[ORF] Support sound.orf.at, updating `ORFRadioIE`

* maintain support for xx.orf.at/player/... URLs
* add `ORFRadioCollectionIE` to support playlists in ORF Sound
* back-port and re-work `ORFPodcastIE` from https://github.com/yt-dlp/yt-dlp/pull/8486, thx Esokrates

youtube_dl/extractor/extractors.py

@@ -900,20 +900,11 @@ from .ora import OraTVIE
 from .orf import (
     ORFONIE,
     ORFONLiveIE,
-    ORFFM4IE,
     ORFFM4StoryIE,
-    ORFOE1IE,
-    ORFOE3IE,
-    ORFNOEIE,
-    ORFWIEIE,
-    ORFBGLIE,
-    ORFOOEIE,
-    ORFSTMIE,
-    ORFKTNIE,
-    ORFSBGIE,
-    ORFTIRIE,
-    ORFVBGIE,
     ORFIPTVIE,
+    ORFPodcastIE,
+    ORFRadioIE,
+    ORFRadioCollectionIE,
 )
 from .outsidetv import OutsideTVIE
 from .packtpub import (

youtube_dl/extractor/orf.py

@@ -9,17 +9,19 @@ from .common import InfoExtractor
 from ..utils import (
     clean_html,
     determine_ext,
+    ExtractorError,
     float_or_none,
     int_or_none,
     merge_dicts,
+    mimetype2ext,
     orderedSet,
     parse_age_limit,
     parse_iso8601,
     remove_end,
-    str_or_none,
     strip_jsonp,
     txt_or_none,
     unified_strdate,
+    update_url_query,
     url_or_none,
 )
 from ..traversal import T, traverse_obj
@@ -27,221 +29,280 @@ from ..traversal import T, traverse_obj
 k_float_or_none = functools.partial(float_or_none, scale=1000)
 
 
-class ORFRadioIE(InfoExtractor):
+class ORFRadioBase(InfoExtractor):
-    def _real_extract(self, url):
+    STATION_INFO = {
-        mobj = re.match(self._VALID_URL, url)
+        'fm4': ('fm4', 'fm4', 'orffm4'),
-        show_date = mobj.group('date')
+        'noe': ('noe', 'oe2n', 'orfnoe'),
-        show_id = mobj.group('show')
+        'wien': ('wie', 'oe2w', 'orfwie'),
+        'burgenland': ('bgl', 'oe2b', 'orfbgl'),
+        'ooe': ('ooe', 'oe2o', 'orfooe'),
+        'steiermark': ('stm', 'oe2st', 'orfstm'),
+        'kaernten': ('ktn', 'oe2k', 'orfktn'),
+        'salzburg': ('sbg', 'oe2s', 'orfsbg'),
+        'tirol': ('tir', 'oe2t', 'orftir'),
+        'vorarlberg': ('vbg', 'oe2v', 'orfvbg'),
+        'oe3': ('oe3', 'oe3', 'orfoe3'),
+        'oe1': ('oe1', 'oe1', 'orfoe1'),
+    }
+    _ID_NAMES = ('id', 'guid', 'program')
 
-        data = self._download_json(
+    @classmethod
-            'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s'
+    def _get_item_id(cls, data):
-            % (self._API_STATION, show_id, show_date), show_id)
+        return traverse_obj(data, *cls._ID_NAMES, expected_type=txt_or_none)
 
