Merge ed4d9a40c1 into e079ffbda6

Authored by: jiru

README.md

@@ -1869,6 +1869,9 @@ The following extractors use this feature:
 #### digitalconcerthall
 * `prefer_combined_hls`: Prefer extracting combined/pre-merged video and audio HLS formats. This will exclude 4K/HEVC video and lossless/FLAC audio formats, which are only available as split video/audio HLS formats
 
+#### sonylivseries
+* `sort_order`: Episode sort order for series extraction - one of `asc` (ascending, oldest first) or `desc` (descending, newest first). Default is `asc`
+
 **Note**: These options may be changed/removed in the future without concern for backward compatibility
 
 <!-- MANPAGE: MOVE "INSTALLATION" SECTION HERE -->

yt_dlp/extractor/ctvnews.py

@@ -1,11 +1,24 @@
+import json
 import re
+import urllib.parse
 
 from .common import InfoExtractor
-from ..utils import orderedSet
+from .ninecninemedia import NineCNineMediaIE
+from ..utils import extract_attributes, orderedSet
+from ..utils.traversal import find_element, traverse_obj
 
 
 class CTVNewsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:.+?\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)(?:$|[#?&])'
+    _BASE_REGEX = r'https?://(?:[^.]+\.)?ctvnews\.ca/'
+    _VIDEO_ID_RE = r'(?P<id>\d{5,})'
+    _PLAYLIST_ID_RE = r'(?P<id>\d\.\d{5,})'
+    _VALID_URL = [
+        rf'{_BASE_REGEX}video/c{_VIDEO_ID_RE}',
+        rf'{_BASE_REGEX}video(?:-gallery)?/?\?clipId={_VIDEO_ID_RE}',
+        rf'{_BASE_REGEX}video/?\?(?:playlist|bin)Id={_PLAYLIST_ID_RE}',
+        rf'{_BASE_REGEX}(?!video/)[^?#]*?{_PLAYLIST_ID_RE}/?(?:$|[?#])',
+        rf'{_BASE_REGEX}(?!video/)[^?#]+\?binId={_PLAYLIST_ID_RE}',
+    ]
     _TESTS = [{
         'url': 'http://www.ctvnews.ca/video?clipId=901995',
         'md5': 'b608f466c7fa24b9666c6439d766ab7e',
@@ -17,13 +30,32 @@ class CTVNewsIE(InfoExtractor):
             'timestamp': 1467286284,
             'upload_date': '20160630',
             'categories': [],
-            'tags': [],
-            'season_id': 57981,
-            'duration': 764.631,
-            'series': 'CTV News National story',
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'season': 'Season 0',
             'season_number': 0,
+            'season': 'Season 0',
+            'tags': [],
+            'series': 'CTV News National | Archive | Stories 2',
+            'season_id': '57981',
+            'thumbnail': r're:https?://.*\.jpg$',
+            'duration': 764.631,
+        },
+    }, {
+        'url': 'https://barrie.ctvnews.ca/video/c3030933-here_s-what_s-making-news-for-nov--15?binId=1272429',
+        'md5': '8b8c2b33c5c1803e3c26bc74ff8694d5',
+        'info_dict': {
+            'id': '3030933',
+            'ext': 'flv',
+            'title': 'Here’s what’s making news for Nov. 15',
+            'description': 'Here are the top stories we’re working on for CTV News at 11 for Nov. 15',
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2021_2_22_a602e68e-1514-410e-a67a-e1f7cccbacab_png_2000x1125.jpg',
+            'season_id': '58104',
+            'season_number': 0,
+            'tags': [],
+            'season': 'Season 0',
+            'categories': [],
+            'series': 'CTV News Barrie',
+            'upload_date': '20241116',
+            'duration': 42.943,
+            'timestamp': 1731722452,
         },
     }, {
         'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224',
@@ -46,6 +78,65 @@ class CTVNewsIE(InfoExtractor):
             'id': '1.5736957',
         },
         'playlist_mincount': 6,
+    }, {
+        'url': 'https://www.ctvnews.ca/business/respondents-to-bank-of-canada-questionnaire-largely-oppose-creating-a-digital-loonie-1.6665797',
+        'md5': '24bc4b88cdc17d8c3fc01dfc228ab72c',
+        'info_dict': {
+            'id': '2695026',
+            'ext': 'flv',
+            'season_id': '89852',
+            'series': 'From CTV News Channel',
+            'description': 'md5:796a985a23cacc7e1e2fafefd94afd0a',
+            'season': '2023',
+            'title': 'Bank of Canada asks public about digital currency',
+            'categories': [],
+            'tags': [],
+            'upload_date': '20230526',
+            'season_number': 2023,
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'timestamp': 1685105157,
+            'duration': 253.553,
+        },
+    }, {
+        'url': 'https://stox.ctvnews.ca/video-gallery?clipId=582589',
+        'md5': '135cc592df607d29dddc931f1b756ae2',
+        'info_dict': {
+            'id': '582589',
+            'ext': 'flv',
+            'categories': [],
+            'timestamp': 1427906183,
+            'season_number': 0,
+            'duration': 125.559,
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'series': 'CTV News Stox',
+            'description': 'CTV original footage of the rise and fall of the Berlin Wall.',
+            'title': 'Berlin Wall',
+            'season_id': '63817',
+            'season': 'Season 0',
+            'tags': [],
+            'upload_date': '20150401',
+        },
+    }, {
+        'url': 'https://ottawa.ctvnews.ca/features/regional-contact/regional-contact-archive?binId=1.1164587#3023759',
+        'md5': 'a14c0603557decc6531260791c23cc5e',
+        'info_dict': {
+            'id': '3023759',
+            'ext': 'flv',
+            'season_number': 2024,
+            'timestamp': 1731798000,
+            'season': '2024',
+            'episode': 'Episode 125',
+            'description': 'CTV News Ottawa at Six',
+            'duration': 2712.076,
+            'episode_number': 125,
+            'upload_date': '20241116',
+            'title': 'CTV News Ottawa at Six for Saturday, November 16, 2024',
+            'thumbnail': 'http://images2.9c9media.com/image_asset/2019_3_28_35f5afc3-10f6-4d92-b194-8b9a86f55c6a_png_1920x1080.jpg',
+            'categories': [],
+            'tags': [],
+            'series': 'CTV News Ottawa at Six',
+            'season_id': '92667',
+        },
     }, {
         'url': 'http://www.ctvnews.ca/1.810401',
         'only_matching': True,
@@ -57,29 +148,35 @@ class CTVNewsIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    def _ninecninemedia_url_result(self, clip_id):
+        return self.url_result(f'9c9media:ctvnews_web:{clip_id}', NineCNineMediaIE, clip_id)
+
     def _real_extract(self, url):
         page_id = self._match_id(url)
 
