Fix "invalid escape sequences" error on Python 3.6

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

diff --git a/youtube_dl/extractor/libsyn.py b/youtube_dl/extractor/libsyn.py
index 4b5029f892cb835e35e02aaa5e020a42447d95af..4750b03a3fb2f47818858338b7eb9a8b4889c012 100644 (file)
--- a/youtube_dl/extractor/libsyn.py
+++ b/youtube_dl/extractor/libsyn.py
@@ -1,41 +1,69 @@
-# encoding: utf-8
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
 from .common import InfoExtractor
-from ..utils import (
-    unified_strdate,
-)
+from ..utils import unified_strdate
+
 
 class LibsynIE(InfoExtractor):
-    _VALID_URL = r'(?:https?:)?//html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+)(?:/.*)?'
+    _VALID_URL = r'(?P<mainurl>https?://html5-player\.libsyn\.com/embed/episode/id/(?P<id>[0-9]+))'
+
+    _TESTS = [{
+        'url': 'http://html5-player.libsyn.com/embed/episode/id/3377616/',
+        'md5': '443360ee1b58007bc3dcf09b41d093bb',
+        'info_dict': {
+            'id': '3377616',
+            'ext': 'mp3',
+            'title': "The Daily Show Podcast without Jon Stewart - Episode 12: Bassem Youssef: Egypt's Jon Stewart",
+            'description': 'md5:601cb790edd05908957dae8aaa866465',
+            'upload_date': '20150220',
+            'thumbnail': 're:^https?://.*',
+        },
+    }, {
+        'url': 'https://html5-player.libsyn.com/embed/episode/id/3727166/height/75/width/200/theme/standard/direction/no/autoplay/no/autonext/no/thumbnail/no/preload/no/no_addthis/no/',
+        'md5': '6c5cb21acd622d754d3b1a92b582ce42',
+        'info_dict': {
+            'id': '3727166',
+            'ext': 'mp3',
+            'title': 'Clients From Hell Podcast - How a Sex Toy Company Kickstarted my Freelance Career',
+            'upload_date': '20150818',
+            'thumbnail': 're:^https?://.*',
+        }
+    }]
 
     def _real_extract(self, url):
-        if url.startswith('//'):
-            url = 'https:' + url
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
-
-        podcast_title         = self._search_regex(r'<h2>(.*?)</h2>', webpage, 'show title')
-        podcast_episode_title = self._search_regex(r'<h3>(.*?)</h3>', webpage, 'episode title')
-        podcast_date          = unified_strdate(self._search_regex(r'<div class="release_date">Released: (.*?)</div>', webpage, 'release date'))
-        podcast_description   = self._search_regex(r'<div id="info_text_body">(.*?)</div>', webpage, 'description')
-
-        url0 = self._search_regex(r'var mediaURLLibsyn = "(?P<url0>https?://.*)";', webpage, 'first media URL')
-        url1 = self._search_regex(r'var mediaURL = "(?P<url1>https?://.*)";', webpage, 'second media URL')
-
-        if url0 != url1:
-            formats = [{
-                'url': url0
-            }, {
-                'url': url1
-            }]
-        else:
-            formats = [{
-                'url': url0
-            }]
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+        url = m.group('mainurl')
+        webpage = self._download_webpage(url, video_id)
+
+        formats = [{
+            'url': media_url,
+        } for media_url in set(re.findall(r'var\s+mediaURL(?:Libsyn)?\s*=\s*"([^"]+)"', webpage))]
+
+        podcast_title = self._search_regex(
+            r'<h2>([^<]+)</h2>', webpage, 'podcast title', default=None)
+        episode_title = self._search_regex(
+            r'(?:<div class="episode-title">|<h3>)([^<]+)</', webpage, 'episode title')
+
+        title = '%s - %s' % (podcast_title, episode_title) if podcast_title else episode_title
+
+        description = self._html_search_regex(
+            r'<div id="info_text_body">(.+?)</div>', webpage,
+            'description', default=None)
+        thumbnail = self._search_regex(
+            r'<img[^>]+class="info-show-icon"[^>]+src="([^"]+)"',
+            webpage, 'thumbnail', fatal=False)
+        release_date = unified_strdate(self._search_regex(
+            r'<div class="release_date">Released: ([^<]+)<', webpage, 'release date', fatal=False))
 
         return {
-            'id': display_id,
-            'title': podcast_episode_title,
-            'description': podcast_description,
-            'upload_date': podcast_date,
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'upload_date': release_date,
             'formats': formats,
         }