-        entries = []
+    @classmethod
-        for info in data['streams']:
+    def _get_api_payload(cls, data, expected_id, in_payload=False):
-            loop_stream_id = str_or_none(info.get('loopStreamId'))
+        if expected_id not in traverse_obj(data, ('payload',)[:1 if in_payload else 0] + (cls._ID_NAMES, T(txt_or_none))):
-            if not loop_stream_id:
+            raise ExtractorError('Unexpected API data result', video_id=expected_id)
-                continue
+        return data['payload']
-            title = str_or_none(data.get('title'))
+
-            if not title:
+    @staticmethod
-                continue
+    def _extract_podcast_upload(data):
-            start = int_or_none(info.get('start'), scale=1000)
+        return traverse_obj(data, {
-            end = int_or_none(info.get('end'), scale=1000)
+            'url': ('enclosures', 0, 'url'),
-            duration = end - start if end and start else None
+            'ext': ('enclosures', 0, 'type', T(mimetype2ext)),
-            entries.append({
+            'filesize': ('enclosures', 0, 'length', T(int_or_none)),
-                'id': loop_stream_id.replace('.mp3', ''),
+            'title': ('title', T(txt_or_none)),
-                'url': 'https://loopstream01.apa.at/?channel=%s&id=%s' % (self._LOOP_STATION, loop_stream_id),
+            'description': ('description', T(clean_html)),
-                'title': title,
+            'timestamp': (('published', 'postDate'), T(parse_iso8601)),
-                'description': clean_html(data.get('subtitle')),
+            'duration': ('duration', T(k_float_or_none)),
-                'duration': duration,
+            'series': ('podcast', 'title'),
-                'timestamp': start,
+            'uploader': ((('podcast', 'author'), 'station'), T(txt_or_none)),
+            'uploader_id': ('podcast', 'channel', T(txt_or_none)),
+        }, get_all=False)
+
+    @classmethod
+    def _entries(cls, data, station, item_type=None):
+        if item_type in ('upload', 'podcast-episode'):
+            yield merge_dicts({
+                'id': cls._get_item_id(data),
                 'ext': 'mp3',
-                'series': data.get('programTitle'),
+                'vcodec': 'none',
-            })
+            }, cls._extract_podcast_upload(data), rev=True)
+            return
+
+        loop_station = cls.STATION_INFO[station][1]
+        for info in traverse_obj(data, ((('streams', Ellipsis), 'stream'), T(lambda v: v if v['loopStreamId'] else None))):
+            item_id = info['loopStreamId']
+            host = info.get('host') or 'loopstream01.apa.at'
+            yield merge_dicts({
+                'id': item_id.replace('.mp3', ''),
+                'ext': 'mp3',
+                'url': update_url_query('https://{0}/'.format(host), {
+                    'channel': loop_station,
+                    'id': item_id,
+                }),
+                'vcodec': 'none',
+                # '_old_archive_ids': [make_archive_id(old_ie, video_id)],
+            }, traverse_obj(data, {
+                'title': ('title', T(txt_or_none)),
+                'description': ('subtitle', T(clean_html)),
+                'uploader': 'station',
+                'series': ('programTitle', T(txt_or_none)),
+            }), traverse_obj(info, {
+                'duration': (('duration',
+                              (None, T(lambda x: x['end'] - x['start']))),
+                             T(k_float_or_none), any),
+                'timestamp': (('start', 'startISO'), T(parse_iso8601), any),
+            }))
 
-        return {
-            '_type': 'playlist',
-            'id': show_id,
-            'title': data.get('title'),
-            'description': clean_html(data.get('subtitle')),
-            'entries': entries,
-        }
 
+class ORFRadioIE(ORFRadioBase):
+    IE_NAME = 'orf:sound'
+    _STATION_RE = '|'.join(map(re.escape, ORFRadioBase.STATION_INFO.keys()))
 
-class ORFFM4IE(ORFRadioIE):
+    _VALID_URL = (
-    IE_NAME = 'orf:fm4'
+        r'https?://sound\.orf\.at/radio/(?P<station>{0})/sendung/(?P<id>\d+)(?:/(?P<show>\w+))?'.format(_STATION_RE),
-    IE_DESC = 'radio FM4'
+        r'https?://(?P<station>{0})\.orf\.at/player/(?P<date>\d{{8}})/(?P<id>\d+)'.format(_STATION_RE),
-    _VALID_URL = r'https?://(?P<station>fm4)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>4\w+)'
+    )
-    _API_STATION = 'fm4'
-    _LOOP_STATION = 'fm4'
 
