[servus] Fix extraction (closes #26872, closes #26967, closes #26983, closes #27000)

youtube_dl/extractor/servus.py

@@ -1,9 +1,15 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    float_or_none,
+    int_or_none,
+    unified_timestamp,
+    urlencode_postdata,
+    url_or_none,
+)
 
 
 class ServusIE(InfoExtractor):
@@ -19,13 +25,22 @@ class ServusIE(InfoExtractor):
     _TESTS = [{
         # new URL schema
         'url': 'https://www.servustv.com/videos/aa-1t6vbu5pw1w12/',
-        'md5': '3e1dd16775aa8d5cbef23628cfffc1f4',
+        'md5': '60474d4c21f3eb148838f215c37f02b9',
         'info_dict': {
             'id': 'AA-1T6VBU5PW1W12',
             'ext': 'mp4',
             'title': 'Die Grünen aus Sicht des Volkes',
+            'alt_title': 'Talk im Hangar-7 Voxpops Gruene',
             'description': 'md5:1247204d85783afe3682644398ff2ec4',
             'thumbnail': r're:^https?://.*\.jpg',
+            'duration': 62.442,
+            'timestamp': 1605193976,
+            'upload_date': '20201112',
+            'series': 'Talk im Hangar-7',
+            'season': 'Season 9',
+            'season_number': 9,
+            'episode': 'Episode 31 - September 14',
+            'episode_number': 31,
         }
     }, {
         # old URL schema
@@ -44,26 +59,87 @@ class ServusIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url).upper()
-        webpage = self._download_webpage(url, video_id)
 
-        title = self._search_regex(
+        token = self._download_json(
-            (r'videoLabel\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
+            'https://auth.redbullmediahouse.com/token', video_id,
-             r'<h\d+[^>]+\bclass=["\']heading--(?:one|two)["\'][^>]*>(?P<title>[^<]+)'),
+            'Downloading token', data=urlencode_postdata({
-            webpage, 'title', default=None,
+                'grant_type': 'client_credentials',
-            group='title') or self._og_search_title(webpage)
+            }), headers={
-        title = re.sub(r'\s*-\s*Servus TV\s*$', '', title)
+                'Authorization': 'Basic SVgtMjJYNEhBNFdEM1cxMTpEdDRVSkFLd2ZOMG5IMjB1NGFBWTBmUFpDNlpoQ1EzNA==',
-        description = self._og_search_description(webpage)
+            })
-        thumbnail = self._og_search_thumbnail(webpage)
+        access_token = token['access_token']
+        token_type = token.get('token_type', 'Bearer')
 
-        formats = self._extract_m3u8_formats(
+        video = self._download_json(
-            'https://stv.rbmbtnx.net/api/v1/manifests/%s.m3u8' % video_id,
+            'https://sparkle-api.liiift.io/api/v1/stv/channels/international/assets/%s' % video_id,
-            video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
+            video_id, 'Downloading video JSON', headers={
+                'Authorization': '%s %s' % (token_type, access_token),
+            })
+
+        formats = []
+        thumbnail = None
+        for resource in video['resources']:
+            if not isinstance(resource, dict):
+                continue
+            format_url = url_or_none(resource.get('url'))
+            if not format_url:
+                continue
+            extension = resource.get('extension')
+            type_ = resource.get('type')
+            if extension == 'jpg' or type_ == 'reference_keyframe':
+                thumbnail = format_url
+                continue
+            ext = determine_ext(format_url)
+            if type_ == 'dash' or ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    format_url, video_id, mpd_id='dash', fatal=False))
+            elif type_ == 'hls' or ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    format_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif extension == 'mp4' or ext == 'mp4':
+                formats.append({
+                    'url': format_url,
+                    'format_id': type_,
+                    'width': int_or_none(resource.get('width')),
+                    'height': int_or_none(resource.get('height')),
+                })
         self._sort_formats(formats)
 
+        attrs = {}
+        for attribute in video['attributes']:
+            if not isinstance(attribute, dict):
+                continue
+            key = attribute.get('fieldKey')
+            value = attribute.get('fieldValue')
+            if not key or not value:
+                continue
+            attrs[key] = value
+
+        title = attrs.get('title_stv') or video_id
+        alt_title = attrs.get('title')
+        description = attrs.get('long_description') or attrs.get('short_description')
+        series = attrs.get('label')
+        season = attrs.get('season')
+        episode = attrs.get('chapter')
+        duration = float_or_none(attrs.get('duration'), scale=1000)
+        season_number = int_or_none(self._search_regex(
+            r'Season (\d+)', season or '', 'season number', default=None))
+        episode_number = int_or_none(self._search_regex(
+            r'Episode (\d+)', episode or '', 'episode number', default=None))
+
         return {
             'id': video_id,
             'title': title,
+            'alt_title': alt_title,
             'description': description,
             'thumbnail': thumbnail,
+            'duration': duration,
+            'timestamp': unified_timestamp(video.get('lastPublished')),
+            'series': series,
+            'season': season,
+            'season_number': season_number,
+            'episode': episode,
+            'episode_number': episode_number,
             'formats': formats,
         }