[ie/Digiteka] Rework extractor (#14903)

Closes #12454
Authored by: beliote

yt_dlp/extractor/_extractors.py

@@ -1284,6 +1284,10 @@ from .nest import (
     NestClipIE,
     NestIE,
 )
+from .netapp import (
+    NetAppCollectionIE,
+    NetAppVideoIE,
+)
 from .neteasemusic import (
     NetEaseMusicAlbumIE,
     NetEaseMusicDjRadioIE,

yt_dlp/extractor/digiteka.py

@@ -1,5 +1,6 @@
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import int_or_none, url_or_none
+from ..utils.traversal import traverse_obj
 
 
 class DigitekaIE(InfoExtractor):
@@ -25,74 +26,56 @@ class DigitekaIE(InfoExtractor):
         )/(?P<id>[\d+a-z]+)'''
     _EMBED_REGEX = [r'<(?:iframe|script)[^>]+src=["\'](?P<url>(?:https?:)?//(?:www\.)?ultimedia\.com/deliver/(?:generic|musique)(?:/[^/]+)*/(?:src|article)/[\d+a-z]+)']
     _TESTS = [{
-        # news
-        'url': 'https://www.ultimedia.com/default/index/videogeneric/id/s8uk0r',
-        'md5': '276a0e49de58c7e85d32b057837952a2',
+        'url': 'https://www.ultimedia.com/default/index/videogeneric/id/3x5x55k',
         'info_dict': {
-            'id': 's8uk0r',
+            'id': '3x5x55k',
             'ext': 'mp4',
-            'title': 'Loi sur la fin de vie: le texte prévoit un renforcement des directives anticipées',
+            'title': 'Il est passionné de DS',
             'thumbnail': r're:^https?://.*\.jpg',
-            'duration': 74,
-            'upload_date': '20150317',
-            'timestamp': 1426604939,
-            'uploader_id': '3fszv',
+            'duration': 89,
+            'upload_date': '20251012',
+            'timestamp': 1760285363,
+            'uploader_id': '3pz33',
         },
-    }, {
-        # music
-        'url': 'https://www.ultimedia.com/default/index/videomusic/id/xvpfp8',
-        'md5': '2ea3513813cf230605c7e2ffe7eca61c',
-        'info_dict': {
-            'id': 'xvpfp8',
-            'ext': 'mp4',
-            'title': 'Two - C\'est La Vie (clip)',
-            'thumbnail': r're:^https?://.*\.jpg',
-            'duration': 233,
-            'upload_date': '20150224',
-            'timestamp': 1424760500,
-            'uploader_id': '3rfzk',
-        },
-    }, {
-        'url': 'https://www.digiteka.net/deliver/generic/iframe/mdtk/01637594/src/lqm3kl/zone/1/showtitle/1/autoplay/yes',
-        'only_matching': True,
+        'params': {'skip_download': True},
     }]
+    _IFRAME_MD_ID = '01836272'   # One static ID working for Ultimedia iframes
 
     def _real_extract(self, url):
-        mobj = self._match_valid_url(url)
-        video_id = mobj.group('id')
-        video_type = mobj.group('embed_type') or mobj.group('site_type')
-        if video_type == 'music':
-            video_type = 'musique'
+        video_id = self._match_id(url)
 
-        deliver_info = self._download_json(
-            f'http://www.ultimedia.com/deliver/video?video={video_id}&topic={video_type}',
-            video_id)
-
-        yt_id = deliver_info.get('yt_id')
-        if yt_id:
-            return self.url_result(yt_id, 'Youtube')
-
-        jwconf = deliver_info['jwconf']
+        video_info = self._download_json(
+            f'https://www.ultimedia.com/player/getConf/{self._IFRAME_MD_ID}/1/{video_id}', video_id,
+            note='Downloading player configuration')['video']
 
         formats = []
-        for source in jwconf['playlist'][0]['sources']:
-            formats.append({
-                'url': source['file'],
-                'format_id': source.get('label'),
-            })
+        subtitles = {}
 