-        def ninecninemedia_url_result(clip_id):
+        if mobj := re.fullmatch(self._VIDEO_ID_RE, urllib.parse.urlparse(url).fragment):
-            return {
+            page_id = mobj.group('id')
-                '_type': 'url_transparent',
-                'id': clip_id,
-                'url': f'9c9media:ctvnews_web:{clip_id}',
-                'ie_key': 'NineCNineMedia',
-            }
 
-        if page_id.isdigit():
+        if re.fullmatch(self._VIDEO_ID_RE, page_id):
-            return ninecninemedia_url_result(page_id)
+            return self._ninecninemedia_url_result(page_id)
-        else:
+
-            webpage = self._download_webpage(f'http://www.ctvnews.ca/{page_id}', page_id, query={
+        webpage = self._download_webpage(f'https://www.ctvnews.ca/{page_id}', page_id, query={
-                'ot': 'example.AjaxPageLayout.ot',
+            'ot': 'example.AjaxPageLayout.ot',
-                'maxItemsPerPage': 1000000,
+            'maxItemsPerPage': 1000000,
-            })
+        })
-            entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet(
+        entries = [self._ninecninemedia_url_result(clip_id)
-                re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
+                   for clip_id in orderedSet(re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
-            if not entries:
+        if not entries:
-                webpage = self._download_webpage(url, page_id)
+            webpage = self._download_webpage(url, page_id)
-                if 'getAuthStates("' in webpage:
+            if 'getAuthStates("' in webpage:
-                    entries = [ninecninemedia_url_result(clip_id) for clip_id in
+                entries = [self._ninecninemedia_url_result(clip_id) for clip_id in
-                               self._search_regex(r'getAuthStates\("([\d+,]+)"', webpage, 'clip ids').split(',')]
+                           self._search_regex(r'getAuthStates\("([\d+,]+)"', webpage, 'clip ids').split(',')]
-            return self.playlist_result(entries, page_id)
+            else:
+                entries = [
+                    self._ninecninemedia_url_result(clip_id) for clip_id in
+                    traverse_obj(webpage, (
+                        {find_element(tag='jasper-player-container', html=True)},
+                        {extract_attributes}, 'axis-ids', {json.loads}, ..., 'axisId'))
+                ]
+
+        return self.playlist_result(entries, page_id)

