[ie/youtube] Add request_no_ads extractor-arg (#15145)

Default is `true` for unauthenticated users.
Default is `false` if logged-in cookies have been passed to yt-dlp.
Using `true` results in a loss of premium formats.

Closes #15144
Authored by: WhatAmISupposedToPutHere

README.md

@@ -1871,6 +1871,7 @@ The following extractors use this feature:
 * `pot_trace`: Enable debug logging for PO Token fetching. Either `true` or `false` (default)
 * `fetch_pot`: Policy to use for fetching a PO Token from providers. One of `always` (always try fetch a PO Token regardless if the client requires one for the given context), `never` (never fetch a PO Token), or `auto` (default; only fetch a PO Token if the client requires one for the given context)
 * `jsc_trace`: Enable debug logging for JS Challenge fetching. Either `true` or `false` (default)
+* `request_no_ads`: Skip preroll ads to eliminate the mandatory wait period before download. Either `true` (the default if unauthenticated) or `false`. The default is `false` when logged-in cookies have been passed to yt-dlp, since `true` will result in a loss of premium formats
 
 #### youtube-ejs
 * `jitless`: Run suported Javascript engines in JIT-less mode. Supported runtimes are `deno`, `node` and `bun`. Provides better security at the cost of performance/speed. Do note that `node` and `bun` are still considered unsecure. Either `true` or `false` (default)

yt_dlp/extractor/fc2.py

@@ -5,6 +5,7 @@ from .common import InfoExtractor
 from ..networking import Request
 from ..utils import (
     ExtractorError,
+    UserNotLive,
     js_to_json,
     traverse_obj,
     update_url_query,
@@ -205,6 +206,9 @@ class FC2LiveIE(InfoExtractor):
                 'client_app': 'browser_hls',
                 'ipv6': '',
             }), headers={'X-Requested-With': 'XMLHttpRequest'})
+        # A non-zero 'status' indicates the stream is not live, so check truthiness
+        if traverse_obj(control_server, ('status', {int})) and 'control_token' not in control_server:
+            raise UserNotLive(video_id=video_id)
         self._set_cookie('live.fc2.com', 'l_ortkn', control_server['orz_raw'])
 
         ws_url = update_url_query(control_server['url'], {'control_token': control_server['control_token']})

yt_dlp/extractor/patreon.py

@@ -598,7 +598,8 @@ class PatreonCampaignIE(PatreonBaseIE):
                 'props', 'pageProps', 'bootstrapEnvelope', 'pageBootstrap', 'campaign', 'data', 'id', {str}))
             if not campaign_id:
                 campaign_id = traverse_obj(self._search_nextjs_v13_data(webpage, vanity), (
-                    lambda _, v: v['type'] == 'campaign', 'id', {str}, any, {require('campaign ID')}))
+                    ((..., 'value', 'campaign', 'data'), lambda _, v: v['type'] == 'campaign'),
+                    'id', {str}, any, {require('campaign ID')}))
 
         params = {
             'json-api-use-default-includes': 'false',

yt_dlp/extractor/tubitv.py

@@ -182,13 +182,13 @@ class TubiTvShowIE(InfoExtractor):
         webpage = self._download_webpage(show_url, playlist_id)
 
         data = self._search_json(
-            r'window\.__data\s*=', webpage, 'data', playlist_id,
+            r'window\.__REACT_QUERY_STATE__\s*=', webpage, 'data', playlist_id,
-            transform_source=js_to_json)['video']
+            transform_source=js_to_json)['queries'][0]['state']['data']
 
         # v['number'] is already a decimal string, but stringify to protect against API changes
         path = [lambda _, v: str(v['number']) == selected_season] if selected_season else [..., {dict}]
 
-        for season in traverse_obj(data, ('byId', lambda _, v: v['type'] == 's', 'seasons', *path)):
+        for season in traverse_obj(data, ('seasons', *path)):
             season_number = int_or_none(season.get('number'))
             for episode in traverse_obj(season, ('episodes', lambda _, v: v['id'])):
                 episode_id = episode['id']

yt_dlp/extractor/youtube/_video.py

@@ -2628,18 +2628,29 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
     def _get_checkok_params():
         return {'contentCheckOk': True, 'racyCheckOk': True}
 
