[ie/KiKA] Add extractor (#5788)

Authored by: 1100101

yt_dlp/extractor/_extractors.py

@@ -945,6 +945,7 @@ from .kick import (
 )
 from .kicker import KickerIE
 from .kickstarter import KickStarterIE
+from .kika import KikaIE
 from .kinja import KinjaEmbedIE
 from .kinopoisk import KinoPoiskIE
 from .kommunetv import KommunetvIE

yt_dlp/extractor/kika.py

@@ -0,0 +1,126 @@
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class KikaIE(InfoExtractor):
+    IE_DESC = 'KiKA.de'
+    _VALID_URL = r'https?://(?:www\.)?kika\.de/[\w/-]+/videos/(?P<id>[a-z-]+\d+)'
+    _GEO_COUNTRIES = ['DE']
+
+    _TESTS = [{
+        'url': 'https://www.kika.de/logo/videos/logo-vom-samstag-einunddreissig-august-zweitausendvierundzwanzig-100',
+        'md5': 'fbfc8da483719ef06f396e5e5b938c69',
+        'info_dict': {
+            'id': 'logo-vom-samstag-einunddreissig-august-zweitausendvierundzwanzig-100',
+            'ext': 'mp4',
+            'upload_date': '20240831',
+            'timestamp': 1725126600,
+            'season_number': 2024,
+            'modified_date': '20240831',
+            'episode': 'Episode 476',
+            'episode_number': 476,
+            'season': 'Season 2024',
+            'duration': 634,
+            'title': 'logo! vom Samstag, 31. August 2024',
+            'modified_timestamp': 1725129983,
+        },
+    }, {
+        'url': 'https://www.kika.de/kaltstart/videos/video92498',
+        'md5': '710ece827e5055094afeb474beacb7aa',
+        'info_dict': {
+            'id': 'video92498',
+            'ext': 'mp4',
+            'title': '7. Wo ist Leo?',
+            'description': 'md5:fb48396a5b75068bcac1df74f1524920',
+            'duration': 436,
+            'timestamp': 1702926876,
+            'upload_date': '20231218',
+            'episode_number': 7,
+            'modified_date': '20240319',
+            'modified_timestamp': 1710880610,
+            'episode': 'Episode 7',
+            'season_number': 1,
+            'season': 'Season 1',
+        },
+    }, {
+        'url': 'https://www.kika.de/bernd-das-brot/astrobrot/videos/video90088',
+        'md5': 'ffd1b700d7de0a6616a1d08544c77294',
+        'info_dict': {
+            'id': 'video90088',
+            'ext': 'mp4',
+            'upload_date': '20221102',
+            'timestamp': 1667390580,
+            'duration': 197,
+            'modified_timestamp': 1711093771,
+            'episode_number': 8,
+            'title': 'Es ist nicht leicht, ein Astrobrot zu sein',
+            'modified_date': '20240322',
+            'description': 'md5:d3641deaf1b5515a160788b2be4159a9',
+            'season_number': 1,
+            'episode': 'Episode 8',
+            'season': 'Season 1',
+        },
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        doc = self._download_json(f'https://www.kika.de/_next-api/proxy/v1/videos/{video_id}', video_id)
+        video_assets = self._download_json(doc['assets']['url'], video_id)
+
+        subtitles = {}
+        if ttml_resource := url_or_none(video_assets.get('videoSubtitle')):
+            subtitles['de'] = [{
+                'url': ttml_resource,
+                'ext': 'ttml',
+            }]
+        if webvtt_resource := url_or_none(video_assets.get('webvttUrl')):
+            subtitles.setdefault('de', []).append({
+                'url': webvtt_resource,
+                'ext': 'vtt',
+            })
+
+        return {
+            'id': video_id,
+            'formats': list(self._extract_formats(video_assets, video_id)),
+            'subtitles': subtitles,
+            **traverse_obj(doc, {
+                'title': ('title', {str}),
+                'description': ('description', {str}),
+                'timestamp': ('date', {parse_iso8601}),
+                'modified_timestamp': ('modificationDate', {parse_iso8601}),
+                'duration': ((
+                    ('durationInSeconds', {int_or_none}),
+                    ('duration', {parse_duration})), any),
+                'episode_number': ('episodeNumber', {int_or_none}),
+                'season_number': ('season', {int_or_none}),
+            }),
+        }
+
+    def _extract_formats(self, media_info, video_id):
+        for media in traverse_obj(media_info, ('assets', lambda _, v: url_or_none(v['url']))):
+            stream_url = media['url']
+            ext = determine_ext(stream_url)
+            if ext == 'm3u8':
+                yield from self._extract_m3u8_formats(
+                    stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
+            else:
+                yield {
+                    'url': stream_url,
+                    'format_id': ext,
+                    **traverse_obj(media, {
+                        'width': ('frameWidth', {int_or_none}),
+                        'height': ('frameHeight', {int_or_none}),
+                        # NB: filesize is 0 if unknown, bitrate is -1 if unknown
+                        'filesize': ('fileSize', {int_or_none}, {lambda x: x or None}),
+                        'abr': ('bitrateAudio', {int_or_none}, {lambda x: None if x == -1 else x}),
+                        'vbr': ('bitrateVideo', {int_or_none}, {lambda x: None if x == -1 else x}),
+                    }),
+                }