yt_dlp/extractor/litv.py

@@ -1,30 +1,32 @@
 import json
+import uuid
 
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
+    join_nonempty,
     smuggle_url,
     traverse_obj,
     try_call,
     unsmuggle_url,
+    urljoin,
 )
 
 
 class LiTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:vod|promo)/[^/]+/(?:content\.do)?\?.*?\b(?:content_)?id=(?P<id>[^&]+)'
+    _VALID_URL = r'https?://(?:www\.)?litv\.tv/(?:[^/?#]+/watch/|vod/[^/?#]+/content\.do\?content_id=)(?P<id>[\w-]+)'
-
+    _URL_TEMPLATE = 'https://www.litv.tv/%s/watch/%s'
-    _URL_TEMPLATE = 'https://www.litv.tv/vod/%s/content.do?content_id=%s'
+    _GEO_COUNTRIES = ['TW']
-
     _TESTS = [{
-        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'url': 'https://www.litv.tv/drama/watch/VOD00041610',
         'info_dict': {
             'id': 'VOD00041606',
             'title': '花千骨',
         },
         'playlist_count': 51,  # 50 episodes + 1 trailer
     }, {
-        'url': 'https://www.litv.tv/vod/drama/content.do?brc_id=root&id=VOD00041610&isUHEnabled=true&autoPlay=1',
+        'url': 'https://www.litv.tv/drama/watch/VOD00041610',
         'md5': 'b90ff1e9f1d8f5cfcd0a44c3e2b34c7a',
         'info_dict': {
             'id': 'VOD00041610',
@@ -32,16 +34,15 @@ class LiTVIE(InfoExtractor):
             'title': '花千骨第1集',
             'thumbnail': r're:https?://.*\.jpg$',
             'description': '《花千骨》陸劇線上看。十六年前，平靜的村莊內，一名女嬰隨異相出生，途徑此地的蜀山掌門清虛道長算出此女命運非同一般，她體內散發的異香易招惹妖魔。一念慈悲下，他在村莊周邊設下結界阻擋妖魔入侵，讓其年滿十六後去蜀山，並賜名花千骨。',
-            'categories': ['奇幻', '愛情', '中國', '仙俠'],
+            'categories': ['奇幻', '愛情', '仙俠', '古裝'],
             'episode': 'Episode 1',
             'episode_number': 1,
         },
         'params': {
             'noplaylist': True,
         },
-        'skip': 'Georestricted to Taiwan',
     }, {
-        'url': 'https://www.litv.tv/promo/miyuezhuan/?content_id=VOD00044841&',
+        'url': 'https://www.litv.tv/drama/watch/VOD00044841',
         'md5': '88322ea132f848d6e3e18b32a832b918',
         'info_dict': {
             'id': 'VOD00044841',
@@ -55,94 +56,62 @@ class LiTVIE(InfoExtractor):
     def _extract_playlist(self, playlist_data, content_type):
         all_episodes = [
             self.url_result(smuggle_url(
-                self._URL_TEMPLATE % (content_type, episode['contentId']),
+                self._URL_TEMPLATE % (content_type, episode['content_id']),
                 {'force_noplaylist': True}))  # To prevent infinite recursion
-            for episode in traverse_obj(playlist_data, ('seasons', ..., 'episode', lambda _, v: v['contentId']))]
+            for episode in traverse_obj(playlist_data, ('seasons', ..., 'episodes', lambda _, v: v['content_id']))]
 
-        return self.playlist_result(all_episodes, playlist_data['contentId'], playlist_data.get('title'))
+        return self.playlist_result(all_episodes, playlist_data['content_id'], playlist_data.get('title'))
 
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
-
         video_id = self._match_id(url)
-
         webpage = self._download_webpage(url, video_id)
+        vod_data = self._search_nextjs_data(webpage, video_id)['props']['pageProps']
 
-        if self._search_regex(
+        program_info = traverse_obj(vod_data, ('programInformation', {dict})) or {}
-                r'(?i)<meta\s[^>]*http-equiv="refresh"\s[^>]*content="[0-9]+;\s*url=https://www\.litv\.tv/"',
+        playlist_data = traverse_obj(vod_data, ('seriesTree'))
-                webpage, 'meta refresh redirect', default=False, group=0):
+        if playlist_data and self._yes_playlist(program_info.get('series_id'), video_id, smuggled_data):
-            raise ExtractorError('No such content found', expected=True)
+            return self._extract_playlist(playlist_data, program_info.get('content_type'))
 
