[trilulilu] extract info from webpage

youtube_dl/extractor/trilulilu.py

@@ -1,11 +1,14 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
-    int_or_none,
     parse_iso8601,
+    js_to_json,
+    parse_duration,
 )
 
 
@@ -22,10 +25,6 @@ class TriluliluIE(InfoExtractor):
             'uploader_id': 'chipy',
             'upload_date': '20120304',
             'timestamp': 1330830647,
-            'uploader': 'chipy',
-            'view_count': int,
-            'like_count': int,
-            'comment_count': int,
         },
     }, {
         'url': 'http://www.trilulilu.ro/adena-ft-morreti-inocenta',
@@ -37,57 +36,59 @@ class TriluliluIE(InfoExtractor):
             'description': 'pop music',
             'uploader_id': 'VEVOmixt',
             'upload_date': '20151204',
-            'uploader': 'VEVOmixt',
             'timestamp': 1449187937,
-            'view_count': int,
-            'like_count': int,
-            'comment_count': int,
         },
     }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
-        media_info = self._download_json('http://m.trilulilu.ro/%s?format=json' % display_id, display_id)
+        webpage = self._download_webpage(url, display_id)
 
-        media_class = media_info.get('class')
+        if re.search(r'Fişierul nu este disponibil pentru vizionare în ţara dumneavoastră', webpage):
+            raise ExtractorError(
+                'This video is not available in your country.', expected=True)
+        elif re.search('Fişierul poate fi accesat doar de către prietenii lui', webpage):
+            raise ExtractorError('This video is private.', expected=True)
+
+        player_vars = self._parse_json(
+            self._search_regex(r'(?s)playerVars\s*=\s*({.+?});', webpage, 'player vars'),
+            display_id, js_to_json)
+
+        uploader_id, media_id = self._search_regex(
+            r'<input type="hidden" id="identifier" value="([^"]+)"',
+            webpage, 'identifier').split('|')
+
+        media_class = player_vars.get('fileClass')
         if media_class not in ('video', 'audio'):
             raise ExtractorError('not a video or an audio')
 
-        user = media_info.get('user', {})
-
-        thumbnail = media_info.get('cover_url')
-        if thumbnail:
-            thumbnail.format(width='1600', height='1200')
-
         # TODO: get correct ext for audio files
-        stream_type = media_info.get('stream_type')
         formats = [{
-            'url': media_info['href'],
+            'url': player_vars['file'],
-            'ext': stream_type,
+            'ext': player_vars.get('streamType'),
         }]
-        if media_info.get('is_hd'):
+
+        hd_url = player_vars.get('fileUrlHD')
+        if hd_url:
             formats.append({
                 'format_id': 'hd',
-                'url': media_info['hrefhd'],
+                'url': hd_url,
-                'ext': stream_type,
+                'ext': player_vars.get('fileTypeHD'),
             })
+
         if media_class == 'audio':
             formats[0]['vcodec'] = 'none'
         else:
             formats[0]['format_id'] = 'sd'
 
         return {
-            'id': media_info['identifier'].split('|')[1],
+            'id': media_id,
             'display_id': display_id,
             'formats': formats,
-            'title': media_info['title'],
+            'title': self._og_search_title(webpage),
-            'description': media_info.get('description'),
+            'description': self._og_search_description(webpage),
-            'thumbnail': thumbnail,
+            'thumbnail': player_vars.get('image'),
-            'uploader_id': user.get('username'),
+            'uploader_id': uploader_id,
-            'uploader': user.get('fullname'),
+            'timestamp': parse_iso8601(self._html_search_meta('uploadDate', webpage), ' '),
-            'timestamp': parse_iso8601(media_info.get('published'), ' '),
+            'duration': parse_duration(self._html_search_meta('duration', webpage)),
-            'duration': int_or_none(media_info.get('duration')),
-            'view_count': int_or_none(media_info.get('count_views')),
-            'like_count': int_or_none(media_info.get('count_likes')),
-            'comment_count': int_or_none(media_info.get('count_comments')),
         }