[ie/TubiTv] Fix extractor (#9975)

Closes #9937
Authored by: chilinux

yt_dlp/extractor/tubitv.py

@@ -7,33 +7,45 @@ from ..utils import (
     int_or_none,
     js_to_json,
     traverse_obj,
+    url_or_none,
     urlencode_postdata,
 )
 
 
 class TubiTvIE(InfoExtractor):
-    _VALID_URL = r'''(?x)
+    _VALID_URL = r'https?://(?:www\.)?tubitv\.com/(?P<type>video|movies|tv-shows)/(?P<id>\d+)'
-                    (?:
-                        tubitv:|
-                        https?://(?:www\.)?tubitv\.com/(?:video|movies|tv-shows)/
-                    )
-                    (?P<id>[0-9]+)'''
     _LOGIN_URL = 'http://tubitv.com/login'
     _NETRC_MACHINE = 'tubitv'
-    _GEO_COUNTRIES = ['US']
     _TESTS = [{
-        'url': 'https://tubitv.com/movies/383676/tracker',
+        'url': 'https://tubitv.com/movies/100004539/the-39-steps',
-        'md5': '566fa0f76870302d11af0de89511d3f0',
         'info_dict': {
-            'id': '383676',
+            'id': '100004539',
             'ext': 'mp4',
-            'title': 'Tracker',
+            'title': 'The 39 Steps',
-            'description': 'md5:ff320baf43d0ad2655e538c1d5cd9706',
+            'description': 'md5:bb2f2dd337f0dc58c06cb509943f54c8',
-            'uploader_id': 'f866e2677ea2f0dff719788e4f7f9195',
+            'uploader_id': 'abc2558d54505d4f0f32be94f2e7108c',
-            'release_year': 2010,
+            'release_year': 1935,
             'thumbnail': r're:^https?://.+\.(jpe?g|png)$',
-            'duration': 6122,
+            'duration': 5187,
         },
+        'params': {'skip_download': 'm3u8'},
+    }, {
+        'url': 'https://tubitv.com/tv-shows/554628/s01-e01-rise-of-the-snakes',
+        'info_dict': {
+            'id': '554628',
+            'ext': 'mp4',
+            'title': 'S01:E01 - Rise of the Snakes',
+            'description': 'md5:ba136f586de53af0372811e783a3f57d',
+            'episode': 'Rise of the Snakes',
+            'episode_number': 1,
+            'season': 'Season 1',
+            'season_number': 1,
+            'uploader_id': '2a9273e728c510d22aa5c57d0646810b',
+            'release_year': 2011,
+            'thumbnail': r're:^https?://.+\.(jpe?g|png)$',
+            'duration': 1376,
+        },
+        'params': {'skip_download': 'm3u8'},
     }, {
         'url': 'http://tubitv.com/video/283829/the_comedian_at_the_friday',
         'md5': '43ac06be9326f41912dc64ccf7a80320',
@@ -81,45 +93,39 @@ class TubiTvIE(InfoExtractor):
                 'Login failed (invalid username/password)', expected=True)
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        video_id, video_type = self._match_valid_url(url).group('id', 'type')
-        video_data = self._download_json(f'https://tubitv.com/oz/videos/{video_id}/content', video_id, query={
+        webpage = self._download_webpage(f'https://tubitv.com/{video_type}/{video_id}/', video_id)
-            'video_resources': ['dash', 'hlsv3', 'hlsv6', *self._UNPLAYABLE_FORMATS],
+        video_data = self._search_json(
-        })
+            r'window\.__data\s*=', webpage, 'data', video_id,
-        title = video_data['title']
+            transform_source=js_to_json)['video']['byId'][video_id]
 