yt_dlp/extractor/mdr.py

@@ -13,8 +13,8 @@ from ..utils import (
 
 
 class MDRIE(InfoExtractor):
-    IE_DESC = 'MDR.DE and KiKA'
+    IE_DESC = 'MDR.DE'
-    _VALID_URL = r'https?://(?:www\.)?(?:mdr|kika)\.de/(?:.*)/[a-z-]+-?(?P<id>\d+)(?:_.+?)?\.html'
+    _VALID_URL = r'https?://(?:www\.)?mdr\.de/(?:.*)/[a-z-]+-?(?P<id>\d+)(?:_.+?)?\.html'
 
     _GEO_COUNTRIES = ['DE']
 
@@ -34,30 +34,6 @@ class MDRIE(InfoExtractor):
             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
         },
         'skip': '404 not found',
-    }, {
-        'url': 'http://www.kika.de/baumhaus/videos/video19636.html',
-        'md5': '4930515e36b06c111213e80d1e4aad0e',
-        'info_dict': {
-            'id': '19636',
-            'ext': 'mp4',
-            'title': 'Baumhaus vom 30. Oktober 2015',
-            'duration': 134,
-            'uploader': 'KIKA',
-        },
-        'skip': '404 not found',
-    }, {
-        'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/videos/video8182.html',
-        'md5': '5fe9c4dd7d71e3b238f04b8fdd588357',
-        'info_dict': {
-            'id': '8182',
-            'ext': 'mp4',
-            'title': 'Beutolomäus und der geheime Weihnachtswunsch',
-            'description': 'md5:b69d32d7b2c55cbe86945ab309d39bbd',
-            'timestamp': 1482541200,
-            'upload_date': '20161224',
-            'duration': 4628,
-            'uploader': 'KIKA',
-        },
     }, {
         # audio with alternative playerURL pattern
         'url': 'http://www.mdr.de/kultur/videos-und-audios/audio-radio/operation-mindfuck-robert-wilson100.html',
@@ -68,28 +44,7 @@ class MDRIE(InfoExtractor):
             'duration': 3239,
             'uploader': 'MITTELDEUTSCHER RUNDFUNK',
         },
-    }, {
+        'skip': '404 not found',
-        # empty bitrateVideo and bitrateAudio
-        'url': 'https://www.kika.de/filme/sendung128372_zc-572e3f45_zs-1d9fb70e.html',
-        'info_dict': {
-            'id': '128372',
-            'ext': 'mp4',
-            'title': 'Der kleine Wichtel kehrt zurück',
-            'description': 'md5:f77fafdff90f7aa1e9dca14f662c052a',
-            'duration': 4876,
-            'timestamp': 1607823300,
-            'upload_date': '20201213',
-            'uploader': 'ZDF',
-        },
-        'params': {
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://www.kika.de/baumhaus/sendungen/video19636_zc-fea7f8a0_zs-4bf89c60.html',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.kika.de/sendungen/einzelsendungen/weihnachtsprogramm/einzelsendung2534.html',
-        'only_matching': True,
     }, {
         'url': 'http://www.mdr.de/mediathek/mdr-videos/a/video-1334.html',
         'only_matching': True,