-    _TEST = {
+    _TESTS = [{
-        'url': 'http://fm4.orf.at/player/20170107/4CC',
+        'url': 'https://sound.orf.at/radio/ooe/sendung/37802/guten-morgen-oberoesterreich-am-feiertag',
-        'md5': '2b0be47375432a7ef104453432a19212',
+        'info_dict': {
+            'id': '37802',
+            'title': 'Guten Morgen Oberösterreich am Feiertag',
+            'description': 'Oberösterreichs meistgehörte regionale Frühsendung.\nRegionale Nachrichten zu jeder halben Stunde.\nModeration: Wolfgang Lehner\nNachrichten:  Stephan Schnabl',
+        },
+        'playlist': [{
+            'md5': 'f9ff8517dd681b642a2c900e2c9e6085',
+            'info_dict': {
+                'id': '2024-05-30_0559_tl_66_7DaysThu1_443862',
+                'ext': 'mp3',
+                'title': 'Guten Morgen Oberösterreich am Feiertag',
+                'description': 'Oberösterreichs meistgehörte regionale Frühsendung.\nRegionale Nachrichten zu jeder halben Stunde.\nModeration: Wolfgang Lehner\nNachrichten:  Stephan Schnabl',
+                'timestamp': 1717041587,
+                'upload_date': '20240530',
+                'uploader': 'ooe',
+                'duration': 14413.0,
+            }
+        }],
+        # 'skip': 'Shows from ORF Sound are only available for 30 days.'
+    }, {
+        'url': 'https://oe1.orf.at/player/20240531/758136',
+        'md5': '2397717aaf3ae9c22a4f090ee3b8d374',
         'info_dict': {
-            'id': '2017-01-07_2100_tl_54_7DaysSat18_31295',
+            'id': '2024-05-31_1905_tl_51_7DaysFri35_2413387',
             'ext': 'mp3',
-            'title': 'Solid Steel Radioshow',
+            'title': '"Who Cares?"',
-            'description': 'Die Mixshow von Coldcut und Ninja Tune.',
+            'description': 'Europas größte Netzkonferenz re:publica 2024',
-            'duration': 3599,
+            'timestamp': 1717175100,
-            'timestamp': 1483819257,
+            'upload_date': '20240531',
-            'upload_date': '20170107',
+            'uploader': 'oe1',
+            'duration': 1500,
         },
-        'skip': 'Shows from ORF radios are only available for 7 days.',
+        # 'skip': 'Shows from ORF Sound are only available for 30 days.'
-        'only_matching': True,
+    }]
-    }
-
-
-class ORFNOEIE(ORFRadioIE):
-    IE_NAME = 'orf:noe'
-    IE_DESC = 'Radio Niederösterreich'
-    _VALID_URL = r'https?://(?P<station>noe)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'noe'
-    _LOOP_STATION = 'oe2n'
-
-    _TEST = {
-        'url': 'https://noe.orf.at/player/20200423/NGM',
-        'only_matching': True,
-    }
-
-
-class ORFWIEIE(ORFRadioIE):
-    IE_NAME = 'orf:wien'
-    IE_DESC = 'Radio Wien'
-    _VALID_URL = r'https?://(?P<station>wien)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'wie'
-    _LOOP_STATION = 'oe2w'
-
-    _TEST = {
-        'url': 'https://wien.orf.at/player/20200423/WGUM',
-        'only_matching': True,
-    }
-
-
-class ORFBGLIE(ORFRadioIE):
-    IE_NAME = 'orf:burgenland'
-    IE_DESC = 'Radio Burgenland'
-    _VALID_URL = r'https?://(?P<station>burgenland)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'bgl'
-    _LOOP_STATION = 'oe2b'
-
-    _TEST = {
-        'url': 'https://burgenland.orf.at/player/20200423/BGM',
-        'only_matching': True,
-    }
-
-
-class ORFOOEIE(ORFRadioIE):
-    IE_NAME = 'orf:oberoesterreich'
-    IE_DESC = 'Radio Oberösterreich'
-    _VALID_URL = r'https?://(?P<station>ooe)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'ooe'
-    _LOOP_STATION = 'oe2o'
-
-    _TEST = {
-        'url': 'https://ooe.orf.at/player/20200423/OGMO',
-        'only_matching': True,
-    }
-
-
-class ORFSTMIE(ORFRadioIE):
-    IE_NAME = 'orf:steiermark'
-    IE_DESC = 'Radio Steiermark'
-    _VALID_URL = r'https?://(?P<station>steiermark)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'stm'
-    _LOOP_STATION = 'oe2st'
 
-    _TEST = {
+    def _real_extract(self, url):
-        'url': 'https://steiermark.orf.at/player/20200423/STGMS',
+        m = self._match_valid_url(url)
-        'only_matching': True,
+        station, show_id = m.group('station', 'id')
-    }
+        api_station, _, _ = self.STATION_INFO[station]
+        if 'date' in m.groupdict():
+            data = self._download_json(
+                'https://audioapi.orf.at/{0}/json/4.0/broadcast/{1}/{2}?_o={3}.orf.at'.format(
+                    api_station, show_id, m.group('date'), station), show_id)
+            show_id = data['id']
+        else:
+            data = self._download_json(
+                'https://audioapi.orf.at/{0}/api/json/5.0/broadcast/{1}?_o=sound.orf.at'.format(
+                    api_station, show_id), show_id)
 
+            data = self._get_api_payload(data, show_id, in_payload=True)
 
-class ORFKTNIE(ORFRadioIE):
+        # site sends ISO8601 GMT date-times with separate TZ offset, ignored
-    IE_NAME = 'orf:kaernten'
+        # TODO: should `..._date` be calculated relative to TZ?
-    IE_DESC = 'Radio Kärnten'
-    _VALID_URL = r'https?://(?P<station>kaernten)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'ktn'
-    _LOOP_STATION = 'oe2k'
 
