[slideslive] Add support for yoda service videos and extract subtitles (closes #27323)
authorSergey M․ <dstftw@gmail.com>
Sun, 6 Dec 2020 20:27:21 +0000 (03:27 +0700)
committerSergey M․ <dstftw@gmail.com>
Sun, 6 Dec 2020 20:27:21 +0000 (03:27 +0700)
youtube_dl/extractor/slideslive.py

index d9ea76831e7ebdc37bad0e722cd92268bfd8eb6f..cd70841a90cad964fb2c654cb01baefdae6bc21c 100644 (file)
@@ -2,7 +2,12 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import smuggle_url
+from ..utils import (
+    bool_or_none,
+    smuggle_url,
+    try_get,
+    url_or_none,
+)
 
 
 class SlidesLiveIE(InfoExtractor):
@@ -18,8 +23,21 @@ class SlidesLiveIE(InfoExtractor):
             'description': 'Watch full version of this video at https://slideslive.com/38902413.',
             'uploader': 'SlidesLive Videos - A',
             'uploader_id': 'UC62SdArr41t_-_fX40QCLRw',
+            'timestamp': 1597615266,
             'upload_date': '20170925',
         }
+    }, {
+        # video_service_name = yoda
+        'url': 'https://slideslive.com/38935785',
+        'md5': '575cd7a6c0acc6e28422fe76dd4bcb1a',
+        'info_dict': {
+            'id': 'RMraDYN5ozA_',
+            'ext': 'mp4',
+            'title': 'Offline Reinforcement Learning: From Algorithms to Practical Challenges',
+        },
+        'params': {
+            'format': 'bestvideo',
+        },
     }, {
         # video_service_name = youtube
         'url': 'https://slideslive.com/38903721/magic-a-scientific-resurrection-of-an-esoteric-legend',
@@ -39,18 +57,47 @@ class SlidesLiveIE(InfoExtractor):
         video_data = self._download_json(
             'https://ben.slideslive.com/player/' + video_id, video_id)
         service_name = video_data['video_service_name'].lower()
-        assert service_name in ('url', 'vimeo', 'youtube')
+        assert service_name in ('url', 'yoda', 'vimeo', 'youtube')
         service_id = video_data['video_service_id']
+        subtitles = {}
+        for sub in try_get(video_data, lambda x: x['subtitles'], list) or []:
+            if not isinstance(sub, dict):
+                continue
+            webvtt_url = url_or_none(sub.get('webvtt_url'))
+            if not webvtt_url:
+                continue
+            lang = sub.get('language') or 'en'
+            subtitles.setdefault(lang, []).append({
+                'url': webvtt_url,
+            })
         info = {
             'id': video_id,
             'thumbnail': video_data.get('thumbnail'),
-            'url': service_id,
+            'is_live': bool_or_none(video_data.get('is_live')),
+            'subtitles': subtitles,
         }
-        if service_name == 'url':
+        if service_name in ('url', 'yoda'):
             info['title'] = video_data['title']
+            if service_name == 'url':
+                info['url'] = service_id
+            else:
+                formats = []
+                _MANIFEST_PATTERN = 'https://01.cdn.yoda.slideslive.com/%s/master.%s'
+                formats.extend(self._extract_m3u8_formats(
+                    _MANIFEST_PATTERN % (service_id, 'm3u8'), service_id, 'mp4',
+                    entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
+                formats.extend(self._extract_mpd_formats(
+                    _MANIFEST_PATTERN % (service_id, 'mpd'), service_id,
+                    mpd_id='dash', fatal=False))
+                self._sort_formats(formats)
+                info.update({
+                    'id': service_id,
+                    'formats': formats,
+                })
         else:
             info.update({
                 '_type': 'url_transparent',
+                'url': service_id,
                 'ie_key': service_name.capitalize(),
                 'title': video_data.get('title'),
             })