From 1666c2709f575a20c7ae1d0c76935b855e872dc9 Mon Sep 17 00:00:00 2001
From: Dan Walker <dwalker@tripwire.com>
Date: Sun, 25 Oct 2020 06:08:57 -0700
Subject: [PATCH 1/2] Fixed tiktok.py The JSON formatting of the response
 changed, corrected.

#TESTING
Successfully downloaded and played a tiktok video
---
 youtube_dlc/extractor/tiktok.py | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/youtube_dlc/extractor/tiktok.py b/youtube_dlc/extractor/tiktok.py
index 52e5f4f1f..f27aa4250 100644
--- a/youtube_dlc/extractor/tiktok.py
+++ b/youtube_dlc/extractor/tiktok.py
@@ -14,17 +14,17 @@
 class TikTokBaseIE(InfoExtractor):
     def _extract_aweme(self, video_data, webpage, url):
         video_info = try_get(
-            video_data, lambda x: x['videoData']['itemInfos'], dict)
+            video_data, lambda x: x['itemInfo']['itemStruct'], dict)
         author_info = try_get(
-            video_data, lambda x: x['videoData']['authorInfos'], dict)
-        share_info = try_get(video_data, lambda x: x['shareMeta'], dict)
+            video_data, lambda x: x['itemInfo']['itemStruct']['author'], dict)
+        share_info = try_get(video_data, lambda x: x['itemInfo']['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)
+        height = try_get(video_info, lambda x: x['video']['height'], int)
+        width = try_get(video_info, lambda x: x['video']['width'], int)
         thumbnails = []
         thumbnails.append({
             'url': video_info.get('thumbnail') or self._og_search_thumbnail(webpage),
@@ -34,7 +34,7 @@ def _extract_aweme(self, video_data, webpage, url):
 
         formats = []
         formats.append({
-            'url': try_get(video_info, lambda x: x['video']['urls'][0]),
+            'url': try_get(video_info, lambda x: x['video']['playAddr']),
             'ext': 'mp4',
             'height': height,
             'width': width

From 4f5a0ad82214456c3d4832bbf4e15e416e1a16cb Mon Sep 17 00:00:00 2001
From: Unknown <blackjack4494@web.de>
Date: Mon, 26 Oct 2020 02:15:46 +0100
Subject: [PATCH 2/2] [tiktok] fallback url + cookie

---
 youtube_dlc/extractor/tiktok.py | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/youtube_dlc/extractor/tiktok.py b/youtube_dlc/extractor/tiktok.py
index f27aa4250..0cfd2168a 100644
--- a/youtube_dlc/extractor/tiktok.py
+++ b/youtube_dlc/extractor/tiktok.py
@@ -12,7 +12,8 @@
 
 
 class TikTokBaseIE(InfoExtractor):
-    def _extract_aweme(self, video_data, webpage, url):
+    def _extract_aweme(self, props_data, webpage, url):
+        video_data = try_get(props_data, lambda x: x['pageProps'], expected_type=dict)
         video_info = try_get(
             video_data, lambda x: x['itemInfo']['itemStruct'], dict)
         author_info = try_get(
@@ -32,14 +33,20 @@ def _extract_aweme(self, video_data, webpage, url):
             'height': height
         })
 
+        url = ''
+        if not url:
+            url = try_get(video_info, lambda x: x['video']['playAddr'])
+        if not url:
+            url = try_get(video_info, lambda x: x['video']['downloadAddr'])
         formats = []
         formats.append({
-            'url': try_get(video_info, lambda x: x['video']['playAddr']),
+            'url': url,
             'ext': 'mp4',
             'height': height,
             'width': width
         })
 
+        tracker = try_get(props_data, lambda x: x['initialProps']['$wid'])
         return {
             'comment_count': int_or_none(video_info.get('commentCount')),
             'duration': try_get(video_info, lambda x: x['video']['videoMeta']['duration'], int),
@@ -63,6 +70,7 @@ def _extract_aweme(self, video_data, webpage, url):
             'formats': formats,
             'http_headers': {
                 'Referer': url,
+                'Cookie': 'tt_webid=%s; tt_webid_v2=%s' % (tracker, tracker),
             }
         }
 
@@ -130,10 +138,10 @@ def _real_extract(self, url):
             r'id=\"__NEXT_DATA__\"\s+type=\"application\/json\"\s*[^>]+>\s*(?P<json_string_ld>[^<]+)',
             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)
+        props_data = try_get(json_data, lambda x: x['props'], expected_type=dict)
 
         # Chech statusCode for success
-        if video_data.get('statusCode') == 0:
-            return self._extract_aweme(video_data, webpage, url)
+        if props_data.get('pageProps').get('statusCode') == 0:
+            return self._extract_aweme(props_data, webpage, url)
 
         raise ExtractorError('Video not available', video_id=video_id)