X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdotsub.py;h=e9ca236d4a03c13b1b29b3386535c4262332dab0;hb=eab7faa0c1e8511bc91c64347d0dffc28c94f101;hp=2afeaba072def6ebbad625484e184c26bc652853;hpb=16ea58cbda0b27d7b43d156092a518a55c69aac4;p=youtube-dl

diff --git a/youtube_dl/extractor/dotsub.py b/youtube_dl/extractor/dotsub.py
index 2afeaba07..e9ca236d4 100644
--- a/youtube_dl/extractor/dotsub.py
+++ b/youtube_dl/extractor/dotsub.py
@@ -1,42 +1,53 @@
-import re
-import json
+from __future__ import unicode_literals
+
 from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    int_or_none,
+)
 
 
 class DotsubIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?dotsub\.com/view/([^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?dotsub\.com/view/(?P<id>[^/]+)'
     _TEST = {
-        u'url': u'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
-        u'file': u'aed3b8b2-1889-4df5-ae63-ad85f5572f27.flv',
-        u'md5': u'0914d4d69605090f623b7ac329fea66e',
-        u'info_dict': {
-            u"title": u"Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary",
-            u"uploader": u"4v4l0n42",
-            u'description': u'Pyramids of Waste (2010) also known as "The lightbulb conspiracy" is a documentary about how our economic system based on consumerism  and planned obsolescence is breaking our planet down.\r\n\r\nSolutions to this can be found at:\r\nhttp://robotswillstealyourjob.com\r\nhttp://www.federicopistono.org\r\n\r\nhttp://opensourceecology.org\r\nhttp://thezeitgeistmovement.com',
-            u'thumbnail': u'http://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p'
+        'url': 'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+        'md5': '0914d4d69605090f623b7ac329fea66e',
+        'info_dict': {
+            'id': 'aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+            'ext': 'flv',
+            'title': 'Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary',
+            'description': 'md5:699a0f7f50aeec6042cb3b1db2d0d074',
+            'thumbnail': 're:^https?://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p',
+            'duration': 3169,
+            'uploader': '4v4l0n42',
+            'timestamp': 1292248482.625,
+            'upload_date': '20101213',
+            'view_count': int,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
-        info_url = "https://dotsub.com/api/media/%s/metadata" %(video_id)
-        webpage = self._download_webpage(info_url, video_id)
-        info = json.loads(webpage)
-        video_url = info['mediaURI']
-        uploader = info['user']
-        description = info['description']
-        view_count = info['numberOfViews']
-        title = info['title']
-        thumbnail_url = info['screenshotURI']
-        ext = 'flv'
-        return [{
-            'id':          video_id,
-            'url':         video_url,
-            'ext':         ext,
-            'title':       title,
-            'thumbnail':   thumbnail_url,
-            'description': description,
-            'uploader':    uploader,
-            'view_count':  view_count,
-        }]
+        video_id = self._match_id(url)
+
+        info = self._download_json(
+            'https://dotsub.com/api/media/%s/metadata' % video_id, video_id)
+        video_url = info.get('mediaURI')
+
+        if not video_url:
+            webpage = self._download_webpage(url, video_id)
+            video_url = self._search_regex(
+                [r'<source[^>]+src="([^"]+)"', r'"file"\s*:\s*\'([^\']+)'],
+                webpage, 'video url')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'flv',
+            'title': info['title'],
+            'description': info.get('description'),
+            'thumbnail': info.get('screenshotURI'),
+            'duration': int_or_none(info.get('duration'), 1000),
+            'uploader': info.get('user'),
+            'timestamp': float_or_none(info.get('dateCreated'), 1000),
+            'view_count': int_or_none(info.get('numberOfViews')),
+        }