[tiktok] fix extraction

src/you_get/extractors/tiktok.py

@@ -5,26 +5,42 @@
 from ..common import *
 
 def tiktok_download(url, output_dir='.', merge=True, info_only=False, **kwargs):
-    html = get_html(url, faker=True)
+    referUrl = url.split('?')[0]
+    headers = fake_headers
 
+    # trick or treat
+    html = get_content(url, headers=headers)
     data = r1(r'<script id="__NEXT_DATA__".*?>(.*?)</script>', html)
     info = json.loads(data)
-    videoData = info['props']['pageProps']['videoData']
-    urls = videoData['itemInfos']['video']['urls']
-    videoId = videoData['itemInfos']['id']
-    uniqueId = videoData['authorInfos'].get('uniqueId')
-    nickName = videoData['authorInfos'].get('nickName')
-
-    for i, videoUrl in enumerate(urls):
-        title = '%s [%s]' % (nickName or uniqueId, videoId)
-        if len(urls) > 1:
-            title = '%s [%s]' % (title, i)
-
-        mime, ext, size = url_info(videoUrl, headers={'Referer': url})
-
-        print_info(site_info, title, mime, size)
-        if not info_only:
-            download_urls([videoUrl], title, ext, size, output_dir=output_dir, merge=merge, headers={'Referer': url})
+    wid = info['props']['initialProps']['$wid']
+    cookie = 'tt_webid=%s; tt_webid_v2=%s' % (wid, wid)
+
+    # here's the cookie
+    headers['Cookie'] = cookie
+
+    # try again
+    html = get_content(url, headers=headers)
+    data = r1(r'<script id="__NEXT_DATA__".*?>(.*?)</script>', html)
+    info = json.loads(data)
+    wid = info['props']['initialProps']['$wid']
+    cookie = 'tt_webid=%s; tt_webid_v2=%s' % (wid, wid)
+
+    videoData = info['props']['pageProps']['itemInfo']['itemStruct']
+    videoId = videoData['id']
+    videoUrl = videoData['video']['downloadAddr']
+    uniqueId = videoData['author'].get('uniqueId')
+    nickName = videoData['author'].get('nickname')
+
+    title = '%s [%s]' % (nickName or uniqueId, videoId)
+
+    # we also need the referer
+    headers['Referer'] = referUrl
+
+    mime, ext, size = url_info(videoUrl, headers=headers)
+
+    print_info(site_info, title, mime, size)
+    if not info_only:
+        download_urls([videoUrl], title, ext, size, output_dir=output_dir, merge=merge, headers=headers)
 
 site_info = "TikTok.com"
 download = tiktok_download