X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/5ad28e7ffd41deccba33776c1609aa7855601739..9f14daf22b4080ae1531a772ee7574959af4e2fa:/yt_dlp/extractor/tubitv.py

diff --git a/yt_dlp/extractor/tubitv.py b/yt_dlp/extractor/tubitv.py
index 2e9b325ba..de8b5da69 100644
--- a/yt_dlp/extractor/tubitv.py
+++ b/yt_dlp/extractor/tubitv.py
@@ -1,6 +1,3 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 import re
 
 from .common import InfoExtractor
@@ -10,6 +7,7 @@
     js_to_json,
     sanitized_Request,
     urlencode_postdata,
+    traverse_obj,
 )
 
 
@@ -24,6 +22,19 @@ class TubiTvIE(InfoExtractor):
     _NETRC_MACHINE = 'tubitv'
     _GEO_COUNTRIES = ['US']
     _TESTS = [{
+        'url': 'https://tubitv.com/movies/383676/tracker',
+        'md5': '566fa0f76870302d11af0de89511d3f0',
+        'info_dict': {
+            'id': '383676',
+            'ext': 'mp4',
+            'title': 'Tracker',
+            'description': 'md5:ff320baf43d0ad2655e538c1d5cd9706',
+            'uploader_id': 'f866e2677ea2f0dff719788e4f7f9195',
+            'release_year': 2010,
+            'thumbnail': r're:^https?://.+\.(jpe?g|png)$',
+            'duration': 6122,
+        },
+    }, {
         'url': 'http://tubitv.com/video/283829/the_comedian_at_the_friday',
         'md5': '43ac06be9326f41912dc64ccf7a80320',
         'info_dict': {
@@ -33,12 +44,10 @@ class TubiTvIE(InfoExtractor):
             'description': 'A stand up comedian is forced to look at the decisions in his life while on a one week trip to the west coast.',
             'uploader_id': 'bc168bee0d18dd1cb3b86c68706ab434',
         },
+        'skip': 'Content Unavailable'
     }, {
         'url': 'http://tubitv.com/tv-shows/321886/s01_e01_on_nom_stories',
         'only_matching': True,
-    }, {
-        'url': 'http://tubitv.com/movies/383676/tracker',
-        'only_matching': True,
     }, {
         'url': 'https://tubitv.com/movies/560057/penitentiary?start=true',
         'info_dict': {
@@ -49,15 +58,14 @@ class TubiTvIE(InfoExtractor):
             'uploader_id': 'd8fed30d4f24fcb22ec294421b9defc2',
             'release_year': 1979,
         },
-        'params': {
-            'skip_download': True,
-        },
+        'skip': 'Content Unavailable'
     }]
 
-    def _login(self):
-        username, password = self._get_login_info()
-        if username is None:
-            return
+    # DRM formats are included only to raise appropriate error
+    _UNPLAYABLE_FORMATS = ('hlsv6_widevine', 'hlsv6_widevine_nonclearlead', 'hlsv6_playready_psshv0',
+                           'hlsv6_fairplay', 'dash_widevine', 'dash_widevine_nonclearlead')
+
+    def _perform_login(self, username, password):
         self.report_login()
         form_data = {
             'username': username,
@@ -72,23 +80,28 @@ def _login(self):
             raise ExtractorError(
                 'Login failed (invalid username/password)', expected=True)
 
-    def _real_initialize(self):
-        self._login()
-
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_data = self._download_json(
-            'http://tubitv.com/oz/videos/%s/content' % video_id, video_id)
+        video_data = self._download_json(f'https://tubitv.com/oz/videos/{video_id}/content', video_id, query={
+            'video_resources': ['dash', 'hlsv3', 'hlsv6', *self._UNPLAYABLE_FORMATS],
+        })
         title = video_data['title']
 
         formats = []
-        url = video_data['url']
-        # URL can be sometimes empty. Does this only happen when there is DRM?
-        if url:
-            formats = self._extract_m3u8_formats(
-                self._proto_relative_url(url),
-                video_id, 'mp4', 'm3u8_native')
-        self._sort_formats(formats)
+        drm_formats = False
+
+        for resource in video_data['video_resources']:
+            if resource['type'] in ('dash', ):
+                formats += self._extract_mpd_formats(resource['manifest']['url'], video_id, mpd_id=resource['type'], fatal=False)
+            elif resource['type'] in ('hlsv3', 'hlsv6'):
+                formats += self._extract_m3u8_formats(resource['manifest']['url'], video_id, 'mp4', m3u8_id=resource['type'], fatal=False)
+            elif resource['type'] in self._UNPLAYABLE_FORMATS:
+                drm_formats = True
+
+        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', []):
@@ -107,6 +120,9 @@ def _real_extract(self, url):
                 'url': self._proto_relative_url(sub_url),
             })
 
+        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))
+
         return {
             'id': video_id,
             'title': title,
@@ -117,6 +133,9 @@ def _real_extract(self, url):
             '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
         }
 
 
@@ -132,10 +151,14 @@ class TubiTvShowIE(InfoExtractor):
 
     def _entries(self, show_url, show_name):
         show_webpage = self._download_webpage(show_url, show_name)
+
         show_json = self._parse_json(self._search_regex(
-            r"window\.__data\s*=\s*({.+?});\s*</script>",
-            show_webpage, 'data',), show_name, transform_source=js_to_json)['video']
+            r'window\.__data\s*=\s*({[^<]+});\s*</script>',
+            show_webpage, 'data'), show_name, transform_source=js_to_json)['video']
+
         for episode_id in show_json['fullContentById'].keys():
+            if traverse_obj(show_json, ('byId', episode_id, 'type')) == 's':
+                continue
             yield self.url_result(
                 'tubitv:%s' % episode_id,
                 ie=TubiTvIE.ie_key(), video_id=episode_id)