[rai] improve subtitles extraction (#27705)

closes #27698

test/test_subtitles.py

@@ -258,16 +258,24 @@ class TestNRKSubtitles(BaseTestSubtitles):
 
 
 class TestRaiPlaySubtitles(BaseTestSubtitles):
-    url = 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html'
     IE = RaiPlayIE
 
-    def test_allsubtitles(self):
+    def test_subtitles_key(self):
+        self.url = 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html'
         self.DL.params['writesubtitles'] = True
         self.DL.params['allsubtitles'] = True
         subtitles = self.getSubtitles()
         self.assertEqual(set(subtitles.keys()), set(['it']))
         self.assertEqual(md5(subtitles['it']), 'b1d90a98755126b61e667567a1f6680a')
 
+    def test_subtitles_array_key(self):
+        self.url = 'https://www.raiplay.it/video/2020/12/Report---04-01-2021-2e90f1de-8eee-4de4-ac0e-78d21db5b600.html'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['it']))
+        self.assertEqual(md5(subtitles['it']), '4b3264186fbb103508abe5311cfcb9cd')
+
 
 class TestVikiSubtitles(BaseTestSubtitles):
     url = 'http://www.viki.com/videos/1060846v-punch-episode-18'

youtube_dl/extractor/rai.py

@@ -103,22 +103,28 @@ class RaiBaseIE(InfoExtractor):
         }.items() if v is not None)
 
     @staticmethod
-    def _extract_subtitles(url, subtitle_url):
+    def _extract_subtitles(url, video_data):
+        STL_EXT = 'stl'
+        SRT_EXT = 'srt'
         subtitles = {}
-        if subtitle_url and isinstance(subtitle_url, compat_str):
+        subtitles_array = video_data.get('subtitlesArray') or []
-            subtitle_url = urljoin(url, subtitle_url)
+        for k in ('subtitles', 'subtitlesUrl'):
-            STL_EXT = '.stl'
+            subtitles_array.append({'url': video_data.get(k)})
-            SRT_EXT = '.srt'
+        for subtitle in subtitles_array:
-            subtitles['it'] = [{
+            sub_url = subtitle.get('url')
-                'ext': 'stl',
+            if sub_url and isinstance(sub_url, compat_str):
-                'url': subtitle_url,
+                sub_lang = subtitle.get('language') or 'it'
-            }]
+                sub_url = urljoin(url, sub_url)
-            if subtitle_url.endswith(STL_EXT):
+                sub_ext = determine_ext(sub_url, SRT_EXT)
-                srt_url = subtitle_url[:-len(STL_EXT)] + SRT_EXT
+                subtitles.setdefault(sub_lang, []).append({
-                subtitles['it'].append({
+                    'ext': sub_ext,
-                    'ext': 'srt',
+                    'url': sub_url,
-                    'url': srt_url,
                 })
+                if STL_EXT == sub_ext:
+                    subtitles[sub_lang].append({
+                        'ext': SRT_EXT,
+                        'url': sub_url[:-len(STL_EXT)] + SRT_EXT,
+                    })
         return subtitles
 
 
@@ -138,6 +144,9 @@ class RaiPlayIE(RaiBaseIE):
             'duration': 6160,
             'series': 'Report',
             'season': '2013/14',
+            'subtitles': {
+                'it': 'count:2',
+            },
         },
         'params': {
             'skip_download': True,
@@ -145,6 +154,10 @@ class RaiPlayIE(RaiBaseIE):
     }, {
         'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
         'only_matching': True,
+    }, {
+        # subtitles at 'subtitlesArray' key (see #27698)
+        'url': 'https://www.raiplay.it/video/2020/12/Report---04-01-2021-2e90f1de-8eee-4de4-ac0e-78d21db5b600.html',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -172,7 +185,7 @@ class RaiPlayIE(RaiBaseIE):
         if date_published and time_published:
             date_published += ' ' + time_published
 
-        subtitles = self._extract_subtitles(url, video.get('subtitles'))
+        subtitles = self._extract_subtitles(url, video)
 
         program_info = media.get('program_info') or {}
         season = media.get('season')
@@ -327,7 +340,7 @@ class RaiIE(RaiBaseIE):
             'skip_download': True,
         },
     }, {
-        # ContentItem in iframe (see #12652)
+        # ContentItem in iframe (see #12652) and subtitle at 'subtitlesUrl' key
         'url': 'http://www.presadiretta.rai.it/dl/portali/site/puntata/ContentItem-3ed19d13-26c2-46ff-a551-b10828262f1b.html',
         'info_dict': {
             'id': '1ad6dc64-444a-42a4-9bea-e5419ad2f5fd',
@@ -335,6 +348,9 @@ class RaiIE(RaiBaseIE):
             'title': 'Partiti acchiappavoti - Presa diretta del 13/09/2015',
             'description': 'md5:d291b03407ec505f95f27970c0b025f4',
             'upload_date': '20150913',
+            'subtitles': {
+                'it': 'count:2',
+            },
         },
         'params': {
             'skip_download': True,
@@ -379,7 +395,7 @@ class RaiIE(RaiBaseIE):
                     'url': compat_urlparse.urljoin(url, thumbnail_url),
                 })
 
-        subtitles = self._extract_subtitles(url, media.get('subtitlesUrl'))
+        subtitles = self._extract_subtitles(url, media)
 
         info = {
             'id': content_id,