[extractor] Deprecate `_sort_formats`

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

diff --git a/yt_dlp/extractor/cbc.py b/yt_dlp/extractor/cbc.py
index 392c778848814bc96c60c3773e36ad308a87a1c4..210f5f8eea4cc14c341380832d823877d9c72a39 100644 (file)
--- a/yt_dlp/extractor/cbc.py
+++ b/yt_dlp/extractor/cbc.py
@@ -1,6 +1,3 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 import re
 import json
 import base64
@@ -127,9 +124,9 @@ def _extract_player_init(self, player_init, display_id):
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
-        title = self._og_search_title(webpage, default=None) or self._html_search_meta(
-            'twitter:title', webpage, 'title', default=None) or self._html_search_regex(
-                r'<title>([^<]+)</title>', webpage, 'title', fatal=False)
+        title = (self._og_search_title(webpage, default=None)
+                 or self._html_search_meta('twitter:title', webpage, 'title', default=None)
+                 or self._html_extract_title(webpage))
         entries = [
             self._extract_player_init(player_init, display_id)
             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
@@ -307,13 +304,13 @@ def claims_token_valid(self):
     def _get_claims_token(self, email, password):
         if not self.claims_token_valid():
             self._claims_token = self._new_claims_token(email, password)
-            self._downloader.cache.store(self._NETRC_MACHINE, 'claims_token', self._claims_token)
+            self.cache.store(self._NETRC_MACHINE, 'claims_token', self._claims_token)
         return self._claims_token
 
     def _real_initialize(self):
         if self.claims_token_valid():
             return
-        self._claims_token = self._downloader.cache.load(self._NETRC_MACHINE, 'claims_token')
+        self._claims_token = self.cache.load(self._NETRC_MACHINE, 'claims_token')
 
     def _find_secret_formats(self, formats, video_id):
         """ Find a valid video url and convert it to the secret variant """
@@ -340,7 +337,8 @@ def _find_secret_formats(self, formats, video_id):
                 yield {
                     **base_format,
                     'format_id': join_nonempty('sec', height),
-                    'url': re.sub(r'(QualityLevels\()\d+(\))', fr'\<1>{bitrate}\2', base_url),
+                    # Note: \g<1> is necessary instead of \1 since bitrate is a number
+                    'url': re.sub(r'(QualityLevels\()\d+(\))', fr'\g<1>{bitrate}\2', base_url),
                     'width': int_or_none(video_quality.attrib.get('MaxWidth')),
                     'tbr': bitrate / 1000.0,
                     'height': height,
@@ -382,8 +380,6 @@ def _real_extract(self, url):
                 if 'descriptive' in format['format_id'].lower():
                     format['preference'] = -2
 
-        self._sort_formats(formats)
-
         return {
             'id': video_id,
             'title': video_info['title'],