Merge tag 'upstream/2016.08.17'

[youtubedl] / youtube_dl / extractor / rbmaradio.py

diff --git a/youtube_dl/extractor/rbmaradio.py b/youtube_dl/extractor/rbmaradio.py
index 0c75eee2a8d5e17a0b9b8f873fbd15e6fc3615ac..471928ef86b5d434953fc694eef0bb7da8edd334 100644 (file)
--- a/youtube_dl/extractor/rbmaradio.py
+++ b/youtube_dl/extractor/rbmaradio.py
@@ -1,44 +1,71 @@
-import json
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    compat_urllib_parse_urlparse,
-
-    ExtractorError,
+    clean_html,
+    int_or_none,
+    unified_timestamp,
+    update_url_query,
 )
 
 
 class RBMARadioIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rbmaradio\.com/shows/(?P<videoID>[^/]+)$'
+    _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',
+            'title': 'Main Stage - Ford & Lopatin',
+            'description': 'md5:4f340fb48426423530af5a9d87bd7b91',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 2452,
+            'timestamp': 1307103164,
+            'upload_date': '20110603',
+        },
+    }
 
     def _real_extract(self, url):
-        m = re.match(self._VALID_URL, url)
-        video_id = m.group('videoID')
-
-        webpage = self._download_webpage(url, video_id)
-
-        json_data = self._search_regex(r'window\.gon.*?gon\.show=(.+?);$',
-            webpage, u'json data', flags=re.MULTILINE)
-
-        try:
-            data = json.loads(json_data)
-        except ValueError as e:
-            raise ExtractorError(u'Invalid JSON: ' + str(e))
-
-        video_url = data['akamai_url'] + '&cbr=256'
-        url_parts = compat_urllib_parse_urlparse(video_url)
-        video_ext = url_parts.path.rpartition('.')[2]
-        info = {
-                'id': video_id,
-                'url': video_url,
-                'ext': video_ext,
-                'title': data['title'],
-                'description': data.get('teaser_text'),
-                'location': data.get('country_of_origin'),
-                'uploader': data.get('host', {}).get('name'),
-                'uploader_id': data.get('host', {}).get('slug'),
-                'thumbnail': data.get('image', {}).get('large_url_2x'),
-                'duration': data.get('duration'),
+        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']
+
+        show_title = episode.get('showTitle')
+        if show_title:
+            title = '%s - %s' % (show_title, title)
+
+        formats = [{
+            'url': update_url_query(episode['audioURL'], query={'cbr': abr}),
+            'format_id': compat_str(abr),
+            'abr': abr,
+            'vcodec': 'none',
+        } for abr in (96, 128, 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': episode_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'timestamp': timestamp,
+            'formats': formats,
         }
-        return [info]