-    _TEST = {
+        return merge_dicts(
-        'url': 'https://kaernten.orf.at/player/20200423/KGUMO',
+            {'_type': 'multi_video'},
-        'only_matching': True,
+            self.playlist_result(
-    }
+                self._entries(data, station), show_id,
+                txt_or_none(data.get('title')),
+                clean_html(data.get('subtitle'))))
 
 
-class ORFSBGIE(ORFRadioIE):
+class ORFRadioCollectionIE(ORFRadioBase):
-    IE_NAME = 'orf:salzburg'
+    IE_NAME = 'orf:collection'
-    IE_DESC = 'Radio Salzburg'
+    _VALID_URL = r'https?://sound\.orf\.at/collection/(?P<coll_id>\d+)(?:/(?P<item_id>\d+))?'
-    _VALID_URL = r'https?://(?P<station>salzburg)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'sbg'
-    _LOOP_STATION = 'oe2s'
-
-    _TEST = {
-        'url': 'https://salzburg.orf.at/player/20200423/SGUM',
-        'only_matching': True,
-    }
 
+    _TESTS = [{
+        'url': 'https://sound.orf.at/collection/4/61908/was-das-uberschreiten-des-15-limits-bedeutet',
+        'info_dict': {
+            'id': '2577582',
+        },
+        'playlist': [{
+            'md5': '5789cec7d75575ff58d19c0428c80eb3',
+            'info_dict': {
+                'id': '2024-06-06_1659_tl_54_7DaysThu6_153926',
+                'ext': 'mp3',
+                'title': 'Klimakrise: Was das Überschreiten des 1,5°-Limits bedeutet',
+                'timestamp': 1717686674,
+                'upload_date': '20240606',
+                'uploader': 'fm4',
+            },
+        }],
+        # 'skip': 'Shows from ORF Sound are only available for 30 days.'
+    }, {
+        'url': 'https://sound.orf.at/collection/4/',
+        'info_dict': {
+            'id': '4',
+        },
+        'playlist_mincount': 10,
+        'playlist_maxcount': 13,
+    }]
 
-class ORFTIRIE(ORFRadioIE):
+    def _real_extract(self, url):
-    IE_NAME = 'orf:tirol'
+        coll_id, item_id = self._match_valid_url(url).group('coll_id', 'item_id')
-    IE_DESC = 'Radio Tirol'
+        data = self._download_json(
-    _VALID_URL = r'https?://(?P<station>tirol)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
+            'https://collector.orf.at/api/frontend/collections/{0}?_o=sound.orf.at'.format(
-    _API_STATION = 'tir'
+                coll_id), coll_id)
-    _LOOP_STATION = 'oe2t'
+        data = self._get_api_payload(data, coll_id, in_payload=True)
+
+        def yield_items():
+            for item in traverse_obj(data, (
+                    'content', 'items', lambda _, v: any(k in v['target']['params'] for k in self._ID_NAMES))):
+                if item_id is None or item_id == txt_or_none(item.get('id')):
+                    target = item['target']
+                    typed_item_id = self._get_item_id(target['params'])
+                    station = target['params'].get('station')
+                    item_type = target.get('type')
+                    if typed_item_id and (station or item_type):
+                        yield station, typed_item_id, item_type
+                    if item_id is not None:
+                        break
+            else:
+                if item_id is not None:
+                    raise ExtractorError('Item not found in collection',
+                                         video_id=coll_id, expected=True)
+
+        def item_playlist(station, typed_item_id, item_type):
+            if item_type == 'upload':
+                item_data = self._download_json('https://audioapi.orf.at/radiothek/api/2.0/upload/{0}?_o=sound.orf.at'.format(
+                    typed_item_id), typed_item_id)
+            elif item_type == 'podcast-episode':
+                item_data = self._download_json('https://audioapi.orf.at/radiothek/api/2.0/episode/{0}?_o=sound.orf.at'.format(
+                    typed_item_id), typed_item_id)
+            else:
+                api_station, _, _ = self.STATION_INFO[station]
+                item_data = self._download_json(
+                    'https://audioapi.orf.at/{0}/api/json/5.0/{1}/{2}?_o=sound.orf.at'.format(
+                        api_station, item_type or 'broadcastitem', typed_item_id), typed_item_id)
 
-    _TEST = {
+            item_data = self._get_api_payload(item_data, typed_item_id, in_payload=True)
-        'url': 'https://tirol.orf.at/player/20200423/TGUMO',
-        'only_matching': True,
-    }
 
