Merge 1cae3bf46d into 10fc719bc7

- Removes MildomClipIE, MildomIE, MildomUserVodIE, MildomVodIE
- Removes PokemonIE, PokemonWatchIE
- Removes VeohIE, VeohUserIE

Closes #3373, Closes #7059
Authored by: doe1080

yt_dlp/extractor/_extractors.py

@@ -946,6 +946,10 @@ from .kaltura import KalturaIE
 from .kankanews import KankaNewsIE
 from .karaoketv import KaraoketvIE
 from .kelbyone import KelbyOneIE
+from .kenh14 import (
+    Kenh14PlaylistIE,
+    Kenh14VideoIE,
+)
 from .khanacademy import (
     KhanAcademyIE,
     KhanAcademyUnitIE,
@@ -1135,12 +1139,6 @@ from .microsoftembed import (
     MicrosoftMediusIE,
 )
 from .microsoftstream import MicrosoftStreamIE
-from .mildom import (
-    MildomClipIE,
-    MildomIE,
-    MildomUserVodIE,
-    MildomVodIE,
-)
 from .minds import (
     MindsChannelIE,
     MindsGroupIE,
@@ -1559,10 +1557,6 @@ from .podbayfm import (
 )
 from .podchaser import PodchaserIE
 from .podomatic import PodomaticIE
-from .pokemon import (
-    PokemonIE,
-    PokemonWatchIE,
-)
 from .pokergo import (
     PokerGoCollectionIE,
     PokerGoIE,
@@ -2284,10 +2278,6 @@ from .utreon import UtreonIE
 from .varzesh3 import Varzesh3IE
 from .vbox7 import Vbox7IE
 from .veo import VeoIE
-from .veoh import (
-    VeohIE,
-    VeohUserIE,
-)
 from .vesti import VestiIE
 from .vevo import (
     VevoIE,

yt_dlp/extractor/chaturbate.py

@@ -79,7 +79,7 @@ class ChaturbateIE(InfoExtractor):
             'formats': self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4', live=True),
         }
 
-    def _extract_from_webpage(self, video_id, tld):
+    def _extract_from_html(self, video_id, tld):
         webpage = self._download_webpage(
             f'https://chaturbate.{tld}/{video_id}/', video_id,
             headers=self.geo_verification_headers(), impersonate=True)
@@ -151,4 +151,4 @@ class ChaturbateIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id, tld = self._match_valid_url(url).group('id', 'tld')
-        return self._extract_from_api(video_id, tld) or self._extract_from_webpage(video_id, tld)
+        return self._extract_from_api(video_id, tld) or self._extract_from_html(video_id, tld)

