Merge 04417e89de into eb15fd5a32

Closes #3937
Authored by: krichbanana, pzhlkj6612

Co-authored-by: Mozi <29089388+pzhlkj6612@users.noreply.github.com>

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,
@@ -1724,6 +1728,7 @@ from .rinsefm import (
     RinseFMIE,
 )
 from .rmcdecouverte import RMCDecouverteIE
+from .roblox import RobloxIE
 from .rockstargames import RockstarGamesIE
 from .rokfin import (
     RokfinChannelIE,

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/roblox.py

@@ -0,0 +1,129 @@
+from .common import InfoExtractor
+from ..postprocessor.ffmpeg import FFmpegPostProcessor
+from ..utils import (
+    ExtractorError,
+    extract_attributes,
+    float_or_none,
+    get_element_by_id,
+    get_element_by_class,
+    get_element_html_by_class,
+    get_element_html_by_id,
+    int_or_none,
+    PostProcessingError,
+    strip_or_none,
+    str_or_none,
+    str_to_int,
+    traverse_obj,
+    try_call,
+    unified_timestamp,
+)
+
+
+class RobloxIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?roblox\.com/library/(?P<id>\d+)'
+    _TESTS = [{
+        # UGC Audio
+        'url': 'https://www.roblox.com/library/7910582982/Backrooms-Ambiance-High-Quality',
+        'md5': '',
+        'info_dict': {
+            'id': '7910582982',
+            'ext': 'ogg',
+            'title': 'Backrooms Ambiance (High Quality)',
+            'description': 'Found an actual higher quality of the sound.',
+            'uploader': 'ChaseDJ549',
+            'uploader_id': '412014916',
+            'categories': ['Horror'],
+            'like_count': int,
+            'timestamp': 1636142127,
+            'modified_timestamp': 1656694893
+        },
+    }]
+
+    def _real_extract(self, url):
+        asset_id = self._match_id(url)
+        webpage = self._download_webpage(url, asset_id)
+
+        item_container_div = get_element_html_by_id('item-container', webpage)
+        item_container_attrs = extract_attributes(item_container_div[:item_container_div.find('>')+1])
+        asset_type = item_container_attrs.get('data-asset-type')
+        if asset_type and (asset_type not in ('Audio', 'Video')):
+            raise ExtractorError('This asset is not an audio/video', expected=True)
+        asset_uploader_id, asset_uploader_name = self._search_regex(
+            r'>By <a.+href=["\']https?://(?:www\.)?roblox\.com/users/(?P<id>\d+)[^"\']*["\'][^>]*>@?(?P<name>\w+)</a',
+            webpage, 'asset creator', fatal=False, group=('id', 'name'))
+
+        is_logged_out = not self._get_cookies('https://roblox.com').get('.ROBLOSECURITY')
+        toolbox_result = traverse_obj(
+            self._download_json(f'https://apis.roblox.com/toolbox-service/v1/items/details', asset_id, query={'assetIds': asset_id},
+                                note='Downloading extra metadata JSON', errnote=False if is_logged_out else 'Unable to download extra metadata JSON',
+                                fatal=False),
+            ('data', ...), default={}, expected_type=dict, get_all=False)
+        toolbox_asset_data = toolbox_result.get('asset') or {}
+        toolbox_creator_data = toolbox_result.get('creator') or {}
+        toolbox_audio_data = toolbox_asset_data.get('audioDetails') or {}
+
+        info_dict = {
+            'id': asset_id,
+            'title': toolbox_asset_data.get('name') or item_container_attrs.get('data-item-name'),
+            'uploader': toolbox_creator_data.get('name') or asset_uploader_name,
+            'uploader_id': str_or_none(toolbox_creator_data.get('id')) or asset_uploader_id,
+            # TODO: Sound effects have separate kinds of categories 
+            'categories': toolbox_asset_data.get('assetGenres') or [strip_or_none(get_element_by_class('item-genre', webpage))],
+            'like_count': str_to_int(extract_attributes(get_element_html_by_class('favoriteCount', webpage)).get('title')),
+            'timestamp': unified_timestamp(toolbox_asset_data.get('createdUtc')),
+            'modified_timestamp': unified_timestamp(toolbox_asset_data.get('updatedUtc')),  # TODO: Extract from webpage
+            'track': toolbox_audio_data.get('title'),
+            'artist': toolbox_audio_data.get('artist'),
+            'genre': try_call(lambda: toolbox_audio_data['musicGenre'].capitalize())
+        }
+
+        cdn_result = self._download_json(
+            f'https://assetdelivery.roblox.com/v1/assetId/{asset_id}',
+            asset_id, note='Downloading file data JSON', headers={
+                'Accept': 'application/json',
+                'roblox-browser-asset-request': 'true'
+            })
+        asset_file_url = cdn_result.get('location')
+        if not asset_file_url:
+            if asset_type == 'Audio':
+                media_play_icon_div = get_element_html_by_class('MediaPlayerIcon')
+                if not media_play_icon_div:
+                    self.raise_no_formats('This audio is unavailable', expected=True, video_id=asset_id)
+                asset_file_url = self._search_regex(r'data-mediathumb-url=["\']https?://[^"\']+["\']', media_play_icon_div, 'audio preview URL')
+            elif is_logged_out:  # assetdelivery API randomly requires auth cookies
+                self.raise_login_required(metadata_available=True)
+            else:
+                self.raise_no_formats(
+                    traverse_obj(cdn_result, ('errors', ..., 'message'), default='Unable to fetch asset', expected_type=str, get_all=False),
+                    video_id=asset_id)
+
+        if asset_file_url:
+            # Assets have no file extension and use binary/octet-stream as Content-Type
+            pp = FFmpegPostProcessor(self._downloader)
+            self.to_screen(f'{asset_id}: Checking file format with ffprobe')
+            try:
+                metadata = pp.get_metadata_object(asset_file_url)
+            except PostProcessingError as err:
+                raise ExtractorError(err.msg, expected=True)
+
+            v_stream = a_stream = {}
+            for stream in metadata['streams']:
+                if stream['codec_type'] == 'video':
+                    v_stream = stream
+                elif stream['codec_type'] == 'audio':
+                    a_stream = stream
+
+            info_dict['formats'] = [{
+                'url': asset_file_url,
+                'ext': 'mp4' if 'mp4' in metadata['format']['format_name'] else metadata['format']['format_name'].split(',')[-1],
+                'vcodec': v_stream.get('codec_name'),
+                'acodec': a_stream.get('codec_name'),
+                'tbr': int_or_none(metadata['format'].get('bit_rate'), scale=1000),
+                'vbr': int_or_none(v_stream.get('bit_rate'), scale=1000),
+                'abr': int_or_none(a_stream.get('bit_rate'), scale=1000),
+                'height': int_or_none(v_stream.get('height')),
+                'width': int_or_none(v_stream.get('width')),
+                'filesize': float_or_none(metadata['format'].get('size'))
+            }]
+
+        return info_dict