Fix "invalid escape sequences" error on Python 3.6

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

diff --git a/youtube_dl/extractor/voicerepublic.py b/youtube_dl/extractor/voicerepublic.py
index 1106c655b871367abf3aad2ae65867d167a36353..59e1359c48628af9b4c53bedc337fa6b9b3d1396 100644 (file)
--- a/youtube_dl/extractor/voicerepublic.py
+++ b/youtube_dl/extractor/voicerepublic.py
@@ -4,13 +4,14 @@ import re
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_request,
+    compat_str,
     compat_urlparse,
 )
 from ..utils import (
     ExtractorError,
     determine_ext,
     int_or_none,
+    sanitized_Request,
 )
 
 
@@ -18,14 +19,14 @@ class VoiceRepublicIE(InfoExtractor):
     _VALID_URL = r'https?://voicerepublic\.com/(?:talks|embed)/(?P<id>[0-9a-z-]+)'
     _TESTS = [{
         'url': 'http://voicerepublic.com/talks/watching-the-watchers-building-a-sousveillance-state',
-        'md5': '0554a24d1657915aa8e8f84e15dc9353',
+        'md5': 'b9174d651323f17783000876347116e3',
         'info_dict': {
             'id': '2296',
             'display_id': 'watching-the-watchers-building-a-sousveillance-state',
             'ext': 'm4a',
             'title': 'Watching the Watchers: Building a Sousveillance State',
-            'description': 'md5:715ba964958afa2398df615809cfecb1',
-            'thumbnail': 're:^https?://.*\.(?:png|jpg)$',
+            'description': 'Secret surveillance programs have metadata too. The people and companies that operate secret surveillance programs can be surveilled.',
+            'thumbnail': r're:^https?://.*\.(?:png|jpg)$',
             'duration': 1800,
             'view_count': int,
         }
@@ -37,7 +38,7 @@ 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')
@@ -47,16 +48,14 @@ 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')
-            talk_id = data.get('talk_id') or display_id
+            talk_id = compat_str(data.get('talk_id') or display_id)
             talk = data['talk']
             duration = int_or_none(talk.get('duration'))
             formats = [{
@@ -74,12 +73,14 @@ 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)