Merge eee4f7163d into eb15fd5a32

Closes #3937
Authored by: krichbanana, pzhlkj6612

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

yt_dlp/YoutubeDL.py

@@ -1470,7 +1470,8 @@ class YoutubeDL:
         return self.get_output_path(dir_type, filename)
 
     def _match_entry(self, info_dict, incomplete=False, silent=False):
-        """Returns None if the file should be downloaded"""
+        """Returns None if the file should be downloaded, False if the file is already present in
+        the download archive, or a string describing another reason to skip the file"""
         _type = 'video' if 'playlist-match-filter' in self.params['compat_opts'] else info_dict.get('_type', 'video')
         assert incomplete or _type == 'video', 'Only video result can be considered complete'
 
@@ -1545,6 +1546,7 @@ class YoutubeDL:
                 format_field(info_dict, 'id', f'{self._format_screen("%s", self.Styles.ID)}: '),
                 format_field(info_dict, 'title', f'{self._format_screen("%s", self.Styles.EMPHASIS)} '),
                 'has already been recorded in the archive'))
+            ret = False
             break_opt, break_err = 'break_on_existing', ExistingVideoReached
         else:
             try:
@@ -1553,12 +1555,13 @@ class YoutubeDL:
                 reason, break_opt, break_err = e.msg, 'match_filter', type(e)
             else:
                 break_opt, break_err = 'break_on_reject', RejectedVideoReached
+            ret = reason
         if reason is not None:
             if not silent:
                 self.to_screen('[download] ' + reason)
             if self.params.get(break_opt, False):
                 raise break_err()
-        return reason
+        return ret
 
     @staticmethod
     def add_extra_info(info_dict, extra_info):
@@ -1832,6 +1835,7 @@ class YoutubeDL:
                 self._raise_pending_errors(info_copy)
                 if self.params.get('force_write_download_archive', False):
                     self.record_download_archive(info_copy)
+                ie_result['__write_download_archive'] = self.params.get('force_write_download_archive', False)
                 return ie_result
 
         if result_type == 'video':
@@ -1971,7 +1975,9 @@ class YoutubeDL:
 
         common_info = self._playlist_infodict(ie_result, strict=True)
         title = common_info.get('playlist') or '<Untitled>'
-        if self._match_entry(common_info, incomplete=True) is not None:
+        skip_reason = self._match_entry(common_info, incomplete=True)
+        if skip_reason is not None:
+            ie_result['__write_download_archive'] = skip_reason is False
             return
         self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
 
@@ -2026,6 +2032,7 @@ class YoutubeDL:
             self.write_debug('The information of all playlist entries will be held in memory')
 
         failures = 0
+        all_write_download_archive = True
         max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
         for i, (playlist_index, entry) in enumerate(entries):
             if lazy:
@@ -2059,6 +2066,8 @@ class YoutubeDL:
             }, extra))
             if not entry_result:
                 failures += 1
+            elif not entry_result.get('__write_download_archive', False):
+                all_write_download_archive = False
             if failures >= max_failures:
                 self.report_error(
                     f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
@@ -2073,6 +2082,12 @@ class YoutubeDL:
             # Do not set for full playlist
             ie_result.pop('requested_entries')
 
+        if ie_result['_type'] == 'multi_video' and not failures:
+            if self.params.get('force_write_download_archive') or (
+                    all_write_download_archive and not self.params.get('simulate')
+                    and not self.params.get('skip_download')):
+                self.record_download_archive(ie_result)
+
         # Write the updated info to json
         if _infojson_written is True and self._write_info_json(
                 'updated playlist', ie_result,
@@ -2919,7 +2934,9 @@ class YoutubeDL:
 
         info_dict, _ = self.pre_process(info_dict)
 
-        if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
+        skip_reason = self._match_entry(info_dict, incomplete=self._format_fields)
+        if skip_reason is not None:
+            info_dict['__write_download_archive'] = skip_reason is False
             return info_dict
 
         self.post_extract(info_dict)
@@ -3022,6 +3039,7 @@ class YoutubeDL:
             assert write_archive.issubset({True, False, 'ignore'})
             if True in write_archive and False not in write_archive:
                 self.record_download_archive(info_dict)
+                info_dict['__write_download_archive'] = True
 
             info_dict['requested_downloads'] = downloaded_formats
             info_dict = self.run_all_pps('after_video', info_dict)

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)})))