fix douyin extractor

src/you_get/extractors/douyin.py

@@ -1,8 +1,6 @@
 # coding=utf-8
 
-import re
 import json
-from urllib.parse import unquote
 
 from ..common import (
     url_size,
@@ -11,25 +9,52 @@ from ..common import (
     fake_headers,
     download_urls,
     playlist_not_supported,
+    match1,
+    get_location,
 )
 
-
 __all__ = ['douyin_download_by_url']
 
 
+def get_value(source: dict, path):
+    try:
+        value = source
+        for key in path:
+            if type(key) is str:
+                if key in value.keys():
+                    value = value[key]
+                else:
+                    value = None
+                    break
+            elif type(key) is int:
+                if len(value) != 0:
+                    value = value[key]
+                else:
+                    value = None
+                    break
+    except:
+        value = None
+    return value
+
+
 def douyin_download_by_url(url, **kwargs):
+    # if short link, get the real url
+    if 'v.douyin.com' in url:
+        url = get_location(url)
+    aweme_id = match1(url, r'/(\d+)/?')
+    # get video info
+    video_info_api = 'https://www.douyin.com/web/api/v2/aweme/iteminfo/?item_ids={}'
+    url = video_info_api.format(aweme_id)
     page_content = get_content(url, headers=fake_headers)
-    # The video player and video source are rendered client-side, the data
+    video_info = json.loads(page_content)
-    # contains in a <script id="RENDER_DATA" type="application/json"> tag
+
-    # quoted, unquote the whole page content then search using regex with
+    # get video id and title
-    # regular string.
+    video_id = get_value(video_info, ['item_list', 0, 'video', 'vid'])
-    page_content = unquote(page_content)
+    title = get_value(video_info, ['item_list', 0, 'desc'])
-    title = re.findall(r'"desc":"([^"]*)"', page_content)[0].strip()
+
+    # get video play url
+    video_url = "https://aweme.snssdk.com/aweme/v1/playwm/?ratio=720p&line=0&video_id={}".format(video_id)
     video_format = 'mp4'
-    # video URLs are in this pattern {"src":"THE_URL"}, in json format
-    urls_pattern = r'"playAddr":(\[.*?\])'
-    urls = json.loads(re.findall(urls_pattern, page_content)[0])
-    video_url = 'https:' + urls[0]['src']
     size = url_size(video_url, faker=True)
     print_info(
         site_info='douyin.com', title=title,