debian/control: Update list of supported sites/extractors.

[youtubedl] / youtube_dl / extractor / dailymail.py

diff --git a/youtube_dl/extractor/dailymail.py b/youtube_dl/extractor/dailymail.py
index b60a1d813fa2c48c81e44306402b303922b65e88..af39780353ac1e044ffb07464dfa64032ba18f0e 100644 (file)
--- a/youtube_dl/extractor/dailymail.py
+++ b/youtube_dl/extractor/dailymail.py
@@ -1,34 +1,54 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

+import re
+

 from .common import InfoExtractor
 from .common import InfoExtractor

+from ..compat import compat_str

 from ..utils import (
     int_or_none,
     determine_protocol,
 from ..utils import (
     int_or_none,
     determine_protocol,

+    try_get,
+    unescapeHTML,

 )
 
 
 class DailyMailIE(InfoExtractor):
 )
 
 
 class DailyMailIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
-    _TEST = {
-        'url': 'http://www.dailymail.co.uk/video/sciencetech/video-1288527/Turn-video-impressionist-masterpiece.html',
-        'md5': '2f639d446394f53f3a33658b518b6615',
+    _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/(?:video/[^/]+/video-|embed/video/)(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
+        'md5': 'f6129624562251f628296c3a9ffde124',

         'info_dict': {
         'info_dict': {

-            'id': '1288527',
+            'id': '1295863',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'Turn any video into an impressionist masterpiece',
-            'description': 'md5:88ddbcb504367987b2708bb38677c9d2',
+            'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
+            'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',

         }
         }

-    }
+    }, {
+        'url': 'http://www.dailymail.co.uk/embed/video/1295863.html',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def _extract_urls(webpage):
+        return re.findall(
+            r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?dailymail\.co\.uk/embed/video/\d+\.html)',
+            webpage)

 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         video_data = self._parse_json(self._search_regex(
             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         video_data = self._parse_json(self._search_regex(
             r"data-opts='({.+?})'", webpage, 'video data'), video_id)

-        title = video_data['title']
-        video_sources = self._download_json(video_data.get(
-            'sources', {}).get('url') or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id, video_id)
+        title = unescapeHTML(video_data['title'])
+
+        sources_url = (try_get(
+            video_data,
+            (lambda x: x['plugins']['sources']['url'],
+             lambda x: x['sources']['url']), compat_str) or
+            'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id)
+
+        video_sources = self._download_json(sources_url, video_id)

 
         formats = []
         for rendition in video_sources['renditions']:
 
         formats = []
         for rendition in video_sources['renditions']:


@@ -55,7 +75,7 @@ class DailyMailIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
         return {
             'id': video_id,
             'title': title,

-            'description': video_data.get('descr'),
+            'description': unescapeHTML(video_data.get('descr')),

             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
             'formats': formats,
         }
             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
             'formats': formats,
         }