[sportdeutschland] fix extraction(closes #21856)(closes #28425)

youtube_dl/extractor/sportdeutschland.py

@@ -1,82 +1,105 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 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,
-    sanitized_Request,
+    strip_or_none,
+    try_get,
 )
 
 
 class SportDeutschlandIE(InfoExtractor):
-    _VALID_URL = r'https?://sportdeutschland\.tv/(?P<sport>[^/?#]+)/(?P<id>[^?#/]+)(?:$|[?#])'
+    _VALID_URL = r'https?://sportdeutschland\.tv/(?P<id>(?:[^/]+/)?[^?#/&]+)'
     _TESTS = [{
         'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0',
         'info_dict': {
-            'id': 're-live-deutsche-meisterschaften-2020-halbfinals',
+            'id': '5318cac0275701382770543d7edaf0a0',
             'ext': 'mp4',
-            'title': 're:Re-live: Deutsche Meisterschaften 2020.*Halbfinals',
+            'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals - Teil 1',
-            'categories': ['Badminton-Deutschland'],
+            'duration': 16106.36,
-            'view_count': int,
-            'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
-            'timestamp': int,
-            'upload_date': '20200201',
-            'description': 're:.*',  # meaningless description for THIS video
         },
+        'params': {
+            'noplaylist': True,
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0',
+        'info_dict': {
+            'id': 'c6e2fdd01f63013854c47054d2ab776f',
+            'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals',
+            'description': 'md5:5263ff4c31c04bb780c9f91130b48530',
+            'duration': 31397,
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://sportdeutschland.tv/freeride-world-tour-2021-fieberbrunn-oesterreich',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        display_id = self._match_id(url)
-        video_id = mobj.group('id')
+        data = self._download_json(
-        sport_id = mobj.group('sport')
+            'https://backend.sportdeutschland.tv/api/permalinks/' + display_id,
-
+            display_id, query={'access_token': 'true'})
-        api_url = 'https://proxy.vidibusdynamic.net/ssl/backend.sportdeutschland.tv/api/permalinks/%s/%s?access_token=true' % (
-            sport_id, video_id)
-        req = sanitized_Request(api_url, headers={
-            'Accept': 'application/vnd.vidibus.v2.html+json',
-            'Referer': url,
-        })
-        data = self._download_json(req, video_id)
-
         asset = data['asset']
-        categories = [data['section']['title']]
+        title = (asset.get('title') or asset['label']).strip()
-
+        asset_id = asset.get('id') or asset.get('uuid')
-        formats = []
+        info = {
-        smil_url = asset['video']
+            'id': asset_id,
-        if '.smil' in smil_url:
+            'title': title,
-            m3u8_url = smil_url.replace('.smil', '.m3u8')
+            'description': clean_html(asset.get('body') or asset.get('description')) or asset.get('teaser'),
-            formats.extend(
+            'duration': int_or_none(asset.get('seconds')),
-                self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4'))
+        }
-
+        videos = asset.get('videos') or []
-            smil_doc = self._download_xml(
+        if len(videos) > 1:
-                smil_url, video_id, note='Downloading SMIL metadata')
+            playlist_id = compat_parse_qs(compat_urllib_parse_urlparse(url).query).get('playlistId', [None])[0]
-            base_url_el = smil_doc.find('./head/meta')
+            if playlist_id:
-            if base_url_el:
+                if self._downloader.params.get('noplaylist'):
-                base_url = base_url_el.attrib['base']
+                    videos = [videos[int(playlist_id)]]
-            formats.extend([{
+                    self.to_screen('Downloading just a single video because of --no-playlist')
-                'format_id': 'rmtp',
-                'url': base_url if base_url_el else n.attrib['src'],
-                'play_path': n.attrib['src'],
-                'ext': 'flv',
-                'preference': -100,
-                'format_note': 'Seems to fail at example stream',
-            } for n in smil_doc.findall('./body/video')])
                 else:
-            formats.append({'url': smil_url})
+                    self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % asset_id)
 
-        self._sort_formats(formats)
+            def entries():
-
+                for i, video in enumerate(videos, 1):
-        return {
+                    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': asset['title'],
+                        'title': title + ' - ' + (video.get('label') or 'Teil %d' % i),
-            'thumbnail': asset.get('image'),
+                        'duration': float_or_none(video.get('duration')),
-            'description': asset.get('teaser'),
-            'duration': asset.get('duration'),
-            'categories': categories,
-            'view_count': asset.get('views'),
-            'rtmp_live': asset.get('live'),
-            'timestamp': parse_iso8601(asset.get('date')),
                     }
+            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