debian/copyright: use spaces rather than tabs to start continuation lines.

[youtubedl] / youtube_dl / extractor / jwplatform.py

diff --git a/youtube_dl/extractor/jwplatform.py b/youtube_dl/extractor/jwplatform.py
index 8e90d59868c380d1f486d838d8b6f3283bdc2b58..2aabd98b5bbaf36efecdc76415bf52854705bd1e 100644 (file)
--- a/youtube_dl/extractor/jwplatform.py
+++ b/youtube_dl/extractor/jwplatform.py
@@ -4,12 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..utils import int_or_none

 
 
 class JWPlatformIE(InfoExtractor):
 
 
 class JWPlatformIE(InfoExtractor):

-    _VALID_URL = r'(?:https?://content\.jwplatform\.com/(?:feeds|players|jw6)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
-    _TEST = {
+    _VALID_URL = r'(?:https?://(?:content\.jwplatform|cdn\.jwplayer)\.com/(?:(?:feed|player|thumb|preview)s|jw6|v2/media)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
+    _TESTS = [{

         'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
         'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
         'info_dict': {
         'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
         'md5': 'fa8899fa601eb7c83a64e9d568bdf325',
         'info_dict': {


@@ -20,51 +19,23 @@ class JWPlatformIE(InfoExtractor):
             'upload_date': '20081127',
             'timestamp': 1227796140,
         }
             'upload_date': '20081127',
             'timestamp': 1227796140,
         }

-    }
+    }, {
+        'url': 'https://cdn.jwplayer.com/players/nPripu9l-ALJ3XQCI.js',
+        'only_matching': True,
+    }]

 
     @staticmethod
     def _extract_url(webpage):
 
     @staticmethod
     def _extract_url(webpage):

-        mobj = re.search(
-            r'<script[^>]+?src=["\'](?P<url>(?:https?:)?//content.jwplatform.com/players/[a-zA-Z0-9]{8})',
+        urls = JWPlatformIE._extract_urls(webpage)
+        return urls[0] if urls else None
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<(?:script|iframe)[^>]+?src=["\']((?:https?:)?//content\.jwplatform\.com/players/[a-zA-Z0-9]{8})',

             webpage)
             webpage)

-        if mobj:
-            return mobj.group('url')

 
     def _real_extract(self, url):
         video_id = self._match_id(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]
-        subtitles = {}
-        for track in video_data['tracks']:
-            if track['kind'] == 'captions':
-                subtitles[track['label']] = [{'url': self._proto_relative_url(track['file'])}]
-
-        formats = []
-        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':
-                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,
-                    'vcodec': 'none',
-                })
-            else:
-                formats.append({
-                    'url': source_url,
-                    'width': int_or_none(source.get('width')),
-                    'height': int_or_none(source.get('height')),
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': video_data['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,
-        }
+        json_data = self._download_json('https://cdn.jwplayer.com/v2/media/' + video_id, video_id)
+        return self._parse_jwplayer_data(json_data, video_id)