+            return merge_dicts(
+                {'_type': 'multi_video'},
+                self.playlist_result(
+                    self._entries(item_data, station, item_type), typed_item_id,
+                    txt_or_none(data.get('title')),
+                    clean_html(data.get('subtitle'))))
 
-class ORFVBGIE(ORFRadioIE):
+        def yield_item_entries():
-    IE_NAME = 'orf:vorarlberg'
+            for station, typed_id, item_type in yield_items():
-    IE_DESC = 'Radio Vorarlberg'
+                yield item_playlist(station, typed_id, item_type)
-    _VALID_URL = r'https?://(?P<station>vorarlberg)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'vbg'
-    _LOOP_STATION = 'oe2v'
 
-    _TEST = {
+        if item_id is not None:
-        'url': 'https://vorarlberg.orf.at/player/20200423/VGUM',
+            # coll_id = '/'.join((coll_id, item_id))
-        'only_matching': True,
+            return next(yield_item_entries())
-    }
 
+        return self.playlist_result(yield_item_entries(), coll_id, data.get('title'))
 
-class ORFOE3IE(ORFRadioIE):
-    IE_NAME = 'orf:oe3'
-    IE_DESC = 'Radio Österreich 3'
-    _VALID_URL = r'https?://(?P<station>oe3)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
-    _API_STATION = 'oe3'
-    _LOOP_STATION = 'oe3'
 
-    _TEST = {
+class ORFPodcastIE(ORFRadioBase):
-        'url': 'https://oe3.orf.at/player/20200424/3WEK',
+    IE_NAME = 'orf:podcast'
-        'only_matching': True,
+    _STATION_RE = '|'.join(map(re.escape, (x[0] for x in ORFRadioBase.STATION_INFO.values()))) + '|tv'
-    }
+    _VALID_URL = r'https?://sound\.orf\.at/podcast/(?P<station>{0})/(?P<show>[\w-]+)/(?P<id>[\w-]+)'.format(_STATION_RE)
+    _TESTS = [{
+        'url': 'https://sound.orf.at/podcast/stm/der-kraeutertipp-von-christine-lackner/rotklee',
+        'md5': '1f2bab2ba90c2ce0c2754196ea78b35f',
+        'info_dict': {
+            'id': 'der-kraeutertipp-von-christine-lackner/rotklee',
+            'ext': 'mp3',
+            'title': 'Rotklee',
+            'description': 'In der Natur weit verbreitet - in der Medizin längst anerkennt: Rotklee. Dieser Podcast begleitet die Sendung "Radio Steiermark am Vormittag", Radio Steiermark, 28. Mai 2024.',
+            'timestamp': 1716891761,
+            'upload_date': '20240528',
+            'uploader_id': 'stm_kraeutertipp',
+            'uploader': 'ORF Radio Steiermark',
+            'duration': 101,
+            'series': 'Der Kräutertipp von Christine Lackner',
+        },
+        # 'skip': 'ORF podcasts are only available for a limited time'
+    }]
 
+    _ID_NAMES = ('slug', 'guid')
 
-class ORFOE1IE(ORFRadioIE):
+    def _real_extract(self, url):
-    IE_NAME = 'orf:oe1'
+        station, show, show_id = self._match_valid_url(url).group('station', 'show', 'id')
-    IE_DESC = 'Radio Österreich 1'
+        data = self._download_json(
-    _VALID_URL = r'https?://(?P<station>oe1)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
+            'https://audioapi.orf.at/radiothek/api/2.0/podcast/{0}/{1}/{2}'.format(
-    _API_STATION = 'oe1'
+                station, show, show_id), show_id)
-    _LOOP_STATION = 'oe1'
+        data = self._get_api_payload(data, show_id, in_payload=True)
 
-    _TEST = {
+        return merge_dicts({
-        'url': 'http://oe1.orf.at/player/20170108/456544',
+            'id': '/'.join((show, show_id)),
-        'md5': '34d8a6e67ea888293741c86a099b745b',
-        'info_dict': {
-            'id': '2017-01-08_0759_tl_51_7DaysSun6_256141',
             'ext': 'mp3',
-            'title': 'Morgenjournal',
+            'vcodec': 'none',
-            'duration': 609,
+        }, self._extract_podcast_upload(data), rev=True)
-            'timestamp': 1483858796,
-            'upload_date': '20170108',
-        },
-        'skip': 'Shows from ORF radios are only available for 7 days.'
-    }
 
 
 class ORFIPTVIE(InfoExtractor):