[nrk] Improve extraction (closes #27634, closes #27635)
authorSergey M․ <dstftw@gmail.com>
Sat, 2 Jan 2021 23:29:52 +0000 (06:29 +0700)
committerSergey M․ <dstftw@gmail.com>
Sat, 2 Jan 2021 23:29:52 +0000 (06:29 +0700)
+ Add support for mp3 formats
* Generalize and delegate all item extractors to nrk, beware ie key breakages
+ Add support for podcasts
+ Generalize nrk shortcut form to support all kind of ids

youtube_dl/extractor/extractors.py
youtube_dl/extractor/nrk.py

index d9d3f4940a11dfaeacdaa36405ed9dadfb0c6531..74743a44915e5725e6c3d32408e8504d5c0e897c 100644 (file)
@@ -790,6 +790,7 @@ from .nrk import (
     NRKSkoleIE,
     NRKTVIE,
     NRKTVDirekteIE,
+    NRKRadioPodkastIE,
     NRKTVEpisodeIE,
     NRKTVEpisodesIE,
     NRKTVSeasonIE,
index 871e4845cad20532507c62acd5a33d104c521296..9621522d43148296e0a73f0b254d032e6d5a0491 100644 (file)
@@ -62,6 +62,75 @@ class NRKBaseIE(InfoExtractor):
             fatal=fatal, query=query,
             headers={'Accept-Encoding': 'gzip, deflate, br'})
 
+
+class NRKIE(NRKBaseIE):
+    _VALID_URL = r'''(?x)
+                        (?:
+                            nrk:|
+                            https?://
+                                (?:
+                                    (?:www\.)?nrk\.no/video/(?:PS\*|[^_]+_)|
+                                    v8[-.]psapi\.nrk\.no/mediaelement/
+                                )
+                            )
+                            (?P<id>[^?\#&]+)
+                        '''
+
+    _TESTS = [{
+        # video
+        'url': 'http://www.nrk.no/video/PS*150533',
+        'md5': 'f46be075326e23ad0e524edfcb06aeb6',
+        'info_dict': {
+            'id': '150533',
+            'ext': 'mp4',
+            'title': 'Dompap og andre fugler i Piip-Show',
+            'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
+            'duration': 262,
+        }
+    }, {
+        # audio
+        'url': 'http://www.nrk.no/video/PS*154915',
+        # MD5 is unstable
+        'info_dict': {
+            'id': '154915',
+            'ext': 'mp4',
+            'title': 'Slik høres internett ut når du er blind',
+            'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
+            'duration': 20,
+        }
+    }, {
+        'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
+        'only_matching': True,
+    }, {
+        'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70',
+        'only_matching': True,
+    }, {
+        'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.nrk.no/video/dompap-og-andre-fugler-i-piip-show_150533',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.nrk.no/video/humor/kommentatorboksen-reiser-til-sjos_d1fda11f-a4ad-437a-a374-0398bc84e999',
+        'only_matching': True,
+    }, {
+        # podcast
+        'url': 'nrk:l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
+        'only_matching': True,
+    }, {
+        # clip
+        'url': 'nrk:150533',
+        'only_matching': True,
+    }, {
+        # episode
+        'url': 'nrk:MDDP12000117',
+        'only_matching': True,
+    }, {
+        # direkte
+        'url': 'nrk:nrk1',
+        'only_matching': True,
+    }]
+
     def _extract_from_playback(self, video_id):
         path_templ = 'playback/%s/' + video_id
 
@@ -84,8 +153,15 @@ class NRKBaseIE(InfoExtractor):
             format_url = url_or_none(asset.get('url'))
             if not format_url:
                 continue
-            if asset.get('format') == 'HLS' or determine_ext(format_url) == 'm3u8':
+            asset_format = (asset.get('format') or '').lower()
+            if asset_format == 'hls' or determine_ext(format_url) == 'm3u8':
                 formats.extend(self._extract_nrk_formats(format_url, video_id))
+            elif asset_format == 'mp3':
+                formats.append({
+                    'url': format_url,
+                    'format_id': asset_format,
+                    'vcodec': 'none',
+                })
         self._sort_formats(formats)
 
         data = call_playback_api('metadata')
@@ -122,65 +198,12 @@ class NRKBaseIE(InfoExtractor):
             'formats': formats,
         }
 