yt_dlp/extractor/kenh14.py

@@ -0,0 +1,160 @@
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    extract_attributes,
+    get_element_by_class,
+    get_element_html_by_attribute,
+    get_elements_html_by_class,
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    remove_start,
+    strip_or_none,
+    unescapeHTML,
+    update_url,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class Kenh14VideoIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.kenh14\.vn/(?:video/)?[\w-]+-(?P<id>[0-9]+)\.chn'
+    _TESTS = [{
+        'url': 'https://video.kenh14.vn/video/mo-hop-iphone-14-pro-max-nguon-unbox-therapy-316173.chn',
+        'md5': '1ed67f9c3a1e74acf15db69590cf6210',
+        'info_dict': {
+            'id': '316173',
+            'ext': 'mp4',
+            'title': 'Video mở hộp iPhone 14 Pro Max (Nguồn: Unbox Therapy)',
+            'description': 'Video mở hộp iPhone 14 Pro MaxVideo mở hộp iPhone 14 Pro Max (Nguồn: Unbox Therapy)',
+            'thumbnail': r're:^https?://videothumbs\.mediacdn\.vn/.*\.jpg$',
+            'tags': [],
+            'uploader': 'Unbox Therapy',
+            'upload_date': '20220517',
+            'view_count': int,
+            'duration': 722.86,
+            'timestamp': 1652764468,
+        },
+    }, {
+        'url': 'https://video.kenh14.vn/video-316174.chn',
+        'md5': '2b41877d2afaf4a3f487ceda8e5c7cbd',
+        'info_dict': {
+            'id': '316174',
+            'ext': 'mp4',
+            'title': 'Khoảnh khắc VĐV nằm gục khóc sau chiến thắng: 7 năm trời Việt Nam mới có HCV kiếm chém nữ, chỉ có 8 tháng để khổ luyện trước khi lên sàn đấu',
+            'description': 'md5:de86aa22e143e2b277bce8ec9c6f17dc',
+            'thumbnail': r're:^https?://videothumbs\.mediacdn\.vn/.*\.jpg$',
+            'tags': [],
+            'upload_date': '20220517',
+            'view_count': int,
+            'duration': 70.04,
+            'timestamp': 1652766021,
+        },
+    }, {
+        'url': 'https://video.kenh14.vn/0-344740.chn',
+        'md5': 'b843495d5e728142c8870c09b46df2a9',
+        'info_dict': {
+            'id': '344740',
+            'ext': 'mov',
+            'title': 'Kỳ Duyên đầy căng thẳng trong buổi ra quân đi Miss Universe, nghi thức tuyên thuệ lần đầu xuất hiện gây nhiều tranh cãi',
+            'description': 'md5:2a2dbb4a7397169fb21ee68f09160497',
+            'thumbnail': r're:^https?://kenh14cdn\.com/.*\.jpg$',
+            'tags': ['kỳ duyên', 'Kỳ Duyên tuyên thuệ', 'miss universe'],
+            'uploader': 'Quang Vũ',
+            'upload_date': '20241024',
+            'view_count': int,
+            'duration': 198.88,
+            'timestamp': 1729741590,
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        attrs = extract_attributes(get_element_html_by_attribute('type', 'VideoStream', webpage) or '')
+        direct_url = attrs['data-vid']
+
+        metadata = self._download_json(
+            'https://api.kinghub.vn/video/api/v1/detailVideoByGet?FileName={}'.format(
+                remove_start(direct_url, 'kenh14cdn.com/')), video_id, fatal=False)
+
+        formats = [{'url': f'https://{direct_url}', 'format_id': 'http', 'quality': 1}]
+        subtitles = {}
+        video_data = self._download_json(
+            f'https://{direct_url}.json', video_id, note='Downloading video data', fatal=False)
+        if hls_url := traverse_obj(video_data, ('hls', {url_or_none})):
+            fmts, subs = self._extract_m3u8_formats_and_subtitles(
+                hls_url, video_id, m3u8_id='hls', fatal=False)
+            formats.extend(fmts)
+            self._merge_subtitles(subs, target=subtitles)
+        if dash_url := traverse_obj(video_data, ('mpd', {url_or_none})):
+            fmts, subs = self._extract_mpd_formats_and_subtitles(
+                dash_url, video_id, mpd_id='dash', fatal=False)
+            formats.extend(fmts)
+            self._merge_subtitles(subs, target=subtitles)
+
+        return {
+            **traverse_obj(metadata, {
+                'duration': ('duration', {parse_duration}),
+                'uploader': ('author', {strip_or_none}),
+                'timestamp': ('uploadtime', {parse_iso8601(delimiter=' ')}),
+                'view_count': ('views', {int_or_none}),
+            }),
+            'id': video_id,
+            'title': (
+                traverse_obj(metadata, ('title', {strip_or_none}))
+                or clean_html(self._og_search_title(webpage))
+                or clean_html(get_element_by_class('vdbw-title', webpage))),
+            'formats': formats,
+            'subtitles': subtitles,
+            'description': (
+                clean_html(self._og_search_description(webpage))
+                or clean_html(get_element_by_class('vdbw-sapo', webpage))),
+            'thumbnail': (self._og_search_thumbnail(webpage) or attrs.get('data-thumb')),
+            'tags': traverse_obj(self._html_search_meta('keywords', webpage), (
+                {lambda x: x.split(';')}, ..., filter)),
+        }
+
+
+class Kenh14PlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.kenh14\.vn/playlist/[\w-]+-(?P<id>[0-9]+)\.chn'
+    _TESTS = [{
+        'url': 'https://video.kenh14.vn/playlist/tran-tinh-naked-love-mua-2-71.chn',
+        'info_dict': {
+            'id': '71',
+            'title': 'Trần Tình (Naked love) mùa 2',
+            'description': 'md5:e9522339304956dea931722dd72eddb2',
+            'thumbnail': r're:^https?://kenh14cdn\.com/.*\.png$',
+        },
+        'playlist_count': 9,
+    }, {
+        'url': 'https://video.kenh14.vn/playlist/0-72.chn',
+        'info_dict': {
+            'id': '72',
+            'title': 'Lau Lại Đầu Từ',
+            'description': 'Cùng xem xưa và nay có gì khác biệt nhé!',
+            'thumbnail': r're:^https?://kenh14cdn\.com/.*\.png$',
+        },
+        'playlist_count': 6,
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+        webpage = self._download_webpage(url, playlist_id)
+
+        category_detail = get_element_by_class('category-detail', webpage) or ''
+        embed_info = traverse_obj(
+            self._yield_json_ld(webpage, playlist_id),
+            (lambda _, v: v['name'] and v['alternateName'], any)) or {}
+
+        return self.playlist_from_matches(
+            get_elements_html_by_class('video-item', webpage), playlist_id,
+            (clean_html(get_element_by_class('name', category_detail)) or unescapeHTML(embed_info.get('name'))),
+            getter=lambda x: 'https://video.kenh14.vn/video/video-{}.chn'.format(extract_attributes(x)['data-id']),
+            ie=Kenh14VideoIE, playlist_description=(
+                clean_html(get_element_by_class('description', category_detail))
+                or unescapeHTML(embed_info.get('alternateName'))),
+            thumbnail=traverse_obj(
+                self._og_search_thumbnail(webpage),
+                ({url_or_none}, {update_url(query=None)})))

