[ie/orf:on] Improve extraction (#9677)

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

diff --git a/yt_dlp/extractor/safari.py b/yt_dlp/extractor/safari.py
index c92e8849bdcc4ea63233d49832bb50a3b15775b0..8d322d7105b06b4901e8e28b615983c191045666 100644 (file)
--- a/yt_dlp/extractor/safari.py
+++ b/yt_dlp/extractor/safari.py
@@ -1,6 +1,3 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 import json
 import re
 
@@ -25,26 +22,19 @@ class SafariBaseIE(InfoExtractor):
 
     LOGGED_IN = False
 
-    def _real_initialize(self):
-        self._login()
-
-    def _login(self):
-        username, password = self._get_login_info()
-        if username is None:
-            return
-
+    def _perform_login(self, username, password):
         _, urlh = self._download_webpage_handle(
             'https://learning.oreilly.com/accounts/login-check/', None,
             'Downloading login page')
 
         def is_logged(urlh):
-            return 'learning.oreilly.com/home/' in urlh.geturl()
+            return 'learning.oreilly.com/home/' in urlh.url
 
         if is_logged(urlh):
             self.LOGGED_IN = True
             return
 
-        redirect_url = urlh.geturl()
+        redirect_url = urlh.url
         parsed_url = compat_urlparse.urlparse(redirect_url)
         qs = compat_parse_qs(parsed_url.query)
         next_uri = compat_urlparse.urljoin(
@@ -139,7 +129,7 @@ def _real_extract(self, url):
 
             webpage, urlh = self._download_webpage_handle(url, video_id)
 
-            mobj = re.match(self._VALID_URL, urlh.geturl())
+            mobj = re.match(self._VALID_URL, urlh.url)
             reference_id = mobj.group('reference_id')
             if not reference_id:
                 reference_id = self._search_regex(
@@ -193,7 +183,12 @@ def _real_extract(self, url):
         part = self._download_json(
             url, '%s/%s' % (mobj.group('course_id'), mobj.group('part')),
             'Downloading part JSON')
-        return self.url_result(part['web_url'], SafariIE.ie_key())
+        web_url = part['web_url']
+        if 'library/view' in web_url:
+            web_url = web_url.replace('library/view', 'videos')
+            natural_keys = part['natural_key']
+            web_url = f'{web_url.rsplit("/", 1)[0]}/{natural_keys[0]}-{natural_keys[1][:-5]}'
+        return self.url_result(web_url, SafariIE.ie_key())
 
 
 class SafariCourseIE(SafariBaseIE):