[telegraaf] fix extraction
authorRemita Amine <remitamine@gmail.com>
Sun, 3 Nov 2019 21:04:03 +0000 (22:04 +0100)
committerRemita Amine <remitamine@gmail.com>
Sun, 3 Nov 2019 21:04:03 +0000 (22:04 +0100)
youtube_dl/extractor/telegraaf.py

index 0f576c1aba1a01491f657d8d892291cfe7934f21..2dc0205373a87381663fd655cf5512e8cfcdeaed 100644 (file)
@@ -4,21 +4,25 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
-    remove_end,
+    int_or_none,
+    parse_iso8601,
+    try_get,
 )
 
 
 class TelegraafIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/tv/(?:[^/]+/)+(?P<id>\d+)/[^/]+\.html'
+    _VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/video/(?P<id>\d+)'
     _TEST = {
-        'url': 'http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html',
+        'url': 'https://www.telegraaf.nl/video/734366489/historisch-scheepswrak-slaat-na-100-jaar-los',
         'info_dict': {
-            'id': '24353229',
+            'id': 'gaMItuoSeUg2',
             'ext': 'mp4',
-            'title': 'Tikibad ontruimd wegens brand',
-            'description': 'md5:05ca046ff47b931f9b04855015e163a4',
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'duration': 33,
+            'title': 'Historisch scheepswrak slaat na 100 jaar los',
+            'description': 'md5:6f53b7c4f55596722ac24d6c0ec00cfb',
+            'thumbnail': r're:^https?://.*\.jpg',
+            'duration': 55,
+            'timestamp': 1572805527,
+            'upload_date': '20191103',
         },
         'params': {
             # m3u8 download
@@ -27,23 +31,30 @@ class TelegraafIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        article_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        video_id = self._download_json(
+            'https://www.telegraaf.nl/graphql', article_id, query={
+                'query': '''{
+  article(uid: %s) {
+    videos {
+      videoId
+    }
+  }
+}''' % article_id,
+            })['data']['article']['videos'][0]['videoId']
 
-        player_url = self._html_search_regex(
-            r'<iframe[^>]+src="([^"]+")', webpage, 'player URL')
-        player_page = self._download_webpage(
-            player_url, video_id, note='Download player webpage')
-        playlist_url = self._search_regex(
-            r'playlist\s*:\s*"([^"]+)"', player_page, 'playlist URL')
-        playlist_data = self._download_json(playlist_url, video_id)
+        item = self._download_json(
+            'https://content.tmgvideo.nl/playlist/item=%s/playlist.json' % video_id,
+            video_id)['items'][0]
+        title = item['title']
 
-        item = playlist_data['items'][0]
         formats = []
-        locations = item['locations']
+        locations = item.get('locations') or {}
         for location in locations.get('adaptive', []):
-            manifest_url = location['src']
+            manifest_url = location.get('src')
+            if not manifest_url:
+                continue
             ext = determine_ext(manifest_url)
             if ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
@@ -54,25 +65,25 @@ class TelegraafIE(InfoExtractor):
             else:
                 self.report_warning('Unknown adaptive format %s' % ext)
         for location in locations.get('progressive', []):
+            src = try_get(location, lambda x: x['sources'][0]['src'])
+            if not src:
+                continue
+            label = location.get('label')
             formats.append({
-                'url': location['sources'][0]['src'],
-                'width': location.get('width'),
-                'height': location.get('height'),
-                'format_id': 'http-%s' % location['label'],
+                'url': src,
+                'width': int_or_none(location.get('width')),
+                'height': int_or_none(location.get('height')),
+                'format_id': 'http' + ('-%s' % label if label else ''),
             })
 
         self._sort_formats(formats)
 
-        title = remove_end(self._og_search_title(webpage), ' - VIDEO')
-        description = self._og_search_description(webpage)
-        duration = item.get('duration')
-        thumbnail = item.get('poster')
-
         return {
             'id': video_id,
             'title': title,
-            'description': description,
+            'description': item.get('description'),
             'formats': formats,
-            'duration': duration,
-            'thumbnail': thumbnail,
+            'duration': int_or_none(item.get('duration')),
+            'thumbnail': item.get('poster'),
+            'timestamp': parse_iso8601(item.get('datecreated'), ' '),
         }