tiktok fix code

test/ci/test_tiktok.py

@@ -4,13 +4,26 @@ import youtube_dl
 
 class TikTokTestYoutubeDl(unittest.TestCase):
     def test_meta_data(self):
-        url = 'https://www.tiktok.com/@danieltbraun/video/6817099671043853574'
+        url = 'https://www.tiktok.com/@oriangaon/video/6807126376001441030'
         params = {}
         ydl = youtube_dl.YoutubeDL(params)
         info = ydl.extract_info(url, download=False)
-        self.assertEquals(info['share_count'], 121)
+        self.assertEquals(info['id'], '6807126376001441030')
-
+        self.assertEquals(info['url'], 'https://www.tiktok.com/@oriangaon/video/6807126376001441030')
-
+        self.assertEquals(info['title'], '#foryou #foyou Mmmmm....,,')
+        self.assertEquals(info['uploader'], 'Oriangaon')
+        self.assertEquals(info['timestamp'], 1584907616)
+        self.assertEquals(info['thumbnail'],
+                          'https://p16-va-default.akamaized.net/obj/tos-maliva-p-0068/d1a8fbd3e42dda3a1baa01ee9edad289')
+        self.assertGreaterEqual(info['view_count'], 79864)
+        self.assertEquals(info['uploader_id'], '6772113344733955077')
+        self.assertFalse(info['is_live'])
+        self.assertEquals(info['live_status'], 'not_live')
+        self.assertGreaterEqual(info['like_count'], 2213)
+        self.assertGreaterEqual(info['share_count'], 109)
+        self.assertGreaterEqual(info['comment_count'], 40)
+        self.assertEquals(info['duration'], 10)
+        self.assertEquals(info['ext'], 'mp.4')
 
 
 if __name__ == '__main__':

youtube_dl/extractor/tiktok.py

@@ -2,7 +2,6 @@
 from __future__ import unicode_literals
 import ast
 from bs4 import BeautifulSoup
-import requests
 import json
 from .common import InfoExtractor
 from ..utils import (
@@ -11,11 +10,7 @@ from ..utils import (
     int_or_none,
     str_or_none,
     try_get,
-    url_or_none,
+    url_or_none)
-)
-
-
-# add to requirements.txt- bs4, newspaper, requests
 
 
 class TikTokBaseIE(InfoExtractor):
@@ -70,8 +65,6 @@ class TikTokBaseIE(InfoExtractor):
         }
 
 
-
-
 class TikTokIE(TikTokBaseIE):
     _VALID_URL = r'''(?x)
                         https?://
@@ -103,52 +96,31 @@ class TikTokIE(TikTokBaseIE):
     }]
 
     def _real_extract(self, url):
-        video_id = url.split('/')[-1]
+        video_id = self._match_id(url)
-
-        # extract meta data using the official api
-        # Response json contains: provider url, title, html, author_namee, height, thumbnail_width, width, version,
-        # author_url, thumbnail_height, thumbnail_url, type, provider_name (tiktok)
-
         json_api = self._download_json('https://www.tiktok.com/oembed?url=' + url, video_id)
 
-        # extract metadata with beautifulSoup
         webpage = self._download_webpage(url, video_id)
         soup = BeautifulSoup(webpage, features="html.parser")
-        h2 = soup.find_all("h2", {"class": "jsx-1038045583 jsx-3192540912 jsx-2150087249 video-meta-count"})
-        data = h2[0].text.split(' ')
-        likes_count = self.numeric_convert(data[0])
-        comments_count = self.numeric_convert(data[3])
         json_next_data = soup.find(id='__NEXT_DATA__')
         props = json_next_data.contents[0]
         json_data_encode = json.dumps(props.encode('utf-8'))
         ast_le = ast.literal_eval(json_data_encode)
         data_dict = json.loads(ast_le)
-        timestamp = self.numeric_convert(data_dict['props']['pageProps']['videoData']['itemInfos']['createTime'])
 
-        shares = data_dict['props']['pageProps']['videoData']['itemInfos']['shareCount']
+        item_info = data_dict['props']['pageProps']['videoData']['itemInfos']
-        views = data_dict['props']['pageProps']['videoData']['itemInfos']['playCount']
+        timestamp = int(item_info['createTime'])
-        duration = data_dict['props']['pageProps']['videoData']['itemInfos']['video']['videoMeta']['duration']
+        shares = item_info['shareCount']
-        provider_id = data_dict['props']['pageProps']['videoData']['itemInfos']['authorId']
+        views = item_info['playCount']
-
+        duration = item_info['video']['videoMeta']['duration']
-        # TO-DO- check on formats
+        provider_id = item_info['authorId']
+        comments_count = item_info['commentCount']
+        likes_count = item_info['diggCount']
 
         return self.info_dict(video_id, str(url), json_api['title'],
                               json_api['author_name'], timestamp, json_api['thumbnail_url'],
-                              views, provider_id, False, 'not_live', likes_count, shares, '', comments_count,duration)
+                              views, provider_id, False, 'not_live', likes_count, shares, '', comments_count, duration)
 
-    def numeric_convert(self, unicode):
+    def info_dict(self, video_id, url, video_title,
-        if 'K' in unicode:
-            unicode=unicode[:-1]
-            return int(float(unicode)*1000)
-        if 'M' in unicode:
-            unicode=unicode[:-1]
-            return int(float(unicode)*100000)
-        else:
-            return int(unicode)
-
-
-
-    def info_dict (self, video_id, url, video_title,
                   uploader, timestamp, thumbnail,
                   view_count, uploader_id, is_live, live_status
                   , likes_count, shares_count, subtitles, comment_count, duration):
@@ -167,11 +139,13 @@ class TikTokIE(TikTokBaseIE):
             'share_count': shares_count,
             'subtitles': subtitles,
             'comment_count': comment_count,
-            'duration': duration
+            'duration': duration,
+            'ext':'mp.4'
 
         }
         return info_dict
 
+
 class TikTokUserIE(TikTokBaseIE):
     _VALID_URL = r'''(?x)
                         https?://