Update extractor for current site behaviour

youtube_dl/extractor/voe.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import base64
+
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -12,38 +14,48 @@
 class VOEIE(InfoExtractor):
     IE_NAME = 'voe'
     IE_DESC = 'VOE.SX'
-    _VALID_URL = r'https?://voe\.sx/(e/)?(?P<id>[a-z0-9]+)'
-    _TEST = {
+    _VALID_URL = r'https?://voe\.sx(?P<path>/(e/)?(?P<id>[a-z0-9]+))'
+    _TESTS = [{
         'url': 'https://voe.sx/e/ng7ja5n5n2y8',
         'info_dict': {
             'id': 'ng7ja5n5n2y8',
             'title': 'md5:a86687fb962742f04652aee19ad34e06',
             'thumbnail': r're:^https?://.*\.jpg$',
             'ext': 'm3u8',
         },
-    }
+    }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        video_id, video_path = self._match_valid_url(url).group('id', 'path')
 
         webpage = self._download_webpage(
             'https://voe.sx/e/%s' % video_id, video_id)
 
-        sources = self._parse_json(
-            self._search_regex(r'\bsources\s*=\s*(\{[^}]+\})', webpage, 'sources'),
-            video_id, transform_source=js_to_json)
+        player_url = self._search_regex(
+            r'''("|')(?P<url>https://(?!voe\.sx/)[^/]+%s)\1\s*;''' % (path,),
+            webpage, 'redirect', group='url', default=None)
+        if player_url:
+            webpage = self._download_webpage(
+                player_url, video_id, note='Redirecting to player page')
+
+
+        sources = self._search_json(
+            r'\bsources\s*=', webpage, 'sources', video_id,  transform_source=js_to_json)      
 
         title = self._search_regex(
             r'<title>(?:Watch\s+)?(?P<title>.+?)(?:-\s+VOE\s+\|.+)?</title>',
             webpage, 'title', group='title')
 
         formats = []
-
-        f_url = url_or_none(sources.get('hls'))
+        for fmt in ('mp4, 'hls'):
+            if fmt not in sources:
+                continue
+            sources[fmt] = url_or_none(sources['fmt']) or url_or_none(base64.b64decode(sources['fmt']))
+            f_url = sources.get('hls')
         if f_url:
             formats.extend(self._extract_m3u8_formats(
                 f_url, video_id, entry_protocol='m3u8_native', fatal=False))
-        f_url = url_or_none(sources.get('mp4'))
+        f_url = sources.get('mp4')
         if f_url:
             formats.append({
                 'url': f_url,
@@ -54,7 +66,7 @@ def _real_extract(self, url):
         self._sort_formats(formats)
 
         thumbnail = url_or_none(self._search_regex(
-            r'(?:VOEPlayer.|data-)poster\s*=\s*(["\'])(?P<thumbnail>(?:(?!\1)\S)+)\1',
+            r'''(?:VOEPlayer\s*\.\s*|data-)poster\s*=\s*("|')(?P<thumbnail>(?:(?!\1)\S)+)\1''',
             webpage, 'thumbnail', group='thumbnail', default=None))
 
         return {