[hitrecord] Improve (closes #11626)
authorSergey M․ <dstftw@gmail.com>
Sun, 8 Jan 2017 13:15:39 +0000 (20:15 +0700)
committerSergey M․ <dstftw@gmail.com>
Sun, 8 Jan 2017 13:17:18 +0000 (20:17 +0700)
youtube_dl/extractor/hitrecord.py

index 35bbb3eb8976e17df00bef6b5e1ed7166aeb0017..01a6946d0d706691cd6ca43c0679cdfb4069924a 100644 (file)
@@ -1,17 +1,17 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     clean_html,
+    float_or_none,
     int_or_none,
-    unified_strdate,
+    try_get,
 )
-from ..compat import compat_str
 
 
 class HitRecordIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?hitrecord\.org/records/(?P<id>\d+)'
-
     _TEST = {
         'url': 'https://hitrecord.org/records/2954362',
         'md5': 'fe1cdc2023bce0bbb95c39c57426aa71',
@@ -20,29 +20,49 @@ class HitRecordIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'A Very Different World (HITRECORD x ACLU)',
             'description': 'md5:e62defaffab5075a5277736bead95a3d',
-            'release_date': '20160818',
+            'duration': 139.327,
             'timestamp': 1471557582,
+            'upload_date': '20160818',
             'uploader': 'Zuzi.C12',
             'uploader_id': '362811',
+            'view_count': int,
+            'like_count': int,
+            'comment_count': int,
+            'tags': list,
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_info = self._download_json('https://hitrecord.org/api/web/records/' + video_id, video_id)
-        user_info = video_info.get('user', {})
+
+        video = self._download_json(
+            'https://hitrecord.org/api/web/records/%s' % video_id, video_id)
+
+        title = video['title']
+        video_url = video['source_url']['mp4_url']
+
+        tags = None
+        tags_list = try_get(video, lambda x: x['tags'], list)
+        if tags_list:
+            tags = [
+                t['text']
+                for t in tags_list
+                if isinstance(t, dict) and t.get('text') and
+                isinstance(t['text'], compat_str)]
 
         return {
             'id': video_id,
-            'title': video_info['title'],
-            'url': video_info['source_url']['mp4_url'],
-            'description': clean_html(video_info.get('body')),
-            'uploader': user_info.get('username'),
-            'uploader_id': compat_str(user_info.get('id')),
-            'release_date': unified_strdate(video_info.get('created_at')),
-            'timestamp': video_info.get('created_at_i'),
-            'view_count': int_or_none(video_info.get('total_views_count')),
-            'like_count': int_or_none(video_info.get('hearts_count')),
-            'comment_count': int_or_none(video_info.get('comments_count')),
-            'tags': [tag.get('text') for tag in video_info.get('tags', [])],
+            'url': video_url,
+            'title': title,
+            'description': clean_html(video.get('body')),
+            'duration': float_or_none(video.get('duration'), 1000),
+            'timestamp': int_or_none(video.get('created_at_i')),
+            'uploader': try_get(
+                video, lambda x: x['user']['username'], compat_str),
+            'uploader_id': try_get(
+                video, lambda x: compat_str(x['user']['id'])),
+            'view_count': int_or_none(video.get('total_views_count')),
+            'like_count': int_or_none(video.get('hearts_count')),
+            'comment_count': int_or_none(video.get('comments_count')),
+            'tags': tags,
         }