[csjw] Fix issues and improve extraction (closes #13525)

youtube_dl/extractor/cjsw.py

@@ -1,41 +1,66 @@
-# coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    unescapeHTML,
+)
 
 
 class CJSWIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?cjsw\.com/program/\S+/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?cjsw\.com/program/(?P<program>[^/]+)/episode/(?P<id>\d+)'
-    IE_NAME = 'cjsw'
     _TEST = {
         'url': 'http://cjsw.com/program/freshly-squeezed/episode/20170620',
         'md5': 'cee14d40f1e9433632c56e3d14977120',
         'info_dict': {
-            'id': '20170620',
+            'id': '91d9f016-a2e7-46c5-8dcb-7cbcd7437c41',
             'ext': 'mp3',
-            'title': 'Freshly Squeezed',
+            'title': 'Freshly Squeezed – Episode June 20, 2017',
-            'description': 'Sled Island artists featured // Live session with Phi Pho, followed by a live session with Sinzere & The Late Nights! // Stay Fresh Y\'all!!',
+            'description': 'md5:c967d63366c3898a80d0c7b0ff337202',
-        }
+            'series': 'Freshly Squeezed',
+            'episode_id': '20170620',
+        },
     }
 
     def _real_extract(self, url):
-        episode_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        program, episode_id = mobj.group('program', 'id')
+        audio_id = '%s/%s' % (program, episode_id)
 
         webpage = self._download_webpage(url, episode_id)
 
-        title = self._search_regex(
+        title = unescapeHTML(self._search_regex(
-            r'<button[^>]+data-showname=(["\'])(?P<title>(?!\1).+?)\1[^>]*>', webpage, 'title', group='title')
+            (r'<h1[^>]+class=["\']episode-header__title["\'][^>]*>(?P<title>[^<]+)',
-        description = self._html_search_regex(
+             r'data-audio-title=(["\'])(?P<title>(?:(?!\1).)+)\1'),
-            r'<p>(?P<description>.+?)</p>', webpage, 'description', fatal=False)
+            webpage, 'title', group='title'))
+
+        audio_url = self._search_regex(
+            r'<button[^>]+data-audio-src=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            webpage, 'audio url', group='url')
+
+        audio_id = self._search_regex(
+            r'/([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})\.mp3',
+            audio_url, 'audio id', default=audio_id)
+
         formats = [{
-            'url': self._search_regex(
+            'url': audio_url,
-                r'<button[^>]+data-audio-src=(["\'])(?P<audio_url>(?!\1).+?)\1[^>]*>', webpage, 'audio_url', group='audio_url'),
+            'ext': determine_ext(audio_url, 'mp3'),
-            'ext': 'mp3',
             'vcodec': 'none',
         }]
+
+        description = self._html_search_regex(
+            r'<p>(?P<description>.+?)</p>', webpage, 'description', fatal=False)
+        series = self._search_regex(
+            r'data-showname=(["\'])(?P<name>(?:(?!\1).)+)\1', webpage,
+            'series', default=program, group='name')
+
         return {
-            'id': episode_id,
+            'id': audio_id,
             'title': title,
             'description': description,
             'formats': formats,
+            'series': series,
+            'episode_id': episode_id,
         }