Update to ytdl-commit-cf2dbec

[yt-dlp.git] / youtube_dlc / extractor / vimeo.py

diff --git a/youtube_dlc/extractor/vimeo.py b/youtube_dlc/extractor/vimeo.py
index 51a0ab2fad49db774997c89ecb092097b98824af..ecfb5f0c5fbe9212f93f777910e8d0231a85dff8 100644 (file)
--- a/youtube_dlc/extractor/vimeo.py
+++ b/youtube_dlc/extractor/vimeo.py
@@ -116,7 +116,8 @@ def _set_vimeo_cookie(self, name, value):
     def _vimeo_sort_formats(self, formats):
         # Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
         # at the same time without actual units specified. This lead to wrong sorting.
-        self._sort_formats(formats, field_preference=('preference', 'height', 'width', 'fps', 'tbr', 'format_id'))
+        # But since yt-dlp prefers 'res,fps' anyway, 'field_preference' is not needed
+        self._sort_formats(formats)
 
     def _parse_config(self, config, video_id):
         video_data = config['video']
@@ -178,15 +179,9 @@ def _parse_config(self, config, video_id):
             formats.append({
                 'format_id': 'live-archive-source',
                 'url': live_archive_source_url,
-                'preference': 1,
+                'quality': 10,
             })
 
-        for f in formats:
-            if f.get('vcodec') == 'none':
-                f['preference'] = -50
-            elif f.get('acodec') == 'none':
-                f['preference'] = -40
-
         subtitles = {}
         text_tracks = config['request'].get('text_tracks')
         if text_tracks:
@@ -226,10 +221,12 @@ def _parse_config(self, config, video_id):
             'is_live': is_live,
         }
 
-    def _extract_original_format(self, url, video_id):
+    def _extract_original_format(self, url, video_id, unlisted_hash=None):
+        query = {'action': 'load_download_config'}
+        if unlisted_hash:
+            query['unlisted_hash'] = unlisted_hash
         download_data = self._download_json(
-            url, video_id, fatal=False,
-            query={'action': 'load_download_config'},
+            url, video_id, fatal=False, query=query,
             headers={'X-Requested-With': 'XMLHttpRequest'})
         if download_data:
             source_file = download_data.get('source_file')
@@ -249,7 +246,7 @@ def _extract_original_format(self, url, video_id):
                             'height': int_or_none(source_file.get('height')),
                             'filesize': parse_filesize(source_file.get('size')),
                             'format_id': source_name,
-                            'preference': 1,
+                            'quality': 1,
                         }
 
 
@@ -509,6 +506,11 @@ class VimeoIE(VimeoBaseInfoExtractor):
         {
             'url': 'https://vimeo.com/160743502/abd0e13fb4',
             'only_matching': True,
+        },
+        {
+            # requires passing unlisted_hash(a52724358e) to load_download_config request
+            'url': 'https://vimeo.com/392479337/a52724358e',
+            'only_matching': True,
         }
         # https://gettingthingsdone.com/workflowmap/
         # vimeo embed with check-password page protected by Referer header
@@ -673,7 +675,8 @@ def _real_extract(self, url):
             if config.get('view') == 4:
                 config = self._verify_player_video_password(redirect_url, video_id, headers)
 
-        vod = config.get('video', {}).get('vod', {})
+        video = config.get('video') or {}
+        vod = video.get('vod') or {}
 
         def is_rented():
             if '>You rented this title.<' in webpage:
@@ -733,7 +736,7 @@ def is_rented():
         formats = []
 
         source_format = self._extract_original_format(
-            'https://vimeo.com/' + video_id, video_id)
+            'https://vimeo.com/' + video_id, video_id, video.get('unlisted_hash'))
         if source_format:
             formats.append(source_format)
 
@@ -1119,6 +1122,12 @@ class VHXEmbedIE(VimeoBaseInfoExtractor):
     IE_NAME = 'vhx:embed'
     _VALID_URL = r'https?://embed\.vhx\.tv/videos/(?P<id>\d+)'
 
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+src="(https?://embed\.vhx\.tv/videos/\d+[^"]*)"', webpage)
+        return unescapeHTML(mobj.group(1)) if mobj else None
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
@@ -1127,5 +1136,6 @@ def _real_extract(self, url):
             'ott data'), video_id, js_to_json)['config_url']
         config = self._download_json(config_url, video_id)
         info = self._parse_config(config, video_id)
+        info['id'] = video_id
         self._vimeo_sort_formats(info['formats'])
         return info