X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdiscovery.py;h=d3e6675283cddcb8f6a6dfffbfbd1e1ea3da11bc;hb=785521bf4fbd99b2916bdab5d847d84424196c1d;hp=14fca3cae00194967f44b27117ded4e7db1a3c71;hpb=96d7b8873ad47c1f52193e84fc6f8cfe12891aa7;p=youtube-dl

diff --git a/youtube_dl/extractor/discovery.py b/youtube_dl/extractor/discovery.py
index 14fca3cae..d3e667528 100644
--- a/youtube_dl/extractor/discovery.py
+++ b/youtube_dl/extractor/discovery.py
@@ -1,40 +1,45 @@
 from __future__ import unicode_literals
 
-import re
-import json
 from .common import InfoExtractor
+from ..utils import (
+    parse_iso8601,
+    int_or_none,
+)
 
 
 class DiscoveryIE(InfoExtractor):
-    _VALID_URL = r'http://dsc\.discovery\.com\/[a-zA-Z0-9\-]*/[a-zA-Z0-9\-]*/videos/(?P<id>[a-zA-Z0-9\-]*)(.htm)?'
+    _VALID_URL = r'http://www\.discovery\.com\/[a-zA-Z0-9\-]*/[a-zA-Z0-9\-]*/videos/(?P<id>[a-zA-Z0-9_\-]*)(?:\.htm)?'
     _TEST = {
-        'url': 'http://dsc.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
-        'file': 'mission-impossible-outtakes.mp4',
-        'md5': 'e12614f9ee303a6ccef415cb0793eba2',
+        'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
+        'md5': '3c69d77d9b0d82bfd5e5932a60f26504',
         'info_dict': {
-            'title': 'MythBusters: Mission Impossible Outtakes'
-        }
+            'id': 'mission-impossible-outtakes',
+            'ext': 'flv',
+            'title': 'Mission Impossible Outtakes',
+            'description': ('Watch Jamie Hyneman and Adam Savage practice being'
+                            ' each other -- to the point of confusing Jamie\'s dog -- and '
+                            'don\'t miss Adam moon-walking as Jamie ... behind Jamie\'s'
+                            ' back.'),
+            'duration': 156,
+            'timestamp': 1303099200,
+            'upload_date': '20110418',
+        },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-        title = self._search_regex(
-            r'(?<=\"name\": ")(?P<title>.*?)(?=\"\,)', webpage, r'Filename')
-        duration = int(self._search_regex(
-            r'(?<=\"duration\"\: )(?P<duration>.*?)(?=,)', webpage, r'Duration'))
-        formats_raw = self._search_regex(
-            r'(?<=\"mp4\":)(.*?)(}])', webpage, r'formats') + '}]'
-        formats_json = json.loads(formats_raw)
-        formats = []
-        for f in formats_json:
-            formats.append(
-                {'url': f['src'], r'ext': r'mp4', 'tbr': int(f['bitrate'][:-1])})
+
+        info = self._parse_json(self._search_regex(
+            r'(?s)<script type="application/ld\+json">(.*?)</script>',
+            webpage, 'video info'), video_id)
 
         return {
             'id': video_id,
-            'duration': duration,
-            'title': title,
-            'formats': formats
+            'title': info['name'],
+            'url': info['contentURL'],
+            'description': info.get('description'),
+            'thumbnail': info.get('thumbnailUrl'),
+            'timestamp': parse_iso8601(info.get('uploadDate')),
+            'duration': int_or_none(info.get('duration')),
         }