]> jfr.im git - yt-dlp.git/blobdiff - yt_dlp/extractor/sportdeutschland.py
[ie/crunchyroll] Fix stream extraction (#10005)
[yt-dlp.git] / yt_dlp / extractor / sportdeutschland.py
index 3e497a9393ea99dc1962848ce1020e0ac5b59de3..30dbcf370a80aaae700454f425b9dd9de283000f 100644 (file)
-# coding: utf-8
-from __future__ import unicode_literals
-
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_urllib_parse_urlparse,
-)
 from ..utils import (
-    clean_html,
-    float_or_none,
-    int_or_none,
-    parse_iso8601,
+    join_nonempty,
     strip_or_none,
-    try_get,
+    traverse_obj,
+    unified_timestamp,
 )
 
 
 class SportDeutschlandIE(InfoExtractor):
     _VALID_URL = r'https?://sportdeutschland\.tv/(?P<id>(?:[^/]+/)?[^?#/&]+)'
     _TESTS = [{
-        'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0',
+        'url': 'https://sportdeutschland.tv/blauweissbuchholztanzsport/buchholzer-formationswochenende-2023-samstag-1-bundesliga-landesliga',
         'info_dict': {
-            'id': '5318cac0275701382770543d7edaf0a0',
+            'id': '9839a5c7-0dbb-48a8-ab63-3b408adc7b54',
             'ext': 'mp4',
-            'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals - Teil 1',
-            'duration': 16106.36,
-        },
-        'params': {
-            'noplaylist': True,
-            # m3u8 download
-            'skip_download': True,
-        },
+            'title': 'Buchholzer Formationswochenende 2023 - Samstag - 1. Bundesliga / Landesliga',
+            'display_id': 'blauweissbuchholztanzsport/buchholzer-formationswochenende-2023-samstag-1-bundesliga-landesliga',
+            'description': 'md5:a288c794a5ee69e200d8f12982f81a87',
+            'live_status': 'was_live',
+            'channel': 'Blau-Weiss Buchholz Tanzsport',
+            'channel_url': 'https://sportdeutschland.tv/blauweissbuchholztanzsport',
+            'channel_id': '93ec33c9-48be-43b6-b404-e016b64fdfa3',
+            'duration': 32447,
+            'upload_date': '20230114',
+            'timestamp': 1673733618,
+        }
+    }, {
+        'url': 'https://sportdeutschland.tv/deutscherbadmintonverband/bwf-tour-1-runde-feld-1-yonex-gainward-german-open-2022-0',
+        'info_dict': {
+            'id': '95c80c52-6b9a-4ae9-9197-984145adfced',
+            'ext': 'mp4',
+            'title': 'BWF Tour: 1. Runde Feld 1 - YONEX GAINWARD German Open 2022',
+            'display_id': 'deutscherbadmintonverband/bwf-tour-1-runde-feld-1-yonex-gainward-german-open-2022-0',
+            'description': 'md5:2afb5996ceb9ac0b2ac81f563d3a883e',
+            'live_status': 'was_live',
+            'channel': 'Deutscher Badminton Verband',
+            'channel_url': 'https://sportdeutschland.tv/deutscherbadmintonverband',
+            'channel_id': '93ca5866-2551-49fc-8424-6db35af58920',
+            'duration': 41097,
+            'upload_date': '20220309',
+            'timestamp': 1646860727.0,
+        }
     }, {
-        'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0',
+        'url': 'https://sportdeutschland.tv/ggcbremen/formationswochenende-latein-2023',
         'info_dict': {
-            'id': 'c6e2fdd01f63013854c47054d2ab776f',
-            'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals',
-            'description': 'md5:5263ff4c31c04bb780c9f91130b48530',
-            'duration': 31397,
+            'id': '9889785e-55b0-4d97-a72a-ce9a9f157cce',
+            'title': 'Formationswochenende Latein 2023 - Samstag',
+            'display_id': 'ggcbremen/formationswochenende-latein-2023',
+            'description': 'md5:6e4060d40ff6a8f8eeb471b51a8f08b2',
+            'live_status': 'was_live',
+            'channel': 'Grün-Gold-Club Bremen e.V.',
+            'channel_id': '9888f04e-bb46-4c7f-be47-df960a4167bb',
+            'channel_url': 'https://sportdeutschland.tv/ggcbremen',
         },
-        'playlist_count': 2,
+        'playlist_count': 3,
+        'playlist': [{
+            'info_dict': {
+                'id': '988e1fea-9d44-4fab-8c72-3085fb667547',
+                'ext': 'mp4',
+                'channel_url': 'https://sportdeutschland.tv/ggcbremen',
+                'channel_id': '9888f04e-bb46-4c7f-be47-df960a4167bb',
+                'channel': 'Grün-Gold-Club Bremen e.V.',
+                'duration': 86,
+                'title': 'Formationswochenende Latein 2023 - Samstag Part 1',
+                'upload_date': '20230225',
+                'timestamp': 1677349909,
+                'live_status': 'was_live',
+            }
+        }]
     }, {
-        'url': 'https://sportdeutschland.tv/freeride-world-tour-2021-fieberbrunn-oesterreich',
-        'only_matching': True,
+        'url': 'https://sportdeutschland.tv/dtb/gymnastik-international-tag-1',
+        'info_dict': {
+            'id': '95d71b8a-370a-4b87-ad16-94680da18528',
+            'ext': 'mp4',
+            'title': r're:Gymnastik International - Tag 1 .+',
+            'display_id': 'dtb/gymnastik-international-tag-1',
+            'channel_id': '936ecef1-2f4a-4e08-be2f-68073cb7ecab',
+            'channel': 'Deutscher Turner-Bund',
+            'channel_url': 'https://sportdeutschland.tv/dtb',
+            'description': 'md5:07a885dde5838a6f0796ee21dc3b0c52',
+            'live_status': 'is_live',
+        },
+        'skip': 'live',
     }]
 
