Merge tag 'upstream/2014.07.11'

[youtubedl] / youtube_dl / extractor / breakcom.py

diff --git a/youtube_dl/extractor/breakcom.py b/youtube_dl/extractor/breakcom.py
index 1f6620d9158d25323e9aeb1c5f172eb74a26a806..1bfc9f35bbd5c7c929c8f21a20f7b9642d00bcb2 100644 (file)
--- a/youtube_dl/extractor/breakcom.py
+++ b/youtube_dl/extractor/breakcom.py
@@ -1,25 +1,40 @@
+from __future__ import unicode_literals
+

 import re
 import re

+import json

 
 from .common import InfoExtractor
 
 
 class BreakIE(InfoExtractor):
 
 from .common import InfoExtractor
 
 
 class BreakIE(InfoExtractor):

-    _VALID_URL = r'(?:http://)?(?:www\.)?break\.com/video/([^/]+)'
+    _VALID_URL = r'http://(?:www\.)?break\.com/video/([^/]+)'
+    _TEST = {
+        'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056',
+        'md5': 'a3513fb1547fba4fb6cfac1bffc6c46b',
+        'info_dict': {
+            'id': '2468056',
+            'ext': 'mp4',
+            'title': 'When Girls Act Like D-Bags',
+        }
+    }

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1).split("-")[-1]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group(1).split("-")[-1]

-        webpage = self._download_webpage(url, video_id)
-        video_url = re.search(r"videoPath: '(.+?)',",webpage).group(1)
-        key = re.search(r"icon: '(.+?)',",webpage).group(1)
-        final_url = str(video_url)+"?"+str(key)
-        thumbnail_url = re.search(r"thumbnailURL: '(.+?)'",webpage).group(1)
-        title = re.search(r"sVidTitle: '(.+)',",webpage).group(1)
-        ext = video_url.split('.')[-1]
-        return [{
-            'id':        video_id,
-            'url':       final_url,
-            'ext':       ext,
-            'title':     title,
-            'thumbnail': thumbnail_url,
-        }]
+        embed_url = 'http://www.break.com/embed/%s' % video_id
+        webpage = self._download_webpage(embed_url, video_id)
+        info_json = self._search_regex(r'var embedVars = ({.*})\s*?</script>',
+            webpage, 'info json', flags=re.DOTALL)
+        info = json.loads(info_json)
+        video_url = info['videoUri']
+        youtube_id = info.get('youtubeId')
+        if youtube_id:
+            return self.url_result(youtube_id, 'Youtube')
+
+        final_url = video_url + '?' + info['AuthToken']
+        return {
+            'id': video_id,
+            'url': final_url,
+            'title': info['contentName'],
+            'thumbnail': info['thumbUri'],
+        }