-        program_info = self._parse_json(self._search_regex(
+        asset_id = traverse_obj(program_info, ('assets', 0, 'asset_id', {str}))
-            r'var\s+programInfo\s*=\s*([^;]+)', webpage, 'VOD data', default='{}'),
+        if asset_id:  # This is a VOD
-            video_id)
+            media_type = 'vod'
+        else:  # This is a live stream
+            asset_id = program_info['content_id']
+            media_type = program_info['content_type']
+        puid = try_call(lambda: self._get_cookies('https://www.litv.tv/')['PUID'].value)
+        if puid:
+            endpoint = 'get-urls'
+        else:
+            puid = str(uuid.uuid4())
+            endpoint = 'get-urls-no-auth'
+        video_data = self._download_json(
+            f'https://www.litv.tv/api/{endpoint}', video_id,
+            data=json.dumps({'AssetId': asset_id, 'MediaType': media_type, 'puid': puid}).encode(),
+            headers={'Content-Type': 'application/json'})
 
-        # In browsers `getProgramInfo` request is always issued. Usually this
+        if error := traverse_obj(video_data, ('error', {dict})):
-        # endpoint gives the same result as the data embedded in the webpage.
+            error_msg = traverse_obj(error, ('message', {str}))
-        # If, for some reason, there are no embedded data, we do an extra request.
+            if error_msg and 'OutsideRegionError' in error_msg:
-        if 'assetId' not in program_info:
-            program_info = self._download_json(
-                'https://www.litv.tv/vod/ajax/getProgramInfo', video_id,
-                query={'contentId': video_id},
-                headers={'Accept': 'application/json'})
-
-        series_id = program_info['seriesId']
-        if self._yes_playlist(series_id, video_id, smuggled_data):
-            playlist_data = self._download_json(
-                'https://www.litv.tv/vod/ajax/getSeriesTree', video_id,
-                query={'seriesId': series_id}, headers={'Accept': 'application/json'})
-            return self._extract_playlist(playlist_data, program_info['contentType'])
-
-        video_data = self._parse_json(self._search_regex(
-            r'uiHlsUrl\s*=\s*testBackendData\(([^;]+)\);',
-            webpage, 'video data', default='{}'), video_id)
-        if not video_data:
-            payload = {'assetId': program_info['assetId']}
-            puid = try_call(lambda: self._get_cookies('https://www.litv.tv/')['PUID'].value)
-            if puid:
-                payload.update({
-                    'type': 'auth',
-                    'puid': puid,
-                })
-                endpoint = 'getUrl'
-            else:
-                payload.update({
-                    'watchDevices': program_info['watchDevices'],
-                    'contentType': program_info['contentType'],
-                })
-                endpoint = 'getMainUrlNoAuth'
-            video_data = self._download_json(
-                f'https://www.litv.tv/vod/ajax/{endpoint}', video_id,
-                data=json.dumps(payload).encode(),
-                headers={'Content-Type': 'application/json'})
-
-        if not video_data.get('fullpath'):
-            error_msg = video_data.get('errorMessage')
-            if error_msg == 'vod.error.outsideregionerror':
                 self.raise_geo_restricted('This video is available in Taiwan only')
-            if error_msg:
+            elif error_msg:
                 raise ExtractorError(f'{self.IE_NAME} said: {error_msg}', expected=True)
-            raise ExtractorError(f'Unexpected result from {self.IE_NAME}')
+            raise ExtractorError(f'Unexpected error from {self.IE_NAME}')
 
         formats = self._extract_m3u8_formats(
-            video_data['fullpath'], video_id, ext='mp4',
+            video_data['result']['AssetURLs'][0], video_id, ext='mp4', m3u8_id='hls')
-            entry_protocol='m3u8_native', m3u8_id='hls')
         for a_format in formats:
             # LiTV HLS segments doesn't like compressions
             a_format.setdefault('http_headers', {})['Accept-Encoding'] = 'identity'
 