yt_dlp/extractor/mildom.py

@@ -1,291 +0,0 @@
-import functools
-import json
-import uuid
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    OnDemandPagedList,
-    determine_ext,
-    dict_get,
-    float_or_none,
-    traverse_obj,
-)
-
-
-class MildomBaseIE(InfoExtractor):
-    _GUEST_ID = None
-
-    def _call_api(self, url, video_id, query=None, note='Downloading JSON metadata', body=None):
-        if not self._GUEST_ID:
-            self._GUEST_ID = f'pc-gp-{uuid.uuid4()}'
-
-        content = self._download_json(
-            url, video_id, note=note, data=json.dumps(body).encode() if body else None,
-            headers={'Content-Type': 'application/json'} if body else {},
-            query={
-                '__guest_id': self._GUEST_ID,
-                '__platform': 'web',
-                **(query or {}),
-            })
-
-        if content['code'] != 0:
-            raise ExtractorError(
-                f'Mildom says: {content["message"]} (code {content["code"]})',
-                expected=True)
-        return content['body']
-
-
-class MildomIE(MildomBaseIE):
-    IE_NAME = 'mildom'
-    IE_DESC = 'Record ongoing live by specific user in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/(?P<id>\d+)'
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(f'https://www.mildom.com/{video_id}', video_id)
-
-        enterstudio = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/live/enterstudio', video_id,
-            note='Downloading live metadata', query={'user_id': video_id})
-        result_video_id = enterstudio.get('log_id', video_id)
-
-        servers = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/live/liveserver', result_video_id,
-            note='Downloading live server list', query={
-                'user_id': video_id,
-                'live_server_type': 'hls',
-            })
-
-        playback_token = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/live/token', result_video_id,
-            note='Obtaining live playback token', body={'host_id': video_id, 'type': 'hls'})
-        playback_token = traverse_obj(playback_token, ('data', ..., 'token'), get_all=False)
-        if not playback_token:
-            raise ExtractorError('Failed to obtain live playback token')
-
-        formats = self._extract_m3u8_formats(
-            f'{servers["stream_server"]}/{video_id}_master.m3u8?{playback_token}',
-            result_video_id, 'mp4', headers={
-                'Referer': 'https://www.mildom.com/',
-                'Origin': 'https://www.mildom.com',
-            })
-
-        for fmt in formats:
-            fmt.setdefault('http_headers', {})['Referer'] = 'https://www.mildom.com/'
-
-        return {
-            'id': result_video_id,
-            'title': self._html_search_meta('twitter:description', webpage, default=None) or traverse_obj(enterstudio, 'anchor_intro'),
-            'description': traverse_obj(enterstudio, 'intro', 'live_intro', expected_type=str),
-            'timestamp': float_or_none(enterstudio.get('live_start_ms'), scale=1000),
-            'uploader': self._html_search_meta('twitter:title', webpage, default=None) or traverse_obj(enterstudio, 'loginname'),
-            'uploader_id': video_id,
-            'formats': formats,
-            'is_live': True,
-        }
-
-
-class MildomVodIE(MildomBaseIE):
-    IE_NAME = 'mildom:vod'
-    IE_DESC = 'VOD in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/playback/(?P<user_id>\d+)/(?P<id>(?P=user_id)-[a-zA-Z0-9]+-?[0-9]*)'
-    _TESTS = [{
-        'url': 'https://www.mildom.com/playback/10882672/10882672-1597662269',
-        'info_dict': {
-            'id': '10882672-1597662269',
-            'ext': 'mp4',
-            'title': '始めてのミルダム配信じゃぃ！',
-            'thumbnail': r're:^https?://.*\.(png|jpg)$',
-            'upload_date': '20200817',
-            'duration': 4138.37,
-            'description': 'ゲームをしたくて！',
-            'timestamp': 1597662269.0,
-            'uploader_id': '10882672',
-            'uploader': 'kson組長(けいそん)',
-        },
-    }, {
-        'url': 'https://www.mildom.com/playback/10882672/10882672-1597758589870-477',
-        'info_dict': {
-            'id': '10882672-1597758589870-477',
-            'ext': 'mp4',
-            'title': '【kson】感染メイズ！麻酔銃で無双する',
-            'thumbnail': r're:^https?://.*\.(png|jpg)$',
-            'timestamp': 1597759093.0,
-            'uploader': 'kson組長(けいそん)',
-            'duration': 4302.58,
-            'uploader_id': '10882672',
-            'description': 'このステージ絶対乗り越えたい',
-            'upload_date': '20200818',
-        },
-    }, {
-        'url': 'https://www.mildom.com/playback/10882672/10882672-buha9td2lrn97fk2jme0',
-        'info_dict': {
-            'id': '10882672-buha9td2lrn97fk2jme0',
-            'ext': 'mp4',
-            'title': '【kson組長】CART RACER!!!',
-            'thumbnail': r're:^https?://.*\.(png|jpg)$',
-            'uploader_id': '10882672',
-            'uploader': 'kson組長(けいそん)',
-            'upload_date': '20201104',
-            'timestamp': 1604494797.0,
-            'duration': 4657.25,
-            'description': 'WTF',
-        },
-    }]
-
-    def _real_extract(self, url):
-        user_id, video_id = self._match_valid_url(url).group('user_id', 'id')
-        webpage = self._download_webpage(f'https://www.mildom.com/playback/{user_id}/{video_id}', video_id)
-
-        autoplay = self._call_api(
-            'https://cloudac.mildom.com/nonolive/videocontent/playback/getPlaybackDetail', video_id,
-            note='Downloading playback metadata', query={
-                'v_id': video_id,
-            })['playback']
-
-        formats = [{
-            'url': autoplay['audio_url'],
