[redbull] improve extraction
authorRemita Amine <remitamine@gmail.com>
Wed, 15 Mar 2017 00:40:54 +0000 (01:40 +0100)
committerRemita Amine <remitamine@gmail.com>
Wed, 15 Mar 2017 00:40:54 +0000 (01:40 +0100)
- extract 1080p quality
- correct ttml subtitle ext
- catch api errors
- reduce request size

youtube_dl/extractor/redbulltv.py

index 5c73d5bcafa7cce19414fe2b9046dbca590cce95..afab624261c6db20e277c93f3f00379bc45f01de 100644 (file)
@@ -2,11 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_HTTPError
 from ..utils import (
     float_or_none,
     int_or_none,
     try_get,
-    unified_timestamp,
+    # unified_timestamp,
+    ExtractorError,
 )
 
 
@@ -15,15 +17,15 @@ class RedBullTVIE(InfoExtractor):
     _TESTS = [{
         # film
         'url': 'https://www.redbull.tv/video/AP-1Q756YYX51W11/abc-of-wrc',
-        'md5': '78e860f631d7a846e712fab8c5fe2c38',
+        'md5': 'fb0445b98aa4394e504b413d98031d1f',
         'info_dict': {
             'id': 'AP-1Q756YYX51W11',
             'ext': 'mp4',
             'title': 'ABC of...WRC',
             'description': 'md5:5c7ed8f4015c8492ecf64b6ab31e7d31',
             'duration': 1582.04,
-            'timestamp': 1488405786,
-            'upload_date': '20170301',
+            'timestamp': 1488405786,
+            'upload_date': '20170301',
         },
     }, {
         # episode
@@ -34,8 +36,8 @@ class RedBullTVIE(InfoExtractor):
             'title': 'Grime - Hashtags S2 E4',
             'description': 'md5:334b741c8c1ce65be057eab6773c1cf5',
             'duration': 904.6,
-            'timestamp': 1487290093,
-            'upload_date': '20170217',
+            'timestamp': 1487290093,
+            'upload_date': '20170217',
             'series': 'Hashtags',
             'season_number': 2,
             'episode_number': 4,
@@ -48,29 +50,40 @@ class RedBullTVIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        access_token = self._download_json(
-            'https://api-v2.redbull.tv/start', video_id,
+        session = self._download_json(
+            'https://api-v2.redbull.tv/session', video_id,
             note='Downloading access token', query={
-                'build': '4.0.9',
-                'category': 'smartphone',
-                'os_version': 23,
-                'os_family': 'android',
-            })['auth']['access_token']
+                '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'])
 
-        info = self._download_json(
-            'https://api-v2.redbull.tv/views/%s' % video_id,
-            video_id, note='Downloading video information',
-            headers={'Authorization': 'Bearer ' + access_token}
-        )['blocks'][0]['top'][0]
+        try:
+            info = self._download_json(
+                'https://api-v2.redbull.tv/content/%s' % video_id,
+                video_id, note='Downloading video information',
+                headers={'Authorization': auth}
+            )
+        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']
+                raise ExtractorError('%s said: %s' % (
+                    self.IE_NAME, error_message), expected=True)
+            raise
 
         video = info['video_product']
 
         title = info['title'].strip()
-        m3u8_url = video['url']
 
         formats = self._extract_m3u8_formats(
-            m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
-            m3u8_id='hls')
+            video['url'], video_id, 'mp4', 'm3u8_native')
+        self._sort_formats(formats)
 
         subtitles = {}
         for _, captions in (try_get(
@@ -82,9 +95,12 @@ class RedBullTVIE(InfoExtractor):
                 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': caption.get('format'),
+                    'ext': ext,
                 })
 
         subheading = info.get('subheading')
@@ -97,7 +113,7 @@ class RedBullTVIE(InfoExtractor):
             'description': info.get('long_description') or info.get(
                 'short_description'),
             'duration': float_or_none(video.get('duration'), scale=1000),
-            'timestamp': unified_timestamp(info.get('published')),
+            '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')),