description': description,
-                                (?:m\.)?tiktok\.com/h5/share/usr|
+            'ext': 'mp4',
-                                (?:www\.)?tiktok\.com/share/user
+            'uploader': uploader,
-                            )
+            'comment_count': int_or_none(commentCount),
-                            /(?P<id>\d+)
+            'view_count': int_or_none(viewCount),
-                    '''
+            'timestamp': int_or_none(timestamp)
-    _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,
-    }]
-
-    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)