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

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

diff --git a/yt_dlp/extractor/espn.py b/yt_dlp/extractor/espn.py
index 7aa454063ed70343b8b3739503f508035ed098c5..7ed824ca804b7a66ab519e36636dc98034b157c8 100644 (file)
--- a/yt_dlp/extractor/espn.py
+++ b/yt_dlp/extractor/espn.py
@@ -1,15 +1,16 @@
 import base64
 import json
 import re
-import urllib
+import urllib.parse
 
-from .common import InfoExtractor
 from .adobepass import AdobePassIE
+from .common import InfoExtractor
 from .once import OnceIE
 from ..utils import (
     determine_ext,
     dict_get,
     int_or_none,
+    traverse_obj,
     unified_strdate,
     unified_timestamp,
 )
@@ -161,7 +162,6 @@ def extract_source(source_url, source_id=None):
         links = clip.get('links', {})
         traverse_source(links.get('source', {}))
         traverse_source(links.get('mobile', {}))
-        self._sort_formats(formats)
 
         description = clip.get('caption') or clip.get('description')
         thumbnail = clip.get('thumbnail')
@@ -197,7 +197,7 @@ class ESPNArticleIE(InfoExtractor):
 
     @classmethod
     def suitable(cls, url):
-        return False if (ESPNIE.suitable(url) or WatchESPNIE.suitable(url)) else super(ESPNArticleIE, cls).suitable(url)
+        return False if (ESPNIE.suitable(url) or WatchESPNIE.suitable(url)) else super().suitable(url)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -240,7 +240,7 @@ def _real_extract(self, url):
 
 
 class ESPNCricInfoIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?espncricinfo\.com/video/[^#$&?/]+-(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?espncricinfo\.com/(?:cricket-)?videos?/[^#$&?/]+-(?P<id>\d+)'
     _TESTS = [{
         'url': 'https://www.espncricinfo.com/video/finch-chasing-comes-with-risks-despite-world-cup-trend-1289135',
         'info_dict': {
@@ -252,6 +252,17 @@ class ESPNCricInfoIE(InfoExtractor):
             'duration': 96,
         },
         'params': {'skip_download': True}
+    }, {
+        'url': 'https://www.espncricinfo.com/cricket-videos/daryl-mitchell-mitchell-santner-is-one-of-the-best-white-ball-spinners-india-vs-new-zealand-1356225',
+        'info_dict': {
+            'id': '1356225',
+            'ext': 'mp4',
+            'description': '"Santner has done it for a long time for New Zealand - we\'re lucky to have him"',
+            'upload_date': '20230128',
+            'title': 'Mitchell: \'Santner is one of the best white-ball spinners at the moment\'',
+            'duration': 87,
+        },
+        'params': {'skip_download': 'm3u8'},
     }]
 
     def _real_extract(self, url):
@@ -268,7 +279,6 @@ def _real_extract(self, url):
                     'url': item['url'],
                     'vcodec': 'none',
                 })
-        self._sort_formats(formats)
         return {
             'id': id,
             'title': data_json.get('title'),
@@ -281,29 +291,43 @@ def _real_extract(self, url):
 
 
 class WatchESPNIE(AdobePassIE):
-    _VALID_URL = r'https://www.espn.com/watch/player/_/id/(?P<id>[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12})'
+    _VALID_URL = r'https?://(?:www\.)?espn\.com/(?:watch|espnplus)/player/_/id/(?P<id>[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12})'
     _TESTS = [{
-        'url': 'https://www.espn.com/watch/player/_/id/ba7d17da-453b-4697-bf92-76a99f61642b',
+        'url': 'https://www.espn.com/watch/player/_/id/dbbc6b1d-c084-4b47-9878-5f13c56ce309',
         'info_dict': {
-            'id': 'ba7d17da-453b-4697-bf92-76a99f61642b',
+            'id': 'dbbc6b1d-c084-4b47-9878-5f13c56ce309',
             'ext': 'mp4',
-            'title': 'Serbia vs. Turkey',
-            'thumbnail': 'https://artwork.api.espn.com/artwork/collections/media/ba7d17da-453b-4697-bf92-76a99f61642b/default?width=640&apikey=1ngjw23osgcis1i1vbj96lmfqs',
+            'title': 'Huddersfield vs. Burnley',
+            'duration': 7500,
+            'thumbnail': 'https://artwork.api.espn.com/artwork/collections/media/dbbc6b1d-c084-4b47-9878-5f13c56ce309/default?width=640&apikey=1ngjw23osgcis1i1vbj96lmfqs',
         },
         'params': {
             'skip_download': True,
         },
     }, {
-        'url': 'https://www.espn.com/watch/player/_/id/4e9b5bd1-4ceb-4482-9d28-1dd5f30d2f34',
+        'url': 'https://www.espn.com/watch/player/_/id/a049a56e-a7ce-477e-aef3-c7e48ef8221c',
         'info_dict': {
-            'id': '4e9b5bd1-4ceb-4482-9d28-1dd5f30d2f34',
+            'id': 'a049a56e-a7ce-477e-aef3-c7e48ef8221c',
             'ext': 'mp4',
-            'title': 'Real Madrid vs. Real Betis (LaLiga)',
+            'title': 'Dynamo Dresden vs. VfB Stuttgart (Round #1) (German Cup)',
+            'duration': 8335,
             'thumbnail': 'https://s.secure.espncdn.com/stitcher/artwork/collections/media/bd1f3d12-0654-47d9-852e-71b85ea695c7/16x9.jpg?timestamp=202201112217&showBadge=true&cb=12&package=ESPN_PLUS',
         },
         'params': {
             'skip_download': True,
         },
+    }, {
+        'url': 'https://www.espn.com/espnplus/player/_/id/317f5fd1-c78a-4ebe-824a-129e0d348421',
+        'info_dict': {
+            'id': '317f5fd1-c78a-4ebe-824a-129e0d348421',
+            'ext': 'mp4',
+            'title': 'The Wheel - Episode 10',
+            'duration': 3352,
+            'thumbnail': 'https://s.secure.espncdn.com/stitcher/artwork/collections/media/317f5fd1-c78a-4ebe-824a-129e0d348421/16x9.jpg?timestamp=202205031523&showBadge=true&cb=12&package=ESPN_PLUS',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     _API_KEY = 'ZXNwbiZicm93c2VyJjEuMC4w.ptUt7QxsteaRruuPmGZFaJByOoqKvDP2a5YkInHrc7c'
@@ -317,9 +341,10 @@ def _call_bamgrid_api(self, path, video_id, payload=None, headers={}):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_data = self._download_json(
+        cdn_data = self._download_json(
             f'https://watch-cdn.product.api.espn.com/api/product/v3/watchespn/web/playback/event?id={video_id}',
-            video_id)['playbackState']
+            video_id)
+        video_data = cdn_data['playbackState']
 
         # ESPN+ subscription required, through cookies
         if 'DTC' in video_data.get('sourceId'):
@@ -384,10 +409,10 @@ def _real_extract(self, url):
             m3u8_url, headers = asset['stream'], {}
 
         formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, video_id, 'mp4', m3u8_id='hls')
-        self._sort_formats(formats)
 
         return {
             'id': video_id,
+            'duration': traverse_obj(cdn_data, ('tracking', 'duration')),
             'title': video_data.get('name'),
             'formats': formats,
             'subtitles': subtitles,