-        title = program_info['title'] + program_info.get('secondaryMark', '')
-        description = program_info.get('description')
-        thumbnail = program_info.get('imageFile')
-        categories = [item['name'] for item in program_info.get('category', [])]
-        episode = int_or_none(program_info.get('episode'))
-
         return {
             'id': video_id,
             'formats': formats,
-            'title': title,
+            'title': join_nonempty('title', 'secondary_mark', delim='', from_dict=program_info),
-            'description': description,
+            **traverse_obj(program_info, {
-            'thumbnail': thumbnail,
+                'description': ('description', {str}),
-            'categories': categories,
+                'thumbnail': ('picture', {urljoin('https://p-cdnstatic.svc.litv.tv/')}),
-            'episode_number': episode,
+                'categories': ('genres', ..., 'name', {str}),
+                'episode_number': ('episode', {int_or_none}),
+            }),
         }

yt_dlp/extractor/mediasite.py

@@ -5,6 +5,7 @@ import urllib.parse
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    determine_ext,
     float_or_none,
     mimetype2ext,
     smuggle_url,
@@ -268,7 +269,29 @@ class MediasiteIE(InfoExtractor):
             formats.extend(stream_formats)
 
         # XXX: Presentation['Presenters']
-        # XXX: Presentation['Transcript']
+        transcripts = presentation.get('Transcripts', {})
+        captions, subtitles = {}, {}
+        for transcript in transcripts:
+            lang_code = traverse_obj(
+                transcript, (('DetailedLanguageCode', 'LanguageCode'), {str}), get_all=False)
+            lang_name = transcript.get('Language')
+            t = {
+                'url': transcript.get('CaptionsUrl'),
+                'name': lang_name,
+            }
+            if 'Auto-Generated' in lang_name:
+                captions.setdefault(lang_code, []).append(t)
+            else:
+                subtitles.setdefault(lang_code, []).append(t)
+        if transcript_url := presentation.get('TranscriptUrl'):
+            if determine_ext(transcript_url) != 'txt':
+                if len(transcripts) == 1 and captions:
+                    captions.setdefault(lang_code, []).append({
+                        'url': transcript_url,
+                        'name': lang_name,
+                    })
+                else:
+                    subtitles.setdefault('und', []).append({'url': transcript_url})
 
         return {
             'id': resource_id,
@@ -277,6 +300,8 @@ class MediasiteIE(InfoExtractor):
             'duration': float_or_none(presentation.get('Duration'), 1000),
             'timestamp': float_or_none(presentation.get('UnixTime'), 1000),
             'formats': formats,
+            'automatic_captions': captions,
+            'subtitles': subtitles,
             'thumbnails': thumbnails,
         }
 

yt_dlp/extractor/sonyliv.py

@@ -199,8 +199,9 @@ class SonyLIVSeriesIE(InfoExtractor):
         },
     }]
     _API_BASE = 'https://apiv2.sonyliv.com/AGL'
+    _SORT_ORDERS = ('asc', 'desc')
 
-    def _entries(self, show_id):
+    def _entries(self, show_id, sort_order):
         headers = {
             'Accept': 'application/json, text/plain, */*',
             'Referer': 'https://www.sonyliv.com',
@@ -215,6 +216,9 @@ class SonyLIVSeriesIE(InfoExtractor):
                 'from': '0',
                 'to': '49',
             }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
+
+        if sort_order == 'desc':
+            seasons = reversed(seasons)
         for season in seasons:
             season_id = str(season['id'])
             note = traverse_obj(season, ('metadata', 'title', {str})) or 'season'
@@ -226,7 +230,7 @@ class SonyLIVSeriesIE(InfoExtractor):
                         'from': str(cursor),
                         'to': str(cursor + 99),
                         'orderBy': 'episodeNumber',
-                        'sortOrder': 'asc',
+                        'sortOrder': sort_order,
                     }), ('resultObj', 'containers', 0, 'containers', lambda _, v: int_or_none(v['id'])))
                 if not episodes:
                     break
@@ -237,4 +241,10 @@ class SonyLIVSeriesIE(InfoExtractor):
 
     def _real_extract(self, url):
         show_id = self._match_id(url)
-        return self.playlist_result(self._entries(show_id), playlist_id=show_id)
+
+        sort_order = self._configuration_arg('sort_order', [self._SORT_ORDERS[0]])[0]
+        if sort_order not in self._SORT_ORDERS:
+            raise ValueError(
+                f'Invalid sort order "{sort_order}". Allowed values are: {", ".join(self._SORT_ORDERS)}')
+
+        return self.playlist_result(self._entries(show_id, sort_order), playlist_id=show_id)