-            'format_id': 'audio',
-            'protocol': 'm3u8_native',
-            'vcodec': 'none',
-            'acodec': 'aac',
-            'ext': 'm4a',
-        }]
-        for fmt in autoplay['video_link']:
-            formats.append({
-                'format_id': 'video-{}'.format(fmt['name']),
-                'url': fmt['url'],
-                'protocol': 'm3u8_native',
-                'width': fmt['level'] * autoplay['video_width'] // autoplay['video_height'],
-                'height': fmt['level'],
-                'vcodec': 'h264',
-                'acodec': 'aac',
-                'ext': 'mp4',
-            })
-
-        return {
-            'id': video_id,
-            'title': self._html_search_meta(('og:description', 'description'), webpage, default=None) or autoplay.get('title'),
-            'description': traverse_obj(autoplay, 'video_intro'),
-            'timestamp': float_or_none(autoplay.get('publish_time'), scale=1000),
-            'duration': float_or_none(autoplay.get('video_length'), scale=1000),
-            'thumbnail': dict_get(autoplay, ('upload_pic', 'video_pic')),
-            'uploader': traverse_obj(autoplay, ('author_info', 'login_name')),
-            'uploader_id': user_id,
-            'formats': formats,
-        }
-
-
-class MildomClipIE(MildomBaseIE):
-    IE_NAME = 'mildom:clip'
-    IE_DESC = 'Clip in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/clip/(?P<id>(?P<user_id>\d+)-[a-zA-Z0-9]+)'
-    _TESTS = [{
-        'url': 'https://www.mildom.com/clip/10042245-63921673e7b147ebb0806d42b5ba5ce9',
-        'info_dict': {
-            'id': '10042245-63921673e7b147ebb0806d42b5ba5ce9',
-            'title': '全然違ったよ',
-            'timestamp': 1619181890,
-            'duration': 59,
-            'thumbnail': r're:https?://.+',
-            'uploader': 'ざきんぽ',
-            'uploader_id': '10042245',
-        },
-    }, {
-        'url': 'https://www.mildom.com/clip/10111524-ebf4036e5aa8411c99fb3a1ae0902864',
-        'info_dict': {
-            'id': '10111524-ebf4036e5aa8411c99fb3a1ae0902864',
-            'title': 'かっこいい',
-            'timestamp': 1621094003,
-            'duration': 59,
-            'thumbnail': r're:https?://.+',
-            'uploader': '(ルーキー',
-            'uploader_id': '10111524',
-        },
-    }, {
-        'url': 'https://www.mildom.com/clip/10660174-2c539e6e277c4aaeb4b1fbe8d22cb902',
-        'info_dict': {
-            'id': '10660174-2c539e6e277c4aaeb4b1fbe8d22cb902',
-            'title': 'あ',
-            'timestamp': 1614769431,
-            'duration': 31,
-            'thumbnail': r're:https?://.+',
-            'uploader': 'ドルゴルスレンギーン＝ダグワドルジ',
-            'uploader_id': '10660174',
-        },
-    }]
-
-    def _real_extract(self, url):
-        user_id, video_id = self._match_valid_url(url).group('user_id', 'id')
-        webpage = self._download_webpage(f'https://www.mildom.com/clip/{video_id}', video_id)
-
-        clip_detail = self._call_api(
-            'https://cloudac-cf-jp.mildom.com/nonolive/videocontent/clip/detail', video_id,
-            note='Downloading playback metadata', query={
-                'clip_id': video_id,
-            })
-
-        return {
-            'id': video_id,
-            'title': self._html_search_meta(
-                ('og:description', 'description'), webpage, default=None) or clip_detail.get('title'),
-            'timestamp': float_or_none(clip_detail.get('create_time')),
-            'duration': float_or_none(clip_detail.get('length')),
-            'thumbnail': clip_detail.get('cover'),
-            'uploader': traverse_obj(clip_detail, ('user_info', 'loginname')),
-            'uploader_id': user_id,
-
-            'url': clip_detail['url'],
-            'ext': determine_ext(clip_detail.get('url'), 'mp4'),
-        }
-
-
-class MildomUserVodIE(MildomBaseIE):
-    IE_NAME = 'mildom:user:vod'
-    IE_DESC = 'Download all VODs from specific user in Mildom'
-    _VALID_URL = r'https?://(?:(?:www|m)\.)mildom\.com/profile/(?P<id>\d+)'
-    _TESTS = [{
-        'url': 'https://www.mildom.com/profile/10093333',
-        'info_dict': {
-            'id': '10093333',
-            'title': 'Uploads from ねこばたけ',
-        },
-        'playlist_mincount': 732,
-    }, {
-        'url': 'https://www.mildom.com/profile/10882672',
-        'info_dict': {
-            'id': '10882672',
-            'title': 'Uploads from kson組長(けいそん)',
-        },
-        'playlist_mincount': 201,
-    }]
-
-    def _fetch_page(self, user_id, page):
-        page += 1
-        reply = self._call_api(
-            'https://cloudac.mildom.com/nonolive/videocontent/profile/playbackList',
-            user_id, note=f'Downloading page {page}', query={
-                'user_id': user_id,
-                'page': page,
-                'limit': '30',
-            })
-        if not reply:
-            return
-        for x in reply:
-            v_id = x.get('v_id')
-            if not v_id:
-                continue
-            yield self.url_result(f'https://www.mildom.com/playback/{user_id}/{v_id}')
-
-    def _real_extract(self, url):
-        user_id = self._match_id(url)
-        self.to_screen(f'This will download all VODs belonging to user. To download ongoing live video, use "https://www.mildom.com/{user_id}" instead')
-
-        profile = self._call_api(
-            'https://cloudac.mildom.com/nonolive/gappserv/user/profileV2', user_id,
-            query={'user_id': user_id}, note='Downloading user profile')['user_info']
-
-        return self.playlist_result(
-            OnDemandPagedList(functools.partial(self._fetch_page, user_id), 30),
-            user_id, f'Uploads from {profile["loginname"]}')

