[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/libsyn.py b/youtube_dl/extractor/libsyn.py
index 4b5029f892cb835e35e02aaa5e020a42447d95af..2cf444258a2b57e7509a5ce83efdab0125bf20e3 100644 (file)
--- a/youtube_dl/extractor/libsyn.py
+++ b/youtube_dl/extractor/libsyn.py
@@ -1,41 +1,93 @@
-# encoding: utf-8
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
+    get_element_by_class,
+    parse_duration,
+    strip_or_none,
     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/6385796/',
+        'md5': '2a55e75496c790cdeb058e7e6c087746',
+        'info_dict': {
+            'id': '6385796',
+            'ext': 'mp3',
+            'title': "Champion Minded - Developing a Growth Mindset",
+            # description fetched using another request:
+            # http://html5-player.libsyn.com/embed/getitemdetails?item_id=6385796
+            # 'description': 'In this episode, Allistair talks about the importance of developing a growth mindset, not only in sports, but in life too.',
+            'upload_date': '20180320',
+            '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
-            }]
+        url, video_id = re.match(self._VALID_URL, url).groups()
+        webpage = self._download_webpage(url, video_id)
+
+        data = self._parse_json(self._search_regex(
+            r'var\s+playlistItem\s*=\s*({.+?});',
+            webpage, 'JSON data block'), video_id)
+
+        episode_title = data.get('item_title') or get_element_by_class('episode-title', webpage)
+        if not episode_title:
+            self._search_regex(
+                [r'data-title="([^"]+)"', r'<title>(.+?)</title>'],
+                webpage, 'episode title')
+        episode_title = episode_title.strip()
+
+        podcast_title = strip_or_none(clean_html(self._search_regex(
+            r'<h3>([^<]+)</h3>', webpage, 'podcast title',
+            default=None) or get_element_by_class('podcast-title', webpage)))
+
+        title = '%s - %s' % (podcast_title, episode_title) if podcast_title else episode_title
+
+        formats = []
+        for k, format_id in (('media_url_libsyn', 'libsyn'), ('media_url', 'main'), ('download_link', 'download')):
+            f_url = data.get(k)
+            if not f_url:
+                continue
+            formats.append({
+                'url': f_url,
+                'format_id': format_id,
+            })
+
+        description = self._html_search_regex(
+            r'<p\s+id="info_text_body">(.+?)</p>', webpage,
+            'description', default=None)
+        if description:
+            # Strip non-breaking and normal spaces
+            description = description.replace('\u00A0', ' ').strip()
+        release_date = unified_strdate(self._search_regex(
+            r'<div class="release_date">Released: ([^<]+)<',
+            webpage, 'release date', default=None) or data.get('release_date'))
 
         return {
-            'id': display_id,
-            'title': podcast_episode_title,
-            'description': podcast_description,
-            'upload_date': podcast_date,
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': data.get('thumbnail_url'),
+            'upload_date': release_date,
+            'duration': parse_duration(data.get('duration')),
             'formats': formats,
         }