-
-class NRKIE(NRKBaseIE):
-    _VALID_URL = r'''(?x)
-                        (?:
-                            nrk:|
-                            https?://
-                                (?:
-                                    (?:www\.)?nrk\.no/video/(?:PS\*|[^_]+_)|
-                                    v8[-.]psapi\.nrk\.no/mediaelement/
-                                )
-                            )
-                            (?P<id>[^?\#&]+)
-                        '''
-
-    _TESTS = [{
-        # video
-        'url': 'http://www.nrk.no/video/PS*150533',
-        'md5': 'f46be075326e23ad0e524edfcb06aeb6',
-        'info_dict': {
-            'id': '150533',
-            'ext': 'mp4',
-            'title': 'Dompap og andre fugler i Piip-Show',
-            'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
-            'duration': 262,
-        }
-    }, {
-        # audio
-        'url': 'http://www.nrk.no/video/PS*154915',
-        # MD5 is unstable
-        'info_dict': {
-            'id': '154915',
-            'ext': 'mp4',
-            'title': 'Slik høres internett ut når du er blind',
-            'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
-            'duration': 20,
-        }
-    }, {
-        'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
-        'only_matching': True,
-    }, {
-        'url': 'nrk:clip/7707d5a3-ebe7-434a-87d5-a3ebe7a34a70',
-        'only_matching': True,
-    }, {
-        'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.nrk.no/video/dompap-og-andre-fugler-i-piip-show_150533',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.nrk.no/video/humor/kommentatorboksen-reiser-til-sjos_d1fda11f-a4ad-437a-a374-0398bc84e999',
-        'only_matching': True,
-    }]
-
     def _real_extract(self, url):
         video_id = self._match_id(url)
         return self._extract_from_playback(video_id)
 
 
-class NRKTVIE(NRKBaseIE):
+class NRKTVIE(InfoExtractor):
     IE_DESC = 'NRK TV and NRK Radio'
     _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
     _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:[^/]+/)*%s' % _EPISODE_RE
@@ -288,7 +311,8 @@ class NRKTVIE(NRKBaseIE):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        return self._extract_from_playback(video_id)
+        return self.url_result(
+            'nrk:%s' % video_id, ie=NRKIE.ie_key(), video_id=video_id)
 
 
 class NRKTVEpisodeIE(InfoExtractor):
@@ -359,8 +383,6 @@ class NRKTVSerieBaseIE(NRKBaseIE):
             nrk_id = episode.get('prfId') or episode.get('episodeId')
             if not nrk_id or not isinstance(nrk_id, compat_str):
                 continue
-            if not re.match(NRKTVIE._EPISODE_RE, nrk_id):
-                continue
             entries.append(self.url_result(
                 'nrk:%s' % nrk_id, ie=NRKIE.ie_key(), video_id=nrk_id))
         return entries
@@ -372,6 +394,10 @@ class NRKTVSerieBaseIE(NRKBaseIE):
             if embedded.get(asset_key):
                 return asset_key
 
+    @staticmethod
+    def _catalog_name(serie_kind):
+        return 'podcast' if serie_kind in ('podcast', 'podkast') else 'series'
+
     def _entries(self, data, display_id):
         for page_num in itertools.count(1):
             embedded = data.get('_embedded') or data
@@ -405,7 +431,16 @@ class NRKTVSerieBaseIE(NRKBaseIE):
 
 
 class NRKTVSeasonIE(NRKTVSerieBaseIE):
-    _VALID_URL = r'https?://(?P<domain>tv|radio)\.nrk\.no/serie/(?P<serie>[^/]+)/(?:sesong/)?(?P<id>\d+)'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?P<domain>tv|radio)\.nrk\.no/
+                        (?P<serie_kind>serie|pod[ck]ast)/
+                        (?P<serie>[^/]+)/
+                        (?:
+                            (?:sesong/)?(?P<id>\d+)|
+                            sesong/(?P<id_2>[^/?#&]+)
+                        )
+                    '''
     _TESTS = [{
         'url': 'https://tv.nrk.no/serie/backstage/sesong/1',
         'info_dict': {
@@ -441,19 +476,34 @@ class NRKTVSeasonIE(NRKTVSerieBaseIE):
         # 180 entries, single page
         'url': 'https://tv.nrk.no/serie/spangas/sesong/1',
         'only_matching': True,
+    }, {
+        'url': 'https://radio.nrk.no/podkast/hele_historien/sesong/diagnose-kverulant',
+        'info_dict': {
+            'id': 'hele_historien/diagnose-kverulant',
+            'title': 'Diagnose kverulant',
+        },
+        'playlist_mincount': 3,
+    }, {
+        'url': 'https://radio.nrk.no/podkast/loerdagsraadet/sesong/202101',
+        'only_matching': True,
     }]
 
     @classmethod
     def suitable(cls, url):
-        return (False if NRKTVIE.suitable(url) or NRKTVEpisodeIE.suitable(url)
+        return (False if NRKTVIE.suitable(url) or NRKTVEpisodeIE.suitable(url) or NRKRadioPodkastIE.suitable(url)
                 else super(NRKTVSeasonIE, cls).suitable(url))
 
     def _real_extract(self, url):
-        domain, serie, season_id = re.match(self._VALID_URL, url).groups()
+        mobj = re.match(self._VALID_URL, url)
+        domain = mobj.group('domain')
+        serie_kind = mobj.group('serie_kind')
+        serie = mobj.group('serie')
+        season_id = mobj.group('id') or mobj.group('id_2')
         display_id = '%s/%s' % (serie, season_id)
 
         data = self._call_api(
-            '%s/catalog/series/%s/seasons/%s' % (domain, serie, season_id),
+            '%s/catalog/%s/%s/seasons/%s'
+            % (domain, self._catalog_name(serie_kind), serie, season_id),
             display_id, 'season', query={'pageSize': 50})
 
