Annotate some highlights of the release.

[youtubedl] / youtube_dl / extractor / fczenit.py

diff --git a/youtube_dl/extractor/fczenit.py b/youtube_dl/extractor/fczenit.py
index 8d1010b88c83dcbfd3e71e9f20275bf6fb9c9d21..8db7c59638704ee4febe9d371c0c85c397fb50f9 100644 (file)
--- a/youtube_dl/extractor/fczenit.py
+++ b/youtube_dl/extractor/fczenit.py
@@ -2,7 +2,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..utils import (
+    int_or_none,
+    float_or_none,
+)
 
 
 class FczenitIE(InfoExtractor):
@@ -14,6 +17,8 @@ class FczenitIE(InfoExtractor):
             'id': '41044',
             'ext': 'mp4',
             'title': 'Так пишется история: казанский разгром ЦСКА на «Зенит-ТВ»',
+            'timestamp': 1462283735,
+            'upload_date': '20160503',
         },
     }
 
@@ -21,28 +26,31 @@ class FczenitIE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        video_title = self._html_search_regex(
-            r'<[^>]+class=\"photoalbum__title\">([^<]+)', webpage, 'title')
+        msi_id = self._search_regex(
+            r"(?s)config\s*=\s*{.+?video_id\s*:\s*'([^']+)'", webpage, 'msi id')
 
-        video_items = self._parse_json(self._search_regex(
-            r'arrPath\s*=\s*JSON\.parse\(\'(.+)\'\)', webpage, 'video items'),
-            video_id)
-
-        def merge_dicts(*dicts):
-            ret = {}
-            for a_dict in dicts:
-                ret.update(a_dict)
-            return ret
+        msi_data = self._download_json(
+            'http://player.fc-zenit.ru/msi/video', msi_id, query={
+                'video': msi_id,
+            })['data']
+        title = msi_data['name']
 
         formats = [{
-            'url': compat_urlparse.urljoin(url, video_url),
-            'tbr': int(tbr),
-        } for tbr, video_url in merge_dicts(*video_items).items()]
+            'format_id': q.get('label'),
+            'url': q['url'],
+            'height': int_or_none(q.get('label')),
+        } for q in msi_data['qualities'] if q.get('url')]
 
         self._sort_formats(formats)
 
+        tags = [tag['label'] for tag in msi_data.get('tags', []) if tag.get('label')]
+
         return {
             'id': video_id,
-            'title': video_title,
+            'title': title,
+            'thumbnail': msi_data.get('preview'),
             'formats': formats,
+            'duration': float_or_none(msi_data.get('duration')),
+            'timestamp': int_or_none(msi_data.get('date')),
+            'tags': tags,
         }