yt_dlp/extractor/pokemon.py

@@ -1,136 +0,0 @@
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    extract_attributes,
-    int_or_none,
-    js_to_json,
-    merge_dicts,
-)
-
-
-class PokemonIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?pokemon\.com/[a-z]{2}(?:.*?play=(?P<id>[a-z0-9]{32})|/(?:[^/]+/)+(?P<display_id>[^/?#&]+))'
-    _TESTS = [{
-        'url': 'https://www.pokemon.com/us/pokemon-episodes/20_30-the-ol-raise-and-switch/',
-        'md5': '2fe8eaec69768b25ef898cda9c43062e',
-        'info_dict': {
-            'id': 'afe22e30f01c41f49d4f1d9eab5cd9a4',
-            'ext': 'mp4',
-            'title': 'The Ol’ Raise and Switch!',
-            'description': 'md5:7db77f7107f98ba88401d3adc80ff7af',
-        },
-        'add_id': ['LimelightMedia'],
-    }, {
-        # no data-video-title
-        'url': 'https://www.pokemon.com/fr/episodes-pokemon/films-pokemon/pokemon-lascension-de-darkrai-2008',
-        'info_dict': {
-            'id': 'dfbaf830d7e54e179837c50c0c6cc0e1',
-            'ext': 'mp4',
-            'title': "Pokémon : L'ascension de Darkrai",
-            'description': 'md5:d1dbc9e206070c3e14a06ff557659fb5',
-        },
-        'add_id': ['LimelightMedia'],
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://www.pokemon.com/uk/pokemon-episodes/?play=2e8b5c761f1d4a9286165d7748c1ece2',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.pokemon.com/fr/episodes-pokemon/18_09-un-hiver-inattendu/',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.pokemon.com/de/pokemon-folgen/01_20-bye-bye-smettbo/',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        video_id, display_id = self._match_valid_url(url).groups()
-        webpage = self._download_webpage(url, video_id or display_id)
-        video_data = extract_attributes(self._search_regex(
-            r'(<[^>]+data-video-id="{}"[^>]*>)'.format(video_id if video_id else '[a-z0-9]{32}'),
-            webpage, 'video data element'))
-        video_id = video_data['data-video-id']
-        title = video_data.get('data-video-title') or self._html_search_meta(
-            'pkm-title', webpage, ' title', default=None) or self._search_regex(
-            r'<h1[^>]+\bclass=["\']us-title[^>]+>([^<]+)', webpage, 'title')
-        return {
-            '_type': 'url_transparent',
-            'id': video_id,
-            'url': f'limelight:media:{video_id}',
-            'title': title,
-            'description': video_data.get('data-video-summary'),
-            'thumbnail': video_data.get('data-video-poster'),
-            'series': 'Pokémon',
-            'season_number': int_or_none(video_data.get('data-video-season')),
-            'episode': title,
-            'episode_number': int_or_none(video_data.get('data-video-episode')),
-            'ie_key': 'LimelightMedia',
-        }
-
-
-class PokemonWatchIE(InfoExtractor):
-    _VALID_URL = r'https?://watch\.pokemon\.com/[a-z]{2}-[a-z]{2}/(?:#/)?player(?:\.html)?\?id=(?P<id>[a-z0-9]{32})'
-    _API_URL = 'https://www.pokemon.com/api/pokemontv/v2/channels/{0:}'
-    _TESTS = [{
-        'url': 'https://watch.pokemon.com/en-us/player.html?id=8309a40969894a8e8d5bc1311e9c5667',
-        'md5': '62833938a31e61ab49ada92f524c42ff',
-        'info_dict': {
-            'id': '8309a40969894a8e8d5bc1311e9c5667',
-            'ext': 'mp4',
-            'title': 'Lillier and the Staff!',
-            'description': 'md5:338841b8c21b283d24bdc9b568849f04',
-        },
-    }, {
-        'url': 'https://watch.pokemon.com/en-us/#/player?id=3fe7752ba09141f0b0f7756d1981c6b2',
-        'only_matching': True,
-    }, {
-        'url': 'https://watch.pokemon.com/de-de/player.html?id=b3c402e111a4459eb47e12160ab0ba07',
-        'only_matching': True,
-    }]
-
-    def _extract_media(self, channel_array, video_id):
-        for channel in channel_array:
-            for media in channel.get('media'):
-                if media.get('id') == video_id:
-                    return media
-        return None
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        info = {
-            '_type': 'url',
-            'id': video_id,
-            'url': f'limelight:media:{video_id}',
-            'ie_key': 'LimelightMedia',
-        }
-
-        # API call can be avoided entirely if we are listing formats
-        if self.get_param('listformats', False):
-            return info
-
-        webpage = self._download_webpage(url, video_id)
-        build_vars = self._parse_json(self._search_regex(
-            r'(?s)buildVars\s*=\s*({.*?})', webpage, 'build vars'),
-            video_id, transform_source=js_to_json)
-        region = build_vars.get('region')
-        channel_array = self._download_json(self._API_URL.format(region), video_id)
-        video_data = self._extract_media(channel_array, video_id)
-
-        if video_data is None:
-            raise ExtractorError(
-                f'Video {video_id} does not exist', expected=True)
-
-        info['_type'] = 'url_transparent'
-        images = video_data.get('images')
-
-        return merge_dicts(info, {
-            'title': video_data.get('title'),
-            'description': video_data.get('description'),
-            'thumbnail': images.get('medium') or images.get('small'),
-            'series': 'Pokémon',
-            'season_number': int_or_none(video_data.get('season')),
-            'episode': video_data.get('title'),
-            'episode_number': int_or_none(video_data.get('episode')),
-        })

