[servus] Fix extraction

youtube_dl/extractor/servus.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import ExtractorError, JSON_LD_RE
 
 
 class ServusIE(InfoExtractor):
@@ -19,13 +20,14 @@ class ServusIE(InfoExtractor):
     _TESTS = [{
         # new URL schema
         'url': 'https://www.servustv.com/videos/aa-1t6vbu5pw1w12/',
-        'md5': '3e1dd16775aa8d5cbef23628cfffc1f4',
+        'md5': '9f825d6ec14b3d8bebc5b23d094e1e51',
         'info_dict': {
             'id': 'AA-1T6VBU5PW1W12',
             'ext': 'mp4',
             'title': 'Die Grünen aus Sicht des Volkes',
             'description': 'md5:1247204d85783afe3682644398ff2ec4',
-            'thumbnail': r're:^https?://.*\.jpg',
+            'upload_date': '20170911',
+            'timestamp': 1505147648,
         }
     }, {
         # old URL schema
@@ -46,24 +48,20 @@ class ServusIE(InfoExtractor):
         video_id = self._match_id(url).upper()
         webpage = self._download_webpage(url, video_id)
 
-        title = self._search_regex(
-            (r'videoLabel\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
-             r'<h\d+[^>]+\bclass=["\']heading--(?:one|two)["\'][^>]*>(?P<title>[^<]+)'),
-            webpage, 'title', default=None,
-            group='title') or self._og_search_title(webpage)
-        title = re.sub(r'\s*-\s*Servus TV\s*$', '', title)
-        description = self._og_search_description(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
+        if 'rbmh-video-player-trigger' not in webpage:
+            raise ExtractorError('Video not available (maybe not aired yet)', expected=True, video_id=video_id)
 
-        formats = self._extract_m3u8_formats(
-            'https://stv.rbmbtnx.net/api/v1/manifests/%s.m3u8' % video_id,
-            video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
-        self._sort_formats(formats)
+        info = {}
+        for match in re.finditer(JSON_LD_RE, webpage):
+            json_ld = match.group('json_ld')
+            info = self._json_ld(json_ld, video_id)
+            if info:
+                break
+        else:
+            raise ExtractorError('Could not extract video URL', video_id=video_id)
 
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'formats': formats,
-        }
+        info['id'] = video_id
+        info['formats'] = self._extract_m3u8_formats(info['url'], video_id, 'mp4')
+        self._sort_formats(info['formats'])
+
+        return info