+    def _process_video(self, asset_id, video):
+        is_live = video['type'] == 'mux_live'
+        token = self._download_json(
+            f'https://api.sportdeutschland.tv/api/frontend/asset-token/{asset_id}',
+            video['id'], query={'type': video['type'], 'playback_id': video['src']})['token']
+        formats, subtitles = self._extract_m3u8_formats_and_subtitles(
+            f'https://stream.mux.com/{video["src"]}.m3u8?token={token}', video['id'], live=is_live)
+
+        return {
+            'is_live': is_live,
+            'formats': formats,
+            'subtitles': subtitles,
+            **traverse_obj(video, {
+                'id': 'id',
+                'duration': ('duration', {lambda x: float(x) > 0 and float(x)}),
+                'timestamp': ('created_at', {unified_timestamp})
+            }),
+        }
+
     def _real_extract(self, url):
         display_id = self._match_id(url)
-        data = self._download_json(
-            'https://backend.sportdeutschland.tv/api/permalinks/' + display_id,
+        meta = self._download_json(
+            f'https://api.sportdeutschland.tv/api/stateless/frontend/assets/{display_id}',
             display_id, query={'access_token': 'true'})
-        asset = data['asset']
-        title = (asset.get('title') or asset['label']).strip()
-        asset_id = asset.get('id') or asset.get('uuid')
+
         info = {
-            'id': asset_id,
-            'title': title,
-            'description': clean_html(asset.get('body') or asset.get('description')) or asset.get('teaser'),
-            'duration': int_or_none(asset.get('seconds')),
+            'display_id': display_id,
+            **traverse_obj(meta, {
+                'id': (('id', 'uuid'), ),
+                'title': (('title', 'name'), {strip_or_none}),
+                'description': 'description',
+                'channel': ('profile', 'name'),
+                'channel_id': ('profile', 'id'),
+                'is_live': 'currently_live',
+                'was_live': 'was_live',
+                'channel_url': ('profile', 'slug', {lambda x: f'https://sportdeutschland.tv/{x}'}),
+            }, get_all=False)
         }
-        videos = asset.get('videos') or []
-        if len(videos) > 1:
-            playlist_id = compat_parse_qs(compat_urllib_parse_urlparse(url).query).get('playlistId', [None])[0]
-            if playlist_id:
-                if self._downloader.params.get('noplaylist'):
-                    videos = [videos[int(playlist_id)]]
-                    self.to_screen('Downloading just a single video because of --no-playlist')
-                else:
-                    self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % asset_id)
 
-            def entries():
-                for i, video in enumerate(videos, 1):
-                    video_id = video.get('uuid')
-                    video_url = video.get('url')
-                    if not (video_id and video_url):
-                        continue
-                    formats = self._extract_m3u8_formats(
-                        video_url.replace('.smil', '.m3u8'), video_id, 'mp4', fatal=False)
-                    if not formats:
-                        continue
-                    yield {
-                        'id': video_id,
-                        'formats': formats,
-                        'title': title + ' - ' + (video.get('label') or 'Teil %d' % i),
-                        'duration': float_or_none(video.get('duration')),
-                    }
-            info.update({
-                '_type': 'multi_video',
-                'entries': entries(),
-            })
-        else:
-            formats = self._extract_m3u8_formats(
-                videos[0]['url'].replace('.smil', '.m3u8'), asset_id, 'mp4')
-            section_title = strip_or_none(try_get(data, lambda x: x['section']['title']))
-            info.update({
-                'formats': formats,
-                'display_id': asset.get('permalink'),
-                'thumbnail': try_get(asset, lambda x: x['images'][0]),
-                'categories': [section_title] if section_title else None,
-                'view_count': int_or_none(asset.get('views')),
-                'is_live': asset.get('is_live') is True,
-                'timestamp': parse_iso8601(asset.get('date') or asset.get('published_at')),
-            })
-        return info
+        parts = traverse_obj(meta, (('livestream', ('videos', ...)), ))
+        entries = [{
+            'title': join_nonempty(info.get('title'), f'Part {i}', delim=' '),
+            **traverse_obj(info, {'channel': 'channel', 'channel_id': 'channel_id',
+                                  'channel_url': 'channel_url', 'was_live': 'was_live'}),
+            **self._process_video(info['id'], video),
+        } for i, video in enumerate(parts, 1)]
+
+        return {
+            '_type': 'multi_video',
+            **info,
+            'entries': entries,
+        } if len(entries) > 1 else {
+            **info,
+            **entries[0],
+            'title': info.get('title'),
+        }