yt_dlp/extractor/veoh.py

@@ -1,189 +0,0 @@
-import functools
-import json
-
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    OnDemandPagedList,
-    int_or_none,
-    parse_duration,
-    qualities,
-    remove_start,
-    strip_or_none,
-)
-
-
-class VeohIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?veoh\.com/(?:watch|videos|embed|iphone/#_Watch)/(?P<id>(?:v|e|yapi-)[\da-zA-Z]+)'
-
-    _TESTS = [{
-        'url': 'http://www.veoh.com/watch/v56314296nk7Zdmz3',
-        'md5': '620e68e6a3cff80086df3348426c9ca3',
-        'info_dict': {
-            'id': 'v56314296nk7Zdmz3',
-            'ext': 'mp4',
-            'title': 'Straight Backs Are Stronger',
-            'description': 'md5:203f976279939a6dc664d4001e13f5f4',
-            'thumbnail': 're:https://fcache\\.veoh\\.com/file/f/th56314296\\.jpg(\\?.*)?',
-            'uploader': 'LUMOback',
-            'duration': 46,
-            'view_count': int,
-            'average_rating': int,
-            'comment_count': int,
-            'age_limit': 0,
-            'categories': ['technology_and_gaming'],
-            'tags': ['posture', 'posture', 'sensor', 'back', 'pain', 'wearable', 'tech', 'lumo'],
-        },
-    }, {
-        'url': 'http://www.veoh.com/embed/v56314296nk7Zdmz3',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.veoh.com/watch/v27701988pbTc4wzN?h1=Chile+workers+cover+up+to+avoid+skin+damage',
-        'md5': '4a6ff84b87d536a6a71e6aa6c0ad07fa',
-        'info_dict': {
-            'id': '27701988',
-            'ext': 'mp4',
-            'title': 'Chile workers cover up to avoid skin damage',
-            'description': 'md5:2bd151625a60a32822873efc246ba20d',
-            'uploader': 'afp-news',
-            'duration': 123,
-        },
-        'skip': 'This video has been deleted.',
-    }, {
-        'url': 'http://www.veoh.com/watch/v69525809F6Nc4frX',
-        'md5': '4fde7b9e33577bab2f2f8f260e30e979',
-        'note': 'Embedded ooyala video',
-        'info_dict': {
-            'id': '69525809',
-            'ext': 'mp4',
-            'title': 'Doctors Alter Plan For Preteen\'s Weight Loss Surgery',
-            'description': 'md5:f5a11c51f8fb51d2315bca0937526891',
-            'uploader': 'newsy-videos',
-        },
-        'skip': 'This video has been deleted.',
-    }, {
-        'url': 'http://www.veoh.com/watch/e152215AJxZktGS',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.veoh.com/videos/v16374379WA437rMH',
-        'md5': 'cceb73f3909063d64f4b93d4defca1b3',
-        'info_dict': {
-            'id': 'v16374379WA437rMH',
-            'ext': 'mp4',
-            'title': 'Phantasmagoria 2, pt. 1-3',
-            'description': 'Phantasmagoria: a Puzzle of Flesh',
-            'thumbnail': 're:https://fcache\\.veoh\\.com/file/f/th16374379\\.jpg(\\?.*)?',
-            'uploader': 'davidspackage',
-            'duration': 968,
-            'view_count': int,
-            'average_rating': int,
-            'comment_count': int,
-            'age_limit': 18,
-            'categories': ['technology_and_gaming', 'gaming'],
-            'tags': ['puzzle', 'of', 'flesh'],
-        },
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        metadata = self._download_json(
-            'https://www.veoh.com/watch/getVideo/' + video_id,
-            video_id)
-        video = metadata['video']
-        title = video['title']
-
-        thumbnail_url = None
-        q = qualities(['Regular', 'HQ'])
-        formats = []
-        for f_id, f_url in video.get('src', {}).items():
-            if not f_url:
-                continue
-            if f_id == 'poster':
-                thumbnail_url = f_url
-            else:
-                formats.append({
-                    'format_id': f_id,
-                    'quality': q(f_id),
-                    'url': f_url,
-                })
-
-        categories = metadata.get('categoryPath')
-        if not categories:
-            category = remove_start(strip_or_none(video.get('category')), 'category_')
-            categories = [category] if category else None
-        tags = video.get('tags')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': video.get('description'),
-            'thumbnail': thumbnail_url,
-            'uploader': video.get('author', {}).get('nickname'),
-            'duration': int_or_none(video.get('lengthBySec')) or parse_duration(video.get('length')),
-            'view_count': int_or_none(video.get('views')),
-            'formats': formats,
-            'average_rating': int_or_none(video.get('rating')),
-            'comment_count': int_or_none(video.get('numOfComments')),
-            'age_limit': 18 if video.get('contentRatingId') == 2 else 0,
-            'categories': categories,
-            'tags': tags.split(', ') if tags else None,
-        }
-
-
-class VeohUserIE(VeohIE):  # XXX: Do not subclass from concrete IE
-    _VALID_URL = r'https?://(?:www\.)?veoh\.com/users/(?P<id>[\w-]+)'
-    IE_NAME = 'veoh:user'
-
-    _TESTS = [
-        {
-            'url': 'https://www.veoh.com/users/valentinazoe',
-            'info_dict': {
-                'id': 'valentinazoe',
-                'title': 'valentinazoe (Uploads)',
-            },
-            'playlist_mincount': 75,
-        },
-        {
-            'url': 'https://www.veoh.com/users/PiensaLibre',
-            'info_dict': {
-                'id': 'PiensaLibre',
-                'title': 'PiensaLibre (Uploads)',
-            },
-            'playlist_mincount': 2,
-        }]
-
-    _PAGE_SIZE = 16
-
-    def _fetch_page(self, uploader, page):
-        response = self._download_json(
-            'https://www.veoh.com/users/published/videos', uploader,
-            note=f'Downloading videos page {page + 1}',
-            headers={
-                'x-csrf-token': self._TOKEN,
-                'content-type': 'application/json;charset=UTF-8',
-            },
-            data=json.dumps({
-                'username': uploader,
-                'maxResults': self._PAGE_SIZE,
-                'page': page + 1,
-                'requestName': 'userPage',
-            }).encode())
-        if not response.get('success'):
-            raise ExtractorError(response['message'])
-
-        for video in response['videos']:
-            yield self.url_result(f'https://www.veoh.com/watch/{video["permalinkId"]}', VeohIE,
-                                  video['permalinkId'], video.get('title'))
-
-    def _real_initialize(self):
-        webpage = self._download_webpage(
-            'https://www.veoh.com', None, note='Downloading authorization token')
-        self._TOKEN = self._search_regex(
-            r'csrfToken:\s*(["\'])(?P<token>[0-9a-zA-Z]{40})\1', webpage,
-            'request token', group='token')
-
-    def _real_extract(self, url):
-        uploader = self._match_id(url)
-        return self.playlist_result(OnDemandPagedList(
-            functools.partial(self._fetch_page, uploader),
-            self._PAGE_SIZE), uploader, f'{uploader} (Uploads)')

