Annotate some highlights of the release.

[youtubedl] / youtube_dl / extractor / telegraaf.py

diff --git a/youtube_dl/extractor/telegraaf.py b/youtube_dl/extractor/telegraaf.py
index 6f8333cfc0d40aee4d3637ed1e58867da1277e9f..2dc0205373a87381663fd655cf5512e8cfcdeaed 100644 (file)
--- a/youtube_dl/extractor/telegraaf.py
+++ b/youtube_dl/extractor/telegraaf.py
@@ -2,34 +2,88 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from __future__ import unicode_literals
 
 from .common import InfoExtractor

-from ..utils import remove_end
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    parse_iso8601,
+    try_get,
+)

 
 
 class TelegraafIE(InfoExtractor):
 
 
 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 = {
     _TEST = {

-        'url': 'http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html',
-        'md5': '83245a9779bcc4a24454bfd53c65b6dc',
+        'url': 'https://www.telegraaf.nl/video/734366489/historisch-scheepswrak-slaat-na-100-jaar-los',

         'info_dict': {
         'info_dict': {

-            'id': '24353229',
+            'id': 'gaMItuoSeUg2',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'Tikibad ontruimd wegens brand',
-            'description': 'md5:05ca046ff47b931f9b04855015e163a4',
-            'thumbnail': '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
+            'skip_download': True,

         },
     }
 
     def _real_extract(self, url):
         },
     }
 
     def _real_extract(self, url):

-        playlist_id = self._match_id(url)
+        article_id = self._match_id(url)
+
+        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']

 
 

-        webpage = self._download_webpage(url, playlist_id)
+        item = self._download_json(
+            'https://content.tmgvideo.nl/playlist/item=%s/playlist.json' % video_id,
+            video_id)['items'][0]
+        title = item['title']

 
 

-        playlist_url = self._search_regex(
-            r"iframe\.loadPlayer\('([^']+)'", webpage, 'player')
+        formats = []
+        locations = item.get('locations') or {}
+        for location in locations.get('adaptive', []):
+            manifest_url = location.get('src')
+            if not manifest_url:
+                continue
+            ext = determine_ext(manifest_url)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    manifest_url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
+            elif ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    manifest_url, video_id, mpd_id='dash', fatal=False))
+            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': src,
+                'width': int_or_none(location.get('width')),
+                'height': int_or_none(location.get('height')),
+                'format_id': 'http' + ('-%s' % label if label else ''),
+            })

 
 

-        entries = self._extract_xspf_playlist(playlist_url, playlist_id)
-        title = remove_end(self._og_search_title(webpage), ' - VIDEO')
-        description = self._og_search_description(webpage)
+        self._sort_formats(formats)

 
 

-        return self.playlist_result(entries, playlist_id, title, description)
+        return {
+            'id': video_id,
+            'title': title,
+            'description': item.get('description'),
+            'formats': formats,
+            'duration': int_or_none(item.get('duration')),
+            'thumbnail': item.get('poster'),
+            'timestamp': parse_iso8601(item.get('datecreated'), ' '),
+        }