         title = try_get(data, lambda x: x['titles']['title'], compat_str) or display_id
@@ -463,7 +513,7 @@ class NRKTVSeasonIE(NRKTVSerieBaseIE):
 
 
 class NRKTVSeriesIE(NRKTVSerieBaseIE):
-    _VALID_URL = r'https?://(?P<domain>(?:tv|radio)\.nrk|(?:tv\.)?nrksuper)\.no/serie/(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?P<domain>(?:tv|radio)\.nrk|(?:tv\.)?nrksuper)\.no/(?P<serie_kind>serie|pod[ck]ast)/(?P<id>[^/]+)'
     _TESTS = [{
         # new layout, instalments
         'url': 'https://tv.nrk.no/serie/groenn-glede',
@@ -523,23 +573,33 @@ class NRKTVSeriesIE(NRKTVSerieBaseIE):
     }, {
         'url': 'https://nrksuper.no/serie/labyrint',
         'only_matching': True,
+    }, {
+        'url': 'https://radio.nrk.no/podkast/ulrikkes_univers',
+        'info_dict': {
+            'id': 'ulrikkes_univers',
+        },
+        'playlist_mincount': 10,
+    }, {
+        'url': 'https://radio.nrk.no/podkast/ulrikkes_univers/nrkno-poddkast-26588-134079-05042018030000',
+        'only_matching': True,
     }]
 
     @classmethod
     def suitable(cls, url):
         return (
             False if any(ie.suitable(url)
-                         for ie in (NRKTVIE, NRKTVEpisodeIE, NRKTVSeasonIE))
+                         for ie in (NRKTVIE, NRKTVEpisodeIE, NRKRadioPodkastIE, NRKTVSeasonIE))
             else super(NRKTVSeriesIE, cls).suitable(url))
 
     def _real_extract(self, url):
-        site, series_id = re.match(self._VALID_URL, url).groups()
+        site, serie_kind, series_id = re.match(self._VALID_URL, url).groups()
         is_radio = site == 'radio.nrk'
         domain = 'radio' if is_radio else 'tv'
 
         size_prefix = 'p' if is_radio else 'embeddedInstalmentsP'
         series = self._call_api(
-            '%s/catalog/series/%s' % (domain, series_id),
+            '%s/catalog/%s/%s'
+            % (domain, self._catalog_name(serie_kind), series_id),
             series_id, 'serie', query={size_prefix + 'ageSize': 50})
         titles = try_get(series, [
             lambda x: x['titles'],
@@ -554,12 +614,14 @@ class NRKTVSeriesIE(NRKTVSerieBaseIE):
         embedded_seasons = embedded.get('seasons') or []
         if len(linked_seasons) > len(embedded_seasons):
             for season in linked_seasons:
-                season_name = season.get('name')
-                if season_name and isinstance(season_name, compat_str):
+                season_url = urljoin(url, season.get('href'))
+                if not season_url:
+                    season_name = season.get('name')
+                    if season_name and isinstance(season_name, compat_str):
+                        season_url = 'https://%s.nrk.no/serie/%s/sesong/%s' % (domain, series_id, season_name)
+                if season_url:
                     entries.append(self.url_result(
-                        'https://%s.nrk.no/serie/%s/sesong/%s'
-                        % (domain, series_id, season_name),
-                        ie=NRKTVSeasonIE.ie_key(),
+                        season_url, ie=NRKTVSeasonIE.ie_key(),
                         video_title=season.get('title')))
         else:
             for season in embedded_seasons:
@@ -584,6 +646,38 @@ class NRKTVDirekteIE(NRKTVIE):
     }]
 
 
+class NRKRadioPodkastIE(InfoExtractor):
+    _VALID_URL = r'https?://radio\.nrk\.no/pod[ck]ast/(?:[^/]+/)+(?P<id>l_[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
+
+    _TESTS = [{
+        'url': 'https://radio.nrk.no/podkast/ulrikkes_univers/l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
+        'md5': '8d40dab61cea8ab0114e090b029a0565',
+        'info_dict': {
+            'id': 'MUHH48000314AA',
+            'ext': 'mp4',
+            'title': '20 spørsmål 23.05.2014',
+            'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
+            'duration': 1741,
+            'series': '20 spørsmål',
+            'episode': '23.05.2014',
+        },
+    }, {
+        'url': 'https://radio.nrk.no/podcast/ulrikkes_univers/l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
+        'only_matching': True,
+    }, {
+        'url': 'https://radio.nrk.no/podkast/ulrikkes_univers/sesong/1/l_96f4f1b0-de54-4e6a-b4f1-b0de54fe6af8',
+        'only_matching': True,
+    }, {
+        'url': 'https://radio.nrk.no/podkast/hele_historien/sesong/bortfoert-i-bergen/l_774d1a2c-7aa7-4965-8d1a-2c7aa7d9652c',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        return self.url_result(
+            'nrk:%s' % video_id, ie=NRKIE.ie_key(), video_id=video_id)
+
+
 class NRKPlaylistBaseIE(InfoExtractor):
     def _extract_description(self, webpage):
         pass