[extractor] Deprecate `_sort_formats`

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

diff --git a/yt_dlp/extractor/vimeo.py b/yt_dlp/extractor/vimeo.py
index 9e17149bede5d132ae5e80df9a91720a7f9056a5..26fe566b03130f500a7e2fe793faa5bffd783bdc 100644 (file)
--- a/yt_dlp/extractor/vimeo.py
+++ b/yt_dlp/extractor/vimeo.py
@@ -123,11 +123,6 @@ def _extract_vimeo_config(self, webpage, video_id, *args, **kwargs):
     def _set_vimeo_cookie(self, name, value):
         self._set_cookie('vimeo.com', name, value)
 
-    def _vimeo_sort_formats(self, formats):
-        # Note: Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
-        # at the same time without actual units specified.
-        self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source'))
-
     def _parse_config(self, config, video_id):
         video_data = config['video']
         video_title = video_data.get('title')
@@ -242,6 +237,9 @@ def _parse_config(self, config, video_id):
             'formats': formats,
             'subtitles': subtitles,
             'is_live': is_live,
+            # Note: Bitrates are completely broken. Single m3u8 may contain entries in kbps and bps
+            # at the same time without actual units specified.
+            '_format_sort_fields': ('quality', 'res', 'fps', 'hdr:12', 'source'),
         }
 
     def _extract_original_format(self, url, video_id, unlisted_hash=None):
@@ -776,7 +774,6 @@ def _extract_from_api(self, video_id, unlisted_hash=None):
             })
         info = self._parse_config(self._download_json(
             video['config_url'], video_id), video_id)
-        self._vimeo_sort_formats(info['formats'])
         get_timestamp = lambda x: parse_iso8601(video.get(x + '_time'))
         info.update({
             'description': video.get('description'),
@@ -870,13 +867,11 @@ def _real_extract(self, url):
 
         if '://player.vimeo.com/video/' in url:
             config = self._parse_json(self._search_regex(
-                r'\bconfig\s*=\s*({.+?})\s*;', webpage, 'info section'), video_id)
+                r'\b(?:playerC|c)onfig\s*=\s*({.+?})\s*;', webpage, 'info section'), video_id)
             if config.get('view') == 4:
                 config = self._verify_player_video_password(
                     redirect_url, video_id, headers)
-            info = self._parse_config(config, video_id)
-            self._vimeo_sort_formats(info['formats'])
-            return info
+            return self._parse_config(config, video_id)
 
         if re.search(r'<form[^>]+?id="pw_form"', webpage):
             video_password = self._get_video_password()
@@ -981,7 +976,7 @@ def is_rented():
 
         info_dict_config = self._parse_config(config, video_id)
         formats.extend(info_dict_config['formats'])
-        self._vimeo_sort_formats(formats)
+        info_dict['_format_sort_fields'] = info_dict_config['_format_sort_fields']
 
         json_ld = self._search_json_ld(webpage, video_id, default={})
 
@@ -1004,7 +999,7 @@ def is_rented():
         return merge_dicts(info_dict, info_dict_config, json_ld)
 
 
-class VimeoOndemandIE(VimeoIE):
+class VimeoOndemandIE(VimeoIE):  # XXX: Do not subclass from concrete IE
     IE_NAME = 'vimeo:ondemand'
     _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?:[^/]+/)?(?P<id>[^/?#&]+)'
     _TESTS = [{
@@ -1129,9 +1124,9 @@ def _real_extract(self, url):
         return self._extract_videos(channel_id, self._BASE_URL_TEMPL % channel_id)
 
 
-class VimeoUserIE(VimeoChannelIE):
+class VimeoUserIE(VimeoChannelIE):  # XXX: Do not subclass from concrete IE
     IE_NAME = 'vimeo:user'
-    _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<id>[^/]+)(?:/videos|[#?]|$)'
+    _VALID_URL = r'https://vimeo\.com/(?!(?:[0-9]+|watchlater)(?:$|[?#/]))(?P<id>[^/]+)(?:/videos)?/?(?:$|[?#])'
     _TITLE_RE = r'<a[^>]+?class="user">([^<>]+?)</a>'
     _TESTS = [{
         'url': 'https://vimeo.com/nkistudio/videos',
@@ -1140,6 +1135,9 @@ class VimeoUserIE(VimeoChannelIE):
             'id': 'nkistudio',
         },
         'playlist_mincount': 66,
+    }, {
+        'url': 'https://vimeo.com/nkistudio/',
+        'only_matching': True,
     }]
     _BASE_URL_TEMPL = 'https://vimeo.com/%s'
 
@@ -1236,7 +1234,7 @@ def _real_extract(self, url):
             entries, album_id, album.get('name'), album.get('description'))
 
 
-class VimeoGroupsIE(VimeoChannelIE):
+class VimeoGroupsIE(VimeoChannelIE):  # XXX: Do not subclass from concrete IE
     IE_NAME = 'vimeo:group'
     _VALID_URL = r'https://vimeo\.com/groups/(?P<id>[^/]+)(?:/(?!videos?/\d+)|$)'
     _TESTS = [{
@@ -1323,12 +1321,11 @@ def _real_extract(self, url):
             page_url + '/action', video_id)
         if source_format:
             info_dict['formats'].append(source_format)
-        self._vimeo_sort_formats(info_dict['formats'])
         info_dict['description'] = clean_html(clip_data.get('description'))
         return info_dict
 
 
-class VimeoWatchLaterIE(VimeoChannelIE):
+class VimeoWatchLaterIE(VimeoChannelIE):  # XXX: Do not subclass from concrete IE
     IE_NAME = 'vimeo:watchlater'
     IE_DESC = 'Vimeo watch later list, ":vimeowatchlater" keyword (requires authentication)'
     _VALID_URL = r'https://vimeo\.com/(?:home/)?watchlater|:vimeowatchlater'
@@ -1351,7 +1348,7 @@ def _real_extract(self, url):
         return self._extract_videos('watchlater', 'https://vimeo.com/watchlater')
 
 
-class VimeoLikesIE(VimeoChannelIE):
+class VimeoLikesIE(VimeoChannelIE):  # XXX: Do not subclass from concrete IE
     _VALID_URL = r'https://(?:www\.)?vimeo\.com/(?P<id>[^/]+)/likes/?(?:$|[?#]|sort:)'
     IE_NAME = 'vimeo:likes'
     IE_DESC = 'Vimeo user likes'
@@ -1395,5 +1392,4 @@ def _real_extract(self, 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