[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / vrv.py

diff --git a/youtube_dl/extractor/vrv.py b/youtube_dl/extractor/vrv.py
index c11da97deb8e890763b27c21236f0a4218134f22..6e51469b094363879749bf2926714662e26a1858 100644 (file)
--- a/youtube_dl/extractor/vrv.py
+++ b/youtube_dl/extractor/vrv.py
@@ -64,7 +64,15 @@ class VRVBaseIE(InfoExtractor):
 
     def _call_cms(self, path, video_id, note):
         if not self._CMS_SIGNING:
-            self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
+            index = self._call_api('index', video_id, 'CMS Signing')
+            self._CMS_SIGNING = index.get('cms_signing') or {}
+            if not self._CMS_SIGNING:
+                for signing_policy in index.get('signing_policies', []):
+                    signing_path = signing_policy.get('path')
+                    if signing_path and signing_path.startswith('/cms/'):
+                        name, value = signing_policy.get('name'), signing_policy.get('value')
+                        if name and value:
+                            self._CMS_SIGNING[name] = value
         return self._download_json(
             self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
             note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
@@ -102,6 +110,15 @@ class VRVIE(VRVBaseIE):
             # m3u8 download
             'skip_download': True,
         },
+    }, {
+        # movie listing
+        'url': 'https://vrv.co/watch/G6NQXZ1J6/Lily-CAT',
+        'info_dict': {
+            'id': 'G6NQXZ1J6',
+            'title': 'Lily C.A.T',
+            'description': 'md5:988b031e7809a6aeb60968be4af7db07',
+        },
+        'playlist_count': 2,
     }]
     _NETRC_MACHINE = 'vrv'
 
@@ -121,25 +138,25 @@ class VRVIE(VRVBaseIE):
         self._TOKEN_SECRET = token_credentials['oauth_token_secret']
 
     def _extract_vrv_formats(self, url, video_id, stream_format, audio_lang, hardsub_lang):
-        if not url or stream_format not in ('hls', 'dash'):
+        if not url or stream_format not in ('hls', 'dash', 'adaptive_hls'):
             return []
-        assert audio_lang or hardsub_lang
         stream_id_list = []
         if audio_lang:
             stream_id_list.append('audio-%s' % audio_lang)
         if hardsub_lang:
             stream_id_list.append('hardsub-%s' % hardsub_lang)
-        stream_id = '-'.join(stream_id_list)
-        format_id = '%s-%s' % (stream_format, stream_id)
-        if stream_format == 'hls':
+        format_id = stream_format
+        if stream_id_list:
+            format_id += '-' + '-'.join(stream_id_list)
+        if 'hls' in stream_format:
             adaptive_formats = self._extract_m3u8_formats(
                 url, video_id, 'mp4', m3u8_id=format_id,
-                note='Downloading %s m3u8 information' % stream_id,
+                note='Downloading %s information' % format_id,
                 fatal=False)
         elif stream_format == 'dash':
             adaptive_formats = self._extract_mpd_formats(
                 url, video_id, mpd_id=format_id,
-                note='Downloading %s MPD information' % stream_id,
+                note='Downloading %s information' % format_id,
                 fatal=False)
         if audio_lang:
             for f in adaptive_formats:
@@ -155,6 +172,23 @@ class VRVIE(VRVBaseIE):
         resource_path = object_data['__links__']['resource']['href']
         video_data = self._call_cms(resource_path, video_id, 'video')
         title = video_data['title']
+        description = video_data.get('description')
+
+        if video_data.get('__class__') == 'movie_listing':
+            items = self._call_cms(
+                video_data['__links__']['movie_listing/movies']['href'],
+                video_id, 'movie listing').get('items') or []
+            if len(items) != 1:
+                entries = []
+                for item in items:
+                    item_id = item.get('id')
+                    if not item_id:
+                        continue
+                    entries.append(self.url_result(
+                        'https://vrv.co/watch/' + item_id,
+                        self.ie_key(), item_id, item.get('title')))
+                return self.playlist_result(entries, video_id, title, description)
+            video_data = items[0]
 
         streams_path = video_data['__links__'].get('streams', {}).get('href')
         if not streams_path:
@@ -172,14 +206,15 @@ class VRVIE(VRVBaseIE):
         self._sort_formats(formats)
 
         subtitles = {}
-        for subtitle in streams_json.get('subtitles', {}).values():
-            subtitle_url = subtitle.get('url')
-            if not subtitle_url:
-                continue
-            subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
-                'url': subtitle_url,
-                'ext': subtitle.get('format', 'ass'),
-            })
+        for k in ('captions', 'subtitles'):
+            for subtitle in streams_json.get(k, {}).values():
+                subtitle_url = subtitle.get('url')
+                if not subtitle_url:
+                    continue
+                subtitles.setdefault(subtitle.get('locale', 'en-US'), []).append({
+                    'url': subtitle_url,
+                    'ext': subtitle.get('format', 'ass'),
+                })
 
         thumbnails = []
         for thumbnail in video_data.get('images', {}).get('thumbnails', []):
@@ -198,7 +233,7 @@ class VRVIE(VRVBaseIE):
             'formats': formats,
             'subtitles': subtitles,
             'thumbnails': thumbnails,
-            'description': video_data.get('description'),
+            'description': description,
             'duration': float_or_none(video_data.get('duration_ms'), 1000),
             'uploader_id': video_data.get('channel_id'),
             'series': video_data.get('series_title'),