Merge branch 'kusi' of https://github.com/mutantmonkey/youtube-dl into mutantmonkey...

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

diff --git a/youtube_dl/extractor/jwplatform.py b/youtube_dl/extractor/jwplatform.py
index cdc095a79aa6ff8ccf0404293b9011a1f5d4d780..6770685d7027c3738fba35f3e057f6be2a3a512c 100644 (file)
--- a/youtube_dl/extractor/jwplatform.py
+++ b/youtube_dl/extractor/jwplatform.py
@@ -7,33 +7,9 @@ from .common import InfoExtractor
 from ..utils import int_or_none
 
 
-class JWPlatformIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
-    _TEST = {
-        'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
-        'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
-        'info_dict': {
-            'id': 'nPripu9l',
-            'ext': 'mov',
-            'title': 'Big Buck Bunny Trailer',
-            'description': 'Big Buck Bunny is a short animated film by the Blender Institute. It is made using free and open source software.',
-            'upload_date': '20081127',
-            'timestamp': 1227796140,
-        }
-    }
-
-    @staticmethod
-    def _extract_url(webpage):
-        mobj = re.search(
-            r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
-            webpage)
-        if mobj:
-            return mobj.group('url')
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        json_data = self._download_json('http://content.jwplatform.com/feeds/%s.json' % video_id, video_id)
-        video_data = json_data['playlist'][0]
+class JWPlatformBaseIE(InfoExtractor):
+    def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True):
+        video_data = jwplayer_data['playlist'][0]
         subtitles = {}
         for track in video_data['tracks']:
             if track['kind'] == 'captions':
@@ -43,10 +19,9 @@ class JWPlatformIE(InfoExtractor):
         for source in video_data['sources']:
             source_url = self._proto_relative_url(source['file'])
             source_type = source.get('type') or ''
-            if source_type == 'application/vnd.apple.mpegurl':
-                m3u8_formats = self._extract_m3u8_formats(source_url, video_id, 'mp4', 'm3u8_native', fatal=None)
-                if m3u8_formats:
-                    formats.extend(m3u8_formats)
+            if source_type in ('application/vnd.apple.mpegurl', 'hls'):
+                formats.extend(self._extract_m3u8_formats(
+                    source_url, video_id, 'mp4', 'm3u8_native', fatal=False))
             elif source_type.startswith('audio'):
                 formats.append({
                     'url': source_url,
@@ -62,10 +37,39 @@ class JWPlatformIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'title': video_data['title'],
+            'title': video_data['title'] if require_title else video_data.get('title'),
             'description': video_data.get('description'),
             'thumbnail': self._proto_relative_url(video_data.get('image')),
             'timestamp': int_or_none(video_data.get('pubdate')),
             'subtitles': subtitles,
             'formats': formats,
         }
+
+
+class JWPlatformIE(JWPlatformBaseIE):
+    _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
+    _TEST = {
+        'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
+        'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
+        'info_dict': {
+            'id': 'nPripu9l',
+            'ext': 'mov',
+            'title': 'Big Buck Bunny Trailer',
+            'description': 'Big Buck Bunny is a short animated film by the Blender Institute. It is made using free and open source software.',
+            'upload_date': '20081127',
+            'timestamp': 1227796140,
+        }
+    }
+
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
+            webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        json_data = self._download_json('http://content.jwplatform.com/feeds/%s.json' % video_id, video_id)
+        return self._parse_jwplayer_data(json_data, video_id)