-    @classmethod
+    def _generate_player_context(self, sts=None):
-    def _generate_player_context(cls, sts=None):
         context = {
             'html5Preference': 'HTML5_PREF_WANTS',
         }
         if sts is not None:
             context['signatureTimestamp'] = sts
+
+        playback_context = {
+            'contentPlaybackContext': context,
+        }
+
+        # The 'adPlaybackContext'/'request_no_ads' workaround results in a loss of premium formats.
+        # Only default to 'true' if the user is unauthenticated, since we can't reliably detect all
+        # types of premium accounts (e.g. YTMusic Premium), and since premium users don't have ads.
+        default_arg_value = 'false' if self.is_authenticated else 'true'
+        if self._configuration_arg('request_no_ads', [default_arg_value])[0] != 'false':
+            playback_context['adPlaybackContext'] = {
+                'pyv': True,
+            }
+
         return {
-            'playbackContext': {
+            'playbackContext': playback_context,
-                'contentPlaybackContext': context,
+            **self._get_checkok_params(),
-            },
-            **cls._get_checkok_params(),
         }
 
     def _get_config_po_token(self, client: str, context: _PoTokenContext):
@@ -4029,6 +4040,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     STREAMING_DATA_CLIENT_NAME: client_name,
                 })
 
+        def set_audio_lang_from_orig_subs_lang(lang_code):
+            for f in formats:
+                if f.get('acodec') != 'none' and not f.get('language'):
+                    f['language'] = lang_code
+
         subtitles = {}
         skipped_subs_clients = set()
 
@@ -4088,7 +4104,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
                 orig_lang = qs.get('lang', [None])[-1]
                 lang_name = self._get_text(caption_track, 'name', max_runs=1)
-                if caption_track.get('kind') != 'asr':
+                is_manual_subs = caption_track.get('kind') != 'asr'
+                if is_manual_subs:
                     if not lang_code:
                         continue
                     process_language(
@@ -4099,16 +4116,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     if not trans_code:
                         continue
                     orig_trans_code = trans_code
-                    if caption_track.get('kind') != 'asr' and trans_code != 'und':
+                    if is_manual_subs and trans_code != 'und':
                         if not get_translated_subs:
                             continue
                         trans_code += f'-{lang_code}'
                         trans_name += format_field(lang_name, None, ' from %s')
                     if lang_code == f'a-{orig_trans_code}':
                         # Set audio language based on original subtitles
-                        for f in formats:
+                        set_audio_lang_from_orig_subs_lang(orig_trans_code)
-                            if f.get('acodec') != 'none' and not f.get('language'):
-                                f['language'] = orig_trans_code
                         # Add an "-orig" label to the original language so that it can be distinguished.
                         # The subs are returned without "-orig" as well for compatibility
                         process_language(
@@ -4119,6 +4134,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                         automatic_captions, base_url, trans_code, trans_name, client_name,
                         pot_params if orig_lang == orig_trans_code else {'tlang': trans_code, **pot_params})
 
+                # Extract automatic captions when the language is not in 'translationLanguages'
+                # e.g. Cantonese [yue], see https://github.com/yt-dlp/yt-dlp/issues/14889
+                lang_code = remove_start(lang_code, 'a-')
+                if is_manual_subs or not lang_code or lang_code in automatic_captions:
+                    continue
+                lang_name = remove_end(lang_name, ' (auto-generated)')
+                if caption_track.get('isTranslatable'):
+                    # We can assume this is the original audio language
+                    set_audio_lang_from_orig_subs_lang(lang_code)
+                    process_language(
+                        automatic_captions, base_url, f'{lang_code}-orig',
+                        f'{lang_name} (Original)', client_name, pot_params)
+                process_language(
+                    automatic_captions, base_url, lang_code, lang_name, client_name, pot_params)
+
             # Avoid duplication if we've already got everything we need
             need_subs_langs.difference_update(subtitles)
             need_caps_langs.difference_update(automatic_captions)