yt_dlp/postprocessor/ffmpeg.py

@@ -219,9 +219,20 @@ class FFmpegPostProcessor(PostProcessor):
     @staticmethod
     def stream_copy_opts(copy=True, *, ext=None):
         yield from ('-map', '0')
+
+        if ext in ('mkv', 'mka'):
+            # Some streams, such as JSON attachments, are considered of unknown
+            # type by FFmpeg but we still want to copy them.
+            yield '-copy_unknown'
+        else:
+            # Most containers don't really like unknown streams. Let's make
+            # sure to get rid of them.
+            yield '-ignore_unknown'
+
         # Don't copy Apple TV chapters track, bin_data
         # See https://github.com/yt-dlp/yt-dlp/issues/2, #19042, #19024, https://trac.ffmpeg.org/ticket/6016
-        yield from ('-dn', '-ignore_unknown')
+        yield '-dn'
+
         if copy:
             yield from ('-c', 'copy')
         if ext in ('mp4', 'mov', 'm4a'):
@@ -556,7 +567,7 @@ class FFmpegVideoConvertorPP(FFmpegPostProcessor):
 
     @staticmethod
     def _options(target_ext):
-        yield from FFmpegPostProcessor.stream_copy_opts(False)
+        yield from FFmpegPostProcessor.stream_copy_opts(False, ext=target_ext)
         if target_ext == 'avi':
             yield from ('-c:v', 'libxvid', '-vtag', 'XVID')
 
@@ -582,7 +593,7 @@ class FFmpegVideoRemuxerPP(FFmpegVideoConvertorPP):
 
     @staticmethod
     def _options(target_ext):
-        return FFmpegPostProcessor.stream_copy_opts()
+        return FFmpegPostProcessor.stream_copy_opts(ext=target_ext)
 
 
 class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
@@ -619,13 +630,18 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
         webm_vtt_warn = False
         mp4_ass_warn = False
 
+        json_subs = {}
+
         for lang, sub_info in subtitles.items():
             if not os.path.exists(sub_info.get('filepath', '')):
                 self.report_warning(f'Skipping embedding {lang} subtitle because the file is missing')
                 continue
             sub_ext = sub_info['ext']
             if sub_ext == 'json':
-                self.report_warning('JSON subtitles cannot be embedded')
+                if info['ext'] in ('mkv', 'mka'):
+                    json_subs[lang] = sub_info['filepath']
+                else:
+                    self.report_warning('JSON subtitles can only be embedded in mkv/mka files.')
             elif ext != 'webm' or ext == 'webm' and sub_ext == 'vtt':
                 sub_langs.append(lang)
                 sub_names.append(sub_info.get('name'))
@@ -638,31 +654,48 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
                 mp4_ass_warn = True
                 self.report_warning('ASS subtitles cannot be properly embedded in mp4 files; expect issues')
 
-        if not sub_langs:
+        if not sub_langs and not json_subs:
             return [], info
 
         input_files = [filename, *sub_filenames]
 
-        opts = [
-            *self.stream_copy_opts(ext=info['ext']),
-            # Don't copy the existing subtitles, we may be running the
-            # postprocessor a second time
-            '-map', '-0:s',
-        ]
+        opts = [*self.stream_copy_opts(ext=info['ext'])]
+
+        if sub_langs and sub_names:
+            # We have regular subtitles available to embed. Don't copy the
+            # existing subtitles, we may be running the postprocessor a second
+            # time.
+            opts.extend([
+                '-map', '-0:s',
+            ])
+
         for i, (lang, name) in enumerate(zip(sub_langs, sub_names)):
-            opts.extend(['-map', f'{i + 1}:0'])
             lang_code = ISO639Utils.short2long(lang) or lang
-            opts.extend([f'-metadata:s:s:{i}', f'language={lang_code}'])
+            opts.extend([
+                '-map', f'{i + 1}:0',
+                f'-metadata:s:s:{i}', f'language={lang_code}',
+            ])
+
             if name:
                 opts.extend([f'-metadata:s:s:{i}', f'handler_name={name}',
                              f'-metadata:s:s:{i}', f'title={name}'])
 
+        for json_lang, json_filename in json_subs.items():
+            escaped_json_filename = self._ffmpeg_filename_argument(json_filename)
+            json_basename = os.path.basename(json_filename)
+            opts.extend([
+                '-map', f'-0:m:filename:{json_lang}.json?',
+                '-attach', escaped_json_filename,
+                f'-metadata:s:m:filename:{json_basename}', 'mimetype=application/json',
+                f'-metadata:s:m:filename:{json_basename}', f'filename={json_lang}.json',
+            ])
+
         temp_filename = prepend_extension(filename, 'temp')
         self.to_screen(f'Embedding subtitles in "{filename}"')
         self.run_ffmpeg_multiple_files(input_files, temp_filename, opts)
         os.replace(temp_filename, filename)
 