         formats = []
         drm_formats = False
 
-        for resource in video_data['video_resources']:
+        for resource in traverse_obj(video_data, ('video_resources', lambda _, v: url_or_none(v['manifest']['url']))):
-            if resource['type'] in ('dash', ):
+            resource_type = resource.get('type')
-                formats += self._extract_mpd_formats(resource['manifest']['url'], video_id, mpd_id=resource['type'], fatal=False)
+            manifest_url = resource['manifest']['url']
-            elif resource['type'] in ('hlsv3', 'hlsv6'):
+            if resource_type == 'dash':
-                formats += self._extract_m3u8_formats(resource['manifest']['url'], video_id, 'mp4', m3u8_id=resource['type'], fatal=False)
+                formats.extend(self._extract_mpd_formats(manifest_url, video_id, mpd_id=resource_type, fatal=False))
-            elif resource['type'] in self._UNPLAYABLE_FORMATS:
+            elif resource_type in ('hlsv3', 'hlsv6'):
+                formats.extend(self._extract_m3u8_formats(manifest_url, video_id, 'mp4', m3u8_id=resource_type, fatal=False))
+            elif resource_type in self._UNPLAYABLE_FORMATS:
                 drm_formats = True
+            else:
+                self.report_warning(f'Skipping unknown resource type "{resource_type}"')
 
         if not formats and drm_formats:
             self.report_drm(video_id)
         elif not formats and not video_data.get('policy_match'):  # policy_match is False if content was removed
             raise ExtractorError('This content is currently unavailable', expected=True)
 
-        thumbnails = []
-        for thumbnail_url in video_data.get('thumbnails', []):
-            if not thumbnail_url:
-                continue
-            thumbnails.append({
-                'url': self._proto_relative_url(thumbnail_url),
-            })
-
         subtitles = {}
-        for sub in video_data.get('subtitles', []):
+        for sub in traverse_obj(video_data, ('subtitles', lambda _, v: url_or_none(v['url']))):
-            sub_url = sub.get('url')
-            if not sub_url:
-                continue
             subtitles.setdefault(sub.get('lang', 'English'), []).append({
-                'url': self._proto_relative_url(sub_url),
+                'url': self._proto_relative_url(sub['url']),
             })
 
+        title = traverse_obj(video_data, ('title', {str}))
         season_number, episode_number, episode_title = self._search_regex(
             r'^S(\d+):E(\d+) - (.+)', title, 'episode info', fatal=False, group=(1, 2, 3), default=(None, None, None))
 
@@ -128,18 +134,21 @@ class TubiTvIE(InfoExtractor):
             'title': title,
             'formats': formats,
             'subtitles': subtitles,
-            'thumbnails': thumbnails,
-            'description': video_data.get('description'),
-            'duration': int_or_none(video_data.get('duration')),
-            'uploader_id': video_data.get('publisher_id'),
-            'release_year': int_or_none(video_data.get('year')),
             'season_number': int_or_none(season_number),
             'episode_number': int_or_none(episode_number),
-            'episode_title': episode_title
+            'episode': episode_title,
+            **traverse_obj(video_data, {
+                'description': ('description', {str}),
+                'duration': ('duration', {int_or_none}),
+                'uploader_id': ('publisher_id', {str}),
+                'release_year': ('year', {int_or_none}),
+                'thumbnails': ('thumbnails', ..., {url_or_none}, {'url': {self._proto_relative_url}}),
+            }),
         }
 
 
 class TubiTvShowIE(InfoExtractor):
+    _WORKING = False
     _VALID_URL = r'https?://(?:www\.)?tubitv\.com/series/[0-9]+/(?P<show_name>[^/?#]+)'
     _TESTS = [{
         'url': 'https://tubitv.com/series/3936/the-joy-of-painting-with-bob-ross?start=true',
@@ -160,7 +169,7 @@ class TubiTvShowIE(InfoExtractor):
             if traverse_obj(show_json, ('byId', episode_id, 'type')) == 's':
                 continue
             yield self.url_result(
-                'tubitv:%s' % episode_id,
+                f'https://tubitv.com/tv-shows/{episode_id}/',
                 ie=TubiTvIE.ie_key(), video_id=episode_id)
 
     def _real_extract(self, url):