Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / redbulltv.py

diff --git a/youtube_dl/extractor/redbulltv.py b/youtube_dl/extractor/redbulltv.py
index afab624261c6db20e277c93f3f00379bc45f01de..7e8d58f38190d857310cd9ecea2074024f38257c 100644 (file)
--- a/youtube_dl/extractor/redbulltv.py
+++ b/youtube_dl/extractor/redbulltv.py
@@ -5,45 +5,38 @@ from .common import InfoExtractor
 from ..compat import compat_HTTPError
 from ..utils import (
     float_or_none,
-    int_or_none,
-    try_get,
-    # unified_timestamp,
     ExtractorError,
 )
 
 
 class RedBullTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?redbull\.tv/(?:video|film)/(?P<id>AP-\w+)'
+    _VALID_URL = r'https?://(?:www\.)?redbull(?:\.tv|\.com/(?:[^/]+/)?tv)/video/(?P<id>AP-\w+)'
     _TESTS = [{
         # film
-        'url': 'https://www.redbull.tv/video/AP-1Q756YYX51W11/abc-of-wrc',
+        'url': 'https://www.redbull.tv/video/AP-1Q6XCDTAN1W11',
         'md5': 'fb0445b98aa4394e504b413d98031d1f',
         'info_dict': {
-            'id': 'AP-1Q756YYX51W11',
+            'id': 'AP-1Q6XCDTAN1W11',
             'ext': 'mp4',
-            'title': 'ABC of...WRC',
+            'title': 'ABC of... WRC - ABC of... S1E6',
             'description': 'md5:5c7ed8f4015c8492ecf64b6ab31e7d31',
             'duration': 1582.04,
-            # 'timestamp': 1488405786,
-            # 'upload_date': '20170301',
         },
     }, {
         # episode
-        'url': 'https://www.redbull.tv/video/AP-1PMT5JCWH1W11/grime?playlist=shows:shows-playall:web',
+        'url': 'https://www.redbull.tv/video/AP-1PMHKJFCW1W11',
         'info_dict': {
-            'id': 'AP-1PMT5JCWH1W11',
+            'id': 'AP-1PMHKJFCW1W11',
             'ext': 'mp4',
-            'title': 'Grime - Hashtags S2 E4',
-            'description': 'md5:334b741c8c1ce65be057eab6773c1cf5',
+            'title': 'Grime - Hashtags S2E4',
+            'description': 'md5:b5f522b89b72e1e23216e5018810bb25',
             'duration': 904.6,
-            # 'timestamp': 1487290093,
-            # 'upload_date': '20170217',
-            'series': 'Hashtags',
-            'season_number': 2,
-            'episode_number': 4,
+        },
+        'params': {
+            'skip_download': True,
         },
     }, {
-        'url': 'https://www.redbull.tv/film/AP-1MSKKF5T92111/in-motion',
+        'url': 'https://www.redbull.com/int-en/tv/video/AP-1UWHCAR9S1W11/rob-meets-sam-gaze?playlist=playlists::3f81040a-2f31-4832-8e2e-545b1d39d173',
         'only_matching': True,
     }]
 
@@ -51,72 +44,57 @@ class RedBullTVIE(InfoExtractor):
         video_id = self._match_id(url)
 
         session = self._download_json(
-            'https://api-v2.redbull.tv/session', video_id,
+            'https://api.redbull.tv/v3/session', video_id,
             note='Downloading access token', query={
-                'build': '4.370.0',
                 'category': 'personal_computer',
-                'os_version': '1.0',
                 'os_family': 'http',
             })
         if session.get('code') == 'error':
             raise ExtractorError('%s said: %s' % (
                 self.IE_NAME, session['message']))
-        auth = '%s %s' % (session.get('token_type', 'Bearer'), session['access_token'])
+        token = session['token']
 
         try:
-            info = self._download_json(
-                'https://api-v2.redbull.tv/content/%s' % video_id,
+            video = self._download_json(
+                'https://api.redbull.tv/v3/products/' + video_id,
                 video_id, note='Downloading video information',
-                headers={'Authorization': auth}
+                headers={'Authorization': token}
             )
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
                 error_message = self._parse_json(
-                    e.cause.read().decode(), video_id)['message']
+                    e.cause.read().decode(), video_id)['error']
                 raise ExtractorError('%s said: %s' % (
                     self.IE_NAME, error_message), expected=True)
             raise
 
-        video = info['video_product']
-
-        title = info['title'].strip()
+        title = video['title'].strip()
 
         formats = self._extract_m3u8_formats(
-            video['url'], video_id, 'mp4', 'm3u8_native')
+            'https://dms.redbull.tv/v3/%s/%s/playlist.m3u8' % (video_id, token),
+            video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
         self._sort_formats(formats)
 
         subtitles = {}
-        for _, captions in (try_get(
-                video, lambda x: x['attachments']['captions'],
-                dict) or {}).items():
-            if not captions or not isinstance(captions, list):
-                continue
-            for caption in captions:
-                caption_url = caption.get('url')
-                if not caption_url:
-                    continue
-                ext = caption.get('format')
-                if ext == 'xml':
-                    ext = 'ttml'
-                subtitles.setdefault(caption.get('lang') or 'en', []).append({
-                    'url': caption_url,
-                    'ext': ext,
-                })
+        for resource in video.get('resources', []):
+            if resource.startswith('closed_caption_'):
+                splitted_resource = resource.split('_')
+                if splitted_resource[2]:
+                    subtitles.setdefault('en', []).append({
+                        'url': 'https://resources.redbull.tv/%s/%s' % (video_id, resource),
+                        'ext': splitted_resource[2],
+                    })
 
-        subheading = info.get('subheading')
+        subheading = video.get('subheading')
         if subheading:
             title += ' - %s' % subheading
 
         return {
             'id': video_id,
             'title': title,
-            'description': info.get('long_description') or info.get(
+            'description': video.get('long_description') or video.get(
                 'short_description'),
             'duration': float_or_none(video.get('duration'), scale=1000),
-            # 'timestamp': unified_timestamp(info.get('published')),
-            'series': info.get('show_title'),
-            'season_number': int_or_none(info.get('season_number')),
-            'episode_number': int_or_none(info.get('episode_number')),
             'formats': formats,
             'subtitles': subtitles,
         }