Merge dfc4769853 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,

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/viu.py

@@ -5,9 +5,11 @@ import urllib.parse
 import uuid
 
 from .common import InfoExtractor
+from ..networking.exceptions import HTTPError
 from ..utils import (
     ExtractorError,
     int_or_none,
+    merge_dicts,
     remove_end,
     smuggle_url,
     strip_or_none,
@@ -150,7 +152,7 @@ class ViuPlaylistIE(ViuBaseIE):
 class ViuOTTIE(InfoExtractor):
     IE_NAME = 'viu:ott'
     _NETRC_MACHINE = 'viu'
-    _VALID_URL = r'https?://(?:www\.)?viu\.com/ott/(?P<country_code>[a-z]{2})/(?P<lang_code>[a-z]{2}-[a-z]{2})/vod/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?viu\.com/ott/(?P<country_code>[a-z]{2})/(?P<lang_code>[a-z]{2}(?:-[a-z]{2})?)/vod/(?P<id>\d+)'
     _TESTS = [{
         'url': 'http://www.viu.com/ott/sg/en-us/vod/3421/The%20Prime%20Minister%20and%20I',
         'info_dict': {
@@ -195,6 +197,19 @@ class ViuOTTIE(InfoExtractor):
             'noplaylist': False,
         },
         'skip': 'Geo-restricted to Hong Kong',
+    }, {
+        'url': 'https://www.viu.com/ott/id/id/vod/2221644/Detective-Conan',
+        'info_dict': {
+            'id': '2221644',
+            'ext': 'mp4',
+            'description': 'md5:b199bcdb07b1e01a03529f155349ddd5',
+            'duration': 1425,
+            'series': 'Detective Conan',
+            'title': 'Detective Conan - Episode 1150',
+            'episode': 'Detective Conan - Episode 1150',
+            'episode_number': 1150,
+            'thumbnail': r're:https?://prod-images\.viu\.com/clip_asset_v6/\d+/\d+/[a-f0-9]+',
+        },
     }]
 
     _AREA_ID = {
@@ -270,27 +285,43 @@ class ViuOTTIE(InfoExtractor):
         url, idata = unsmuggle_url(url, {})
         country_code, lang_code, video_id = self._match_valid_url(url).groups()
 
+        webpage = self._download_webpage(url, video_id, fatal=False)
+        json_ld = self._search_json_ld(webpage, video_id, fatal=False)
+        next_js_data = (self._search_nextjs_data(webpage, video_id, fatal=False) or {}).get('props')
+        runtime_info = traverse_obj(next_js_data, ('initialState', 'app', 'runtimeInfo'))
+
         query = {
             'r': 'vod/ajax-detail',
             'platform_flag_label': 'web',
             'product_id': video_id,
         }
 
-        area_id = self._AREA_ID.get(country_code.upper())
+        area_id = self._AREA_ID.get(country_code.upper()) or runtime_info.get('areaId')
         if area_id:
             query['area_id'] = area_id
 
-        product_data = self._download_json(
+        try:
-            f'http://www.viu.com/ott/{country_code}/index.php', video_id,
+            product_data = self._download_json(
-            'Downloading video info', query=query)['data']
+                f'http://www.viu.com/ott/{country_code}/index.php', video_id,
+                'Downloading video info', query=query)['data']
+        # The `fatal` in `_download_json` didn't prevent json error
+        # FIXME: probably the error still too broad
+        except ExtractorError as e:
+            if not isinstance(e.cause, (json.JSONDecodeError, HTTPError)):
+                raise
+            # NOTE: some geo-blocked like https://www.viu.com/ott/sg/en/vod/108599/The-Beauty-Inside actually can bypassed
+            # on other region (like in ID)
+            product_data = traverse_obj(
+                next_js_data, ('pageProps', 'fallback', lambda k, v: v if re.match(r'@"PRODUCT_DETAIL"[^:]+', k) else None),
+                get_all=False)['data']
 
         video_data = product_data.get('current_product')
         if not video_data:
             self.raise_geo_restricted()
 
-        series_id = video_data.get('series_id')
+        series_id = video_data.get('series_id') or traverse_obj(product_data, ('series', 'series_id'))
         if self._yes_playlist(series_id, video_id, idata):
-            series = product_data.get('series') or {}
+            series = product_data.get('series') or traverse_obj(product_data, ('series', 'name')) or {}
             product = series.get('product')
             if product:
                 entries = []
@@ -308,7 +339,9 @@ class ViuOTTIE(InfoExtractor):
         duration_limit = False
         query = {
             'ccs_product_id': video_data['ccs_product_id'],
-            'language_flag_id': self._LANGUAGE_FLAG.get(lang_code.lower()) or '3',
+            'language_flag_id': self._LANGUAGE_FLAG.get(lang_code.lower()) or runtime_info.get('languageFlagId') or '3',
+            'platform_flag_label': 'web',
+            'countryCode': country_code.upper(),
         }
 
         def download_playback():
@@ -384,7 +417,7 @@ class ViuOTTIE(InfoExtractor):
                 })
 
         title = strip_or_none(video_data.get('synopsis'))
-        return {
+        return merge_dicts({
             'id': video_id,
             'title': title,
             'description': video_data.get('description'),
@@ -395,7 +428,12 @@ class ViuOTTIE(InfoExtractor):
             'thumbnail': url_or_none(video_data.get('cover_image_url')),
             'formats': formats,
             'subtitles': subtitles,
-        }
+        }, traverse_obj(json_ld, {
+            'thumbnails': 'thumbnails',
+            'title': 'title',
+            'episode': 'episode',
+            'episode_number': 'episode_number',
+        }))
 
 
 class ViuOTTIndonesiaBaseIE(InfoExtractor):