[ie/crunchyroll] Fix stream extraction (#10005)

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

diff --git a/yt_dlp/extractor/curiositystream.py b/yt_dlp/extractor/curiositystream.py
index 628c83631950a5e167ff36ee126bb500a0f70bdc..941cf4e79c9549e4adef4ababae2a576809cdee4 100644 (file)
--- a/yt_dlp/extractor/curiositystream.py
+++ b/yt_dlp/extractor/curiositystream.py
@@ -1,21 +1,14 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 import re
+import urllib.parse
 
 from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    urlencode_postdata,
-    compat_str,
-    ExtractorError,
-)
+from ..compat import compat_str
+from ..utils import ExtractorError, int_or_none, urlencode_postdata
 
 
 class CuriosityStreamBaseIE(InfoExtractor):
     _NETRC_MACHINE = 'curiositystream'
     _auth_token = None
-    _API_BASE_URL = 'https://api.curiositystream.com/v1/'
 
     def _handle_errors(self, result):
         error = result.get('error', {}).get('message')
@@ -27,6 +20,11 @@ def _handle_errors(self, result):
 
     def _call_api(self, path, video_id, query=None):
         headers = {}
+        if not self._auth_token:
+            auth_cookie = self._get_cookies('https://curiositystream.com').get('auth_token')
+            if auth_cookie:
+                self.write_debug('Obtained auth_token cookie')
+                self._auth_token = urllib.parse.unquote(auth_cookie.value)
         if self._auth_token:
             headers['X-Auth-Token'] = self._auth_token
         result = self._download_json(
@@ -34,13 +32,11 @@ def _call_api(self, path, video_id, query=None):
         self._handle_errors(result)
         return result['data']
 
-    def _real_initialize(self):
-        email, password = self._get_login_info()
-        if email is None:
-            return
+    def _perform_login(self, username, password):
         result = self._download_json(
-            self._API_BASE_URL + 'login', None, data=urlencode_postdata({
-                'email': email,
+            'https://api.curiositystream.com/v1/login', None,
+            note='Logging in', data=urlencode_postdata({
+                'email': username,
                 'password': password,
             }))
         self._handle_errors(result)
@@ -50,26 +46,35 @@ def _real_initialize(self):
 class CuriosityStreamIE(CuriosityStreamBaseIE):
     IE_NAME = 'curiositystream'
     _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/video/(?P<id>\d+)'
-    _TEST = {
-        'url': 'https://app.curiositystream.com/video/2',
+    _TESTS = [{
+        'url': 'http://app.curiositystream.com/video/2',
         'info_dict': {
             'id': '2',
             'ext': 'mp4',
             'title': 'How Did You Develop The Internet?',
             'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.',
+            'channel': 'Curiosity Stream',
+            'categories': ['Technology', 'Interview'],
+            'average_rating': float,
+            'series_id': '2',
+            'thumbnail': r're:https://img.curiositystream.com/.+\.jpg',
+            'tags': [],
+            'duration': 158
         },
         'params': {
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }]
+
+    _API_BASE_URL = 'https://api.curiositystream.com/v1/media/'
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
         formats = []
         for encoding_format in ('m3u8', 'mpd'):
-            media = self._call_api('media/' + video_id, video_id, query={
+            media = self._call_api(video_id, video_id, query={
                 'encodingsNew': 'true',
                 'encodingsFormat': encoding_format,
             })
@@ -116,7 +121,6 @@ def _real_extract(self, url):
                             'format_id': 'http',
                         })
                     formats.append(fmt)
-        self._sort_formats(formats)
 
         title = media['title']
 
@@ -139,6 +143,10 @@ def _real_extract(self, url):
             'duration': int_or_none(media.get('duration')),
             'tags': media.get('tags'),
             'subtitles': subtitles,
+            'channel': media.get('producer'),
+            'categories': [media.get('primary_category'), media.get('type')],
+            'average_rating': media.get('rating_percentage'),
+            'series_id': str(media.get('collection_id') or '') or None,
         }