Switch codebase to use sanitized_Request instead of

[youtube-dl] / youtube_dl / extractor / voicerepublic.py

diff --git a/youtube_dl/extractor/voicerepublic.py b/youtube_dl/extractor/voicerepublic.py
index 1106c655b871367abf3aad2ae65867d167a36353..93d15a556dedb6e0589dc5f393e0a01ad5b4a8a0 100644 (file)
--- a/youtube_dl/extractor/voicerepublic.py
+++ b/youtube_dl/extractor/voicerepublic.py
@@ -3,14 +3,12 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_request,
-    compat_urlparse,
-)
+from ..compat import compat_urlparse
 from ..utils import (
     ExtractorError,
     determine_ext,
     int_or_none,
+    sanitized_Request,
 )
 
 
@@ -37,7 +35,7 @@ class VoiceRepublicIE(InfoExtractor):
     def _real_extract(self, url):
         display_id = self._match_id(url)
 
-        req = compat_urllib_request.Request(
+        req = sanitized_Request(
             compat_urlparse.urljoin(url, '/talks/%s' % display_id))
         # Older versions of Firefox get redirected to an "upgrade browser" page
         req.add_header('User-Agent', 'youtube-dl')
@@ -47,12 +45,10 @@ class VoiceRepublicIE(InfoExtractor):
             raise ExtractorError(
                 'Audio is still queued for processing', expected=True)
 
-        data = self._parse_json(
-            self._search_regex(
-                r'(?s)return ({.+?});\s*\n', webpage,
-                'data', default=None),
-            display_id, fatal=False)
-
+        config = self._search_regex(
+            r'(?s)return ({.+?});\s*\n', webpage,
+            'data', default=None)
+        data = self._parse_json(config, display_id, fatal=False) if config else None
         if data:
             title = data['title']
             description = data.get('teaser')
@@ -74,12 +70,14 @@ class VoiceRepublicIE(InfoExtractor):
                 [r"id='jc-(\d+)'", r"data-shareable-id='(\d+)'"],
                 webpage, 'talk id', default=None) or display_id
             duration = None
+            player = self._search_regex(
+                r"class='vr-player jp-jplayer'([^>]+)>", webpage, 'player')
             formats = [{
                 'url': compat_urlparse.urljoin(url, talk_url),
                 'format_id': format_id,
                 'ext': determine_ext(talk_url) or format_id,
                 'vcodec': 'none',
-            } for format_id, talk_url in re.findall(r"data-([^=]+)='([^']+)'", webpage)]
+            } for format_id, talk_url in re.findall(r"data-([^=]+)='([^']+)'", player)]
         self._sort_formats(formats)
 
         thumbnail = self._og_search_thumbnail(webpage)