[vimple] Fix extraction (Closes #5448)

youtube_dl/extractor/vimple.py

@@ -1,75 +1,54 @@
-# coding: utf-8
 from __future__ import unicode_literals
 
-import base64
-import re
-import xml.etree.ElementTree
-import zlib
-
 from .common import InfoExtractor
 from ..utils import int_or_none
 
 
 class VimpleIE(InfoExtractor):
-    IE_DESC = 'Vimple.ru'
+    IE_DESC = 'Vimple - one-click video hosting'
-    _VALID_URL = r'https?://(player.vimple.ru/iframe|vimple.ru)/(?P<id>[a-f0-9]{10,})'
+    _VALID_URL = r'https?://(?:player\.vimple\.ru/iframe|vimple\.ru)/(?P<id>[\da-f-]{32,36})'
     _TESTS = [
         {
             'url': 'http://vimple.ru/c0f6b1687dcd4000a97ebe70068039cf',
             'md5': '2e750a330ed211d3fd41821c6ad9a279',
             'info_dict': {
-                'id': 'c0f6b1687dcd4000a97ebe70068039cf',
+                'id': 'c0f6b168-7dcd-4000-a97e-be70068039cf',
                 'ext': 'mp4',
                 'title': 'Sunset',
                 'duration': 20,
                 'thumbnail': 're:https?://.*?\.jpg',
             },
-        },
+        }, {
+            'url': 'http://player.vimple.ru/iframe/52e1beec-1314-4a83-aeac-c61562eadbf9',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
-
-        iframe_url = 'http://player.vimple.ru/iframe/%s' % video_id
-
-        iframe = self._download_webpage(
-            iframe_url, video_id,
-            note='Downloading iframe', errnote='unable to fetch iframe')
-        player_url = self._html_search_regex(
-            r'"(http://player.vimple.ru/flash/.+?)"', iframe, 'player url')
 
-        player = self._request_webpage(
+        webpage = self._download_webpage(
-            player_url, video_id, note='Downloading swf player').read()
+            'http://player.vimple.ru/iframe/%s' % video_id, video_id)
 
-        player = zlib.decompress(player[8:])
+        playlist = self._parse_json(
+            self._search_regex(
+                r'sprutoData\s*:\s*({.+?}),\r\n', webpage, 'spruto data'),
+            video_id)['playlist'][0]
 
-        xml_pieces = re.findall(b'([a-zA-Z0-9 =+/]{500})', player)
+        title = playlist['title']
-        xml_pieces = [piece[1:-1] for piece in xml_pieces]
+        video_id = playlist.get('videoId') or video_id
+        thumbnail = playlist.get('posterUrl') or playlist.get('thumbnailUrl')
+        duration = int_or_none(playlist.get('duration'))
 
-        xml_data = b''.join(xml_pieces)
+        formats = [{
-        xml_data = base64.b64decode(xml_data)
+            'url': f['url'],
-
+        } for f in playlist['video']]
-        xml_data = xml.etree.ElementTree.fromstring(xml_data)
+        self._sort_formats(formats)
-
-        video = xml_data.find('Video')
-        quality = video.get('quality')
-        q_tag = video.find(quality.capitalize())
-
-        formats = [
-            {
-                'url': q_tag.get('url'),
-                'tbr': int(q_tag.get('bitrate')),
-                'filesize': int(q_tag.get('filesize')),
-                'format_id': quality,
-            },
-        ]
 
         return {
             'id': video_id,
-            'title': video.find('Title').text,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
             'formats': formats,
-            'thumbnail': video.find('Poster').get('url'),
-            'duration': int_or_none(video.get('duration')),
-            'webpage_url': video.find('Share').get('videoPageUrl'),
         }