-        title = deliver_info['title']
-        thumbnail = jwconf.get('image')
-        duration = int_or_none(deliver_info.get('duration'))
-        timestamp = int_or_none(deliver_info.get('release_time'))
-        uploader_id = deliver_info.get('owner_id')
+        if hls_url := traverse_obj(video_info, ('media_sources', 'hls', 'hls_auto', {url_or_none})):
+            fmts, subs = self._extract_m3u8_formats_and_subtitles(
+                hls_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
+            formats.extend(fmts)
+            self._merge_subtitles(subs, target=subtitles)
+
+        for format_id, mp4_url in traverse_obj(video_info, ('media_sources', 'mp4', {dict.items}, ...)):
+            if not mp4_url:
+                continue
+            formats.append({
+                'url': mp4_url,
+                'format_id': format_id,
+                'height': int_or_none(format_id.partition('_')[2]),
+                'ext': 'mp4',
+            })
 
         return {
             'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'timestamp': timestamp,
-            'uploader_id': uploader_id,
             'formats': formats,
+            'subtitles': subtitles,
+            **traverse_obj(video_info, {
+                'title': ('title', {str}),
+                'thumbnail': ('image', {url_or_none}),
+                'duration': ('duration', {int_or_none}),
+                'timestamp': ('creationDate', {int_or_none}),
+                'uploader_id': ('ownerId', {str}),
+            }),
         }

yt_dlp/extractor/netapp.py

@@ -0,0 +1,79 @@
+from .brightcove import BrightcoveNewIE
+from .common import InfoExtractor
+from ..utils import parse_iso8601
+from ..utils.traversal import require, traverse_obj
+
+
+class NetAppBaseIE(InfoExtractor):
+    _BC_URL = 'https://players.brightcove.net/6255154784001/default_default/index.html?videoId={}'
+
+    @staticmethod
+    def _parse_metadata(item):
+        return traverse_obj(item, {
+            'title': ('name', {str}),
+            'description': ('description', {str}),
+            'timestamp': ('createdAt', {parse_iso8601}),
+        })
+
+
+class NetAppVideoIE(NetAppBaseIE):
+    _VALID_URL = r'https?://media\.netapp\.com/video-detail/(?P<id>[0-9a-f-]+)'
+
+    _TESTS = [{
+        'url': 'https://media.netapp.com/video-detail/da25fc01-82ad-5284-95bc-26920200a222/seamless-storage-for-modern-kubernetes-deployments',
+        'info_dict': {
+            'id': '1843620950167202073',
+            'ext': 'mp4',
+            'title': 'Seamless storage for modern Kubernetes deployments',
+            'description': 'md5:1ee39e315243fe71fb90af2796037248',
+            'uploader_id': '6255154784001',
+            'duration': 2159.41,
+            'thumbnail': r're:https://house-fastly-signed-us-east-1-prod\.brightcovecdn\.com/image/.*\.jpg',
+            'tags': 'count:15',
+            'timestamp': 1758213949,
+            'upload_date': '20250918',
+        },
+    }, {
+        'url': 'https://media.netapp.com/video-detail/45593e5d-cf1c-5996-978c-c9081906e69f/unleash-ai-innovation-with-your-data-with-the-netapp-platform',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_uuid = self._match_id(url)
+        metadata = self._download_json(
+            f'https://api.media.netapp.com/client/detail/{video_uuid}', video_uuid)
+
+        brightcove_video_id = traverse_obj(metadata, (
+            'sections', lambda _, v: v['type'] == 'Player', 'video', {str}, any, {require('brightcove video id')}))
+
+        video_item = traverse_obj(metadata, ('sections', lambda _, v: v['type'] == 'VideoDetail', any))
+
+        return self.url_result(
+            self._BC_URL.format(brightcove_video_id), BrightcoveNewIE, brightcove_video_id,
+            url_transparent=True, **self._parse_metadata(video_item))
+
+
+class NetAppCollectionIE(NetAppBaseIE):
+    _VALID_URL = r'https?://media\.netapp\.com/collection/(?P<id>[0-9a-f-]+)'
+    _TESTS = [{
+        'url': 'https://media.netapp.com/collection/9820e190-f2a6-47ac-9c0a-98e5e64234a4',
+        'info_dict': {
+            'title': 'Featured sessions',
+            'id': '9820e190-f2a6-47ac-9c0a-98e5e64234a4',
+        },
+        'playlist_count': 4,
+    }]
+
+    def _entries(self, metadata):
+        for item in traverse_obj(metadata, ('items', lambda _, v: v['brightcoveVideoId'])):
+            brightcove_video_id = item['brightcoveVideoId']
+            yield self.url_result(
+                self._BC_URL.format(brightcove_video_id), BrightcoveNewIE, brightcove_video_id,
+                url_transparent=True, **self._parse_metadata(item))
+
+    def _real_extract(self, url):
+        collection_uuid = self._match_id(url)
+        metadata = self._download_json(
+            f'https://api.media.netapp.com/client/collection/{collection_uuid}', collection_uuid)
+
+        return self.playlist_result(self._entries(metadata), collection_uuid, playlist_title=metadata.get('name'))