[stv] improve episode id extraction(closes #23083)

youtube_dl/extractor/stv.py

@@ -8,13 +8,17 @@ from ..utils import (
     compat_str,
     float_or_none,
     int_or_none,
+    smuggle_url,
+    str_or_none,
+    try_get,
 )
 
 
 class STVPlayerIE(InfoExtractor):
     IE_NAME = 'stv:player'
     _VALID_URL = r'https?://player\.stv\.tv/(?P<type>episode|video)/(?P<id>[a-z0-9]{4})'
-    _TEST = {
+    _TESTS = [{
+        # shortform
         'url': 'https://player.stv.tv/video/4gwd/emmerdale/60-seconds-on-set-with-laura-norton/',
         'md5': '5adf9439c31d554f8be0707c7abe7e0a',
         'info_dict': {
@@ -27,7 +31,11 @@ class STVPlayerIE(InfoExtractor):
             'uploader_id': '1486976045',
         },
         'skip': 'this resource is unavailable outside of the UK',
-    }
+    }, {
+        # episodes
+        'url': 'https://player.stv.tv/episode/4125/jennifer-saunders-memory-lane',
+        'only_matching': True,
+    }]
     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1486976045/default_default/index.html?videoId=%s'
     _PTYPE_MAP = {
         'episode': 'episodes',
@@ -36,11 +44,31 @@ class STVPlayerIE(InfoExtractor):
 
     def _real_extract(self, url):
         ptype, video_id = re.match(self._VALID_URL, url).groups()
-        resp = self._download_json(
-            'https://player.api.stv.tv/v1/%s/%s' % (self._PTYPE_MAP[ptype], video_id),
-            video_id)
 
-        result = resp['results']
+        webpage = self._download_webpage(url, video_id, fatal=False) or ''
+        props = (self._parse_json(self._search_regex(
+            r'<script[^>]+id="__NEXT_DATA__"[^>]*>({.+?})</script>',
+            webpage, 'next data', default='{}'), video_id,
+            fatal=False) or {}).get('props') or {}
+        player_api_cache = try_get(
+            props, lambda x: x['initialReduxState']['playerApiCache']) or {}
+
+        api_path, resp = None, {}
+        for k, v in player_api_cache.items():
+            if k.startswith('/episodes/') or k.startswith('/shortform/'):
+                api_path, resp = k, v
+                break
+        else:
+            episode_id = str_or_none(try_get(
+                props, lambda x: x['pageProps']['episodeId']))
+            api_path = '/%s/%s' % (self._PTYPE_MAP[ptype], episode_id or video_id)
+
+        result = resp.get('results')
+        if not result:
+            resp = self._download_json(
+                'https://player.api.stv.tv/v1' + api_path, video_id)
+            result = resp['results']
+
         video = result['video']
         video_id = compat_str(video['id'])
 
@@ -57,7 +85,7 @@ class STVPlayerIE(InfoExtractor):
         return {
             '_type': 'url_transparent',
             'id': video_id,
-            'url': self.BRIGHTCOVE_URL_TEMPLATE % video_id,
+            'url': smuggle_url(self.BRIGHTCOVE_URL_TEMPLATE % video_id, {'geo_countries': ['GB']}),
             'description': result.get('summary'),
             'duration': float_or_none(video.get('length'), 1000),
             'subtitles': subtitles,