[rbmaradio] Improve, simplify and extract all formats (Closes #10242)

youtube_dl/extractor/rbmaradio.py

@@ -1,46 +1,71 @@
-# encoding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    clean_html
+    clean_html,
+    int_or_none,
+    unified_timestamp,
+    update_url_query,
 )
 
 
 class RBMARadioIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rbmaradio\.com/shows/[^/]+/episodes/(?P<id>[^/]+)$'
+    _VALID_URL = r'https?://(?:www\.)?rbmaradio\.com/shows/(?P<show_id>[^/]+)/episodes/(?P<id>[^/?#&]+)'
     _TEST = {
         'url': 'https://www.rbmaradio.com/shows/main-stage/episodes/ford-lopatin-live-at-primavera-sound-2011',
         'md5': '6bc6f9bcb18994b4c983bc3bf4384d95',
         'info_dict': {
             'id': 'ford-lopatin-live-at-primavera-sound-2011',
             'ext': 'mp3',
-            'description': 'Joel Ford and Daniel ’Oneohtrix Point Never’ Lopatin fly their midified pop extravaganza to Spain. Live at Primavera Sound 2011.',
+            'title': 'Main Stage - Ford & Lopatin',
-            'title': 'Ford & Lopatin - Main Stage',
+            'description': 'md5:4f340fb48426423530af5a9d87bd7b91',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 2452,
+            'timestamp': 1307103164,
+            'upload_date': '20110603',
         },
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        show_id = mobj.group('show_id')
+        episode_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, episode_id)
+
+        episode = self._parse_json(
+            self._search_regex(
+                r'__INITIAL_STATE__\s*=\s*({.+?})\s*</script>',
+                webpage, 'json data'),
+            episode_id)['episodes'][show_id][episode_id]
+
+        title = episode['title']
 
-        webpage = self._download_webpage(url, video_id)
+        show_title = episode.get('showTitle')
-        json_data = self._search_regex(r'<script>window\.__INITIAL_STATE__\s*=\s*(.+?)</script>',
+        if show_title:
-                                       webpage, 'json data')
+            title = '%s - %s' % (show_title, title)
-        data = self._parse_json(json_data, video_id)
 
-        item = None
+        formats = [{
-        for episode in data['episodes']:
+            'url': update_url_query(episode['audioURL'], query={'cbr': abr}),
-            items = data['episodes'][episode]
+            'format_id': compat_str(abr),
-            if video_id in items:
+            'abr': abr,
-                item = items[video_id]
+            'vcodec': 'none',
+        } for abr in (96, 128, 256)]
 
-        video_url = item['audioURL'] + '?cbr=256'
+        description = clean_html(episode.get('longTeaser'))
+        thumbnail = self._proto_relative_url(episode.get('imageURL', {}).get('landscape'))
+        duration = int_or_none(episode.get('duration'))
+        timestamp = unified_timestamp(episode.get('publishedAt'))
 
         return {
-            'id': video_id,
+            'id': episode_id,
-            'url': video_url,
+            'title': title,
-            'title': item.get('title') + ' - ' + item.get('showTitle'),
+            'description': description,
-            'description': clean_html(item.get('longTeaser')),
+            'thumbnail': thumbnail,
-            'thumbnail': self._proto_relative_url(item.get('imageURL', {}).get('landscape')),
+            'duration': duration,
-            'duration': item.get('duration'),
+            'timestamp': timestamp,
+            'formats': formats,
         }