-        files_to_delete = [] if self._already_have_subtitle else sub_filenames
+        files_to_delete = [] if self._already_have_subtitle else [*sub_filenames, *json_subs.values()]
         return files_to_delete, info
 
 
@@ -677,7 +710,7 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
     @staticmethod
     def _options(target_ext):
         audio_only = target_ext == 'm4a'
-        yield from FFmpegPostProcessor.stream_copy_opts(not audio_only)
+        yield from FFmpegPostProcessor.stream_copy_opts(not audio_only, ext=target_ext)
         if audio_only:
             yield from ('-vn', '-acodec', 'copy')
 
@@ -805,15 +838,20 @@ class FFmpegMetadataPP(FFmpegPostProcessor):
             write_json_file(self._downloader.sanitize_info(info, self.get_param('clean_infojson', True)), infofn)
             info['infojson_filename'] = infofn
 
-        old_stream, new_stream = self.get_stream_number(info['filepath'], ('tags', 'mimetype'), 'application/json')
-        if old_stream is not None:
-            yield ('-map', f'-0:{old_stream}')
-            new_stream -= 1
+        escaped_name = self._ffmpeg_filename_argument(infofn)
+        info_basename = os.path.basename(infofn)
 
         yield (
-            '-attach', self._ffmpeg_filename_argument(infofn),
-            f'-metadata:s:{new_stream}', 'mimetype=application/json',
-            f'-metadata:s:{new_stream}', 'filename=info.json',
+            # In order to override any old info.json reliably we need to
+            # instruct FFmpeg to consider valid tracks without a codec id, like
+            # JSON attachments.
+            '-copy_unknown',
+            # This map operation allows us to actually replace any previous
+            # info.json data.
+            '-map', '-0:m:filename:info.json?',
+            '-attach', escaped_name,
+            f'-metadata:s:m:filename:{info_basename}', 'mimetype=application/json',
+            f'-metadata:s:m:filename:{info_basename}', 'filename=info.json',
         )
 
 
@@ -872,7 +910,7 @@ class FFmpegFixupStretchedPP(FFmpegFixupPostProcessor):
         stretched_ratio = info.get('stretched_ratio')
         if stretched_ratio not in (None, 1):
             self._fixup('Fixing aspect ratio', info['filepath'], [
-                *self.stream_copy_opts(), '-aspect', f'{stretched_ratio:f}'])
+                *self.stream_copy_opts(ext=info['ext']), '-aspect', f'{stretched_ratio:f}'])
         return [], info
 
 
@@ -880,7 +918,7 @@ class FFmpegFixupM4aPP(FFmpegFixupPostProcessor):
     @PostProcessor._restrict_to(images=False, video=False)
     def run(self, info):
         if info.get('container') == 'm4a_dash':
-            self._fixup('Correcting container', info['filepath'], [*self.stream_copy_opts(), '-f', 'mp4'])
+            self._fixup('Correcting container', info['filepath'], [*self.stream_copy_opts(ext=info['ext']), '-f', 'mp4'])
         return [], info
 
 
@@ -903,7 +941,7 @@ class FFmpegFixupM3u8PP(FFmpegFixupPostProcessor):
             if self.get_audio_codec(info['filepath']) == 'aac':
                 args.extend(['-bsf:a', 'aac_adtstoasc'])
             self._fixup('Fixing MPEG-TS in MP4 container', info['filepath'], [
-                *self.stream_copy_opts(), *args])
+                *self.stream_copy_opts(ext=info['ext']), *args])
         return [], info
 
 
@@ -924,7 +962,7 @@ class FFmpegFixupTimestampPP(FFmpegFixupPostProcessor):
             opts = ['-vf', 'setpts=PTS-STARTPTS']
         else:
             opts = ['-c', 'copy', '-bsf', 'setts=ts=TS-STARTPTS']
-        self._fixup('Fixing frame timestamp', info['filepath'], [*opts, *self.stream_copy_opts(False), '-ss', self.trim])
+        self._fixup('Fixing frame timestamp', info['filepath'], [*opts, *self.stream_copy_opts(False, ext=info['ext']), '-ss', self.trim])
         return [], info
 
 
@@ -933,7 +971,7 @@ class FFmpegCopyStreamPP(FFmpegFixupPostProcessor):
 
     @PostProcessor._restrict_to(images=False)
     def run(self, info):
-        self._fixup(self.MESSAGE, info['filepath'], self.stream_copy_opts())
+        self._fixup(self.MESSAGE, info['filepath'], self.stream_copy_opts(ext=info['ext']))
         return [], info
 
 
@@ -1062,7 +1100,7 @@ class FFmpegSplitChaptersPP(FFmpegPostProcessor):
         self.to_screen(f'Splitting video by chapters; {len(chapters)} chapters found')
         for idx, chapter in enumerate(chapters):
             destination, opts = self._ffmpeg_args_for_chapter(idx + 1, chapter, info)
-            self.real_run_ffmpeg([(in_file, opts)], [(destination, self.stream_copy_opts())])
+            self.real_run_ffmpeg([(in_file, opts)], [(destination, self.stream_copy_opts(ext=info['ext']))])
         if in_file != info['filepath']:
             self._delete_downloaded_files(in_file, msg=None)
         return [], info