[ie/ondemandkorea] Fix upgraded format extraction (#8677)

Closes #8675
Authored by: seproDev

yt_dlp/extractor/francetv.py

@@ -1,12 +1,14 @@
 from .common import InfoExtractor
+from .dailymotion import DailymotionIE
 from ..utils import (
-    determine_ext,
     ExtractorError,
+    determine_ext,
     format_field,
+    int_or_none,
+    join_nonempty,
     parse_iso8601,
     parse_qs,
 )
-from .dailymotion import DailymotionIE
 
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
@@ -82,6 +84,8 @@ class FranceTVIE(InfoExtractor):
         videos = []
         title = None
         subtitle = None
+        episode_number = None
+        season_number = None
         image = None
         duration = None
         timestamp = None
@@ -112,7 +116,9 @@ class FranceTVIE(InfoExtractor):
             if meta:
                 if title is None:
                     title = meta.get('title')
-                # XXX: what is meta['pre_title']?
+                # meta['pre_title'] contains season and episode number for series in format "S<ID> E<ID>"
+                season_number, episode_number = self._search_regex(
+                    r'S(\d+)\s*E(\d+)', meta.get('pre_title'), 'episode info', group=(1, 2), default=(None, None))
                 if subtitle is None:
                     subtitle = meta.get('additional_title')
                 if image is None:
@@ -191,19 +197,19 @@ class FranceTVIE(InfoExtractor):
                 } for sheet in spritesheets]
             })
 
-        if subtitle:
-            title += ' - %s' % subtitle
-        title = title.strip()
-
         return {
             'id': video_id,
-            'title': title,
+            'title': join_nonempty(title, subtitle, delim=' - ').strip(),
             'thumbnail': image,
             'duration': duration,
             'timestamp': timestamp,
             'is_live': is_live,
             'formats': formats,
             'subtitles': subtitles,
+            'episode': subtitle if episode_number else None,
+            'series': title if episode_number else None,
+            'episode_number': int_or_none(episode_number),
+            'season_number': int_or_none(season_number),
         }
 
     def _real_extract(self, url):
@@ -230,14 +236,31 @@ class FranceTVSiteIE(FranceTVBaseInfoExtractor):
             'id': 'ec217ecc-0733-48cf-ac06-af1347b849d1',
             'ext': 'mp4',
             'title': '13h15, le dimanche... - Les mystères de Jésus',
-            'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
             'timestamp': 1502623500,
+            'duration': 2580,
+            'thumbnail': r're:^https?://.*\.jpg$',
             'upload_date': '20170813',
         },
         'params': {
             'skip_download': True,
         },
         'add_ie': [FranceTVIE.ie_key()],
+    }, {
+        'url': 'https://www.france.tv/enfants/six-huit-ans/foot2rue/saison-1/3066387-duel-au-vieux-port.html',
+        'info_dict': {
+            'id': 'a9050959-eedd-4b4a-9b0d-de6eeaa73e44',
+            'ext': 'mp4',
+            'title': 'Foot2Rue - Duel au vieux port',
+            'episode': 'Duel au vieux port',
+            'series': 'Foot2Rue',
+            'episode_number': 1,
+            'season_number': 1,
+            'timestamp': 1642761360,
+            'upload_date': '20220121',
+            'season': 'Season 1',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1441,
+        },
     }, {
         # france3
         'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',

yt_dlp/extractor/mediastream.py

@@ -3,8 +3,11 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     clean_html,
+    filter_dict,
+    parse_qs,
     remove_end,
     traverse_obj,
+    update_url_query,
     urljoin,
 )
 
@@ -108,7 +111,9 @@ class MediaStreamIE(MediaStreamBaseIE):
 
         for message in [
             'Debido a tu ubicación no puedes ver el contenido',
-            'You are not allowed to watch this video: Geo Fencing Restriction'
+            'You are not allowed to watch this video: Geo Fencing Restriction',
+            'Este contenido no está disponible en tu zona geográfica.',
+            'El contenido sólo está disponible dentro de',
         ]:
             if message in webpage:
                 self.raise_geo_restricted()
@@ -118,7 +123,16 @@ class MediaStreamIE(MediaStreamBaseIE):
         formats, subtitles = [], {}
         for video_format in player_config['src']:
             if video_format == 'hls':
-                fmts, subs = self._extract_m3u8_formats_and_subtitles(player_config['src'][video_format], video_id)
+                params = {
+                    'at': 'web-app',
+                    'access_token': traverse_obj(parse_qs(url), ('access_token', 0)),
+                }
+                for name, key in (('MDSTRMUID', 'uid'), ('MDSTRMSID', 'sid'), ('MDSTRMPID', 'pid'), ('VERSION', 'av')):
+                    params[key] = self._search_regex(
+                        rf'window\.{name}\s*=\s*["\']([^"\']+)["\'];', webpage, key, default=None)
+
+                fmts, subs = self._extract_m3u8_formats_and_subtitles(
+                    update_url_query(player_config['src'][video_format], filter_dict(params)), video_id)
                 formats.extend(fmts)
                 self._merge_subtitles(subs, target=subtitles)
             elif video_format == 'mpd':

yt_dlp/extractor/ondemandkorea.py

@@ -3,7 +3,6 @@ import re
 import uuid
 
 from .common import InfoExtractor
-from ..networking import HEADRequest
 from ..utils import (
     ExtractorError,
     OnDemandPagedList,
@@ -84,15 +83,17 @@ class OnDemandKoreaIE(InfoExtractor):
         def try_geo_bypass(url):
             return traverse_obj(url, ({parse_qs}, 'stream_url', 0, {url_or_none})) or url
 
-        def try_upgrade_quality(url):
-            mod_url = re.sub(r'_720(p?)\.m3u8', r'_1080\1.m3u8', url)
-            return mod_url if mod_url != url and self._request_webpage(
-                HEADRequest(mod_url), video_id, note='Checking for higher quality format',
-                errnote='No higher quality format found', fatal=False) else url
-
         formats = []
         for m3u8_url in traverse_obj(data, (('sources', 'manifest'), ..., 'url', {url_or_none}, {try_geo_bypass})):
-            formats.extend(self._extract_m3u8_formats(try_upgrade_quality(m3u8_url), video_id, fatal=False))
+            mod_url = re.sub(r'_720(p?)\.m3u8', r'_1080\1.m3u8', m3u8_url)
+            if mod_url != m3u8_url:
+                mod_format = self._extract_m3u8_formats(
+                    mod_url, video_id, note='Checking for higher quality format',
+                    errnote='No higher quality format found', fatal=False)
+                if mod_format:
+                    formats.extend(mod_format)
+                    continue
+            formats.extend(self._extract_m3u8_formats(m3u8_url, video_id, fatal=False))
 
         subtitles = {}
         for track in traverse_obj(data, ('text_tracks', lambda _, v: url_or_none(v['url']))):