]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/breakcom.py
debian/copyright: Fix lintian's dep5-copyright-license-name-not-unique.
[youtubedl] / youtube_dl / extractor / breakcom.py
index 53a898de3707ce9a2f235d95e1d7fa0be58edb20..809287d144ca7d629bf42bad7ac4e213a323e6dd 100644 (file)
@@ -1,38 +1,63 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+from ..utils import (
+    int_or_none,
+    parse_age_limit,
+)
 
 
 class BreakIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?break\.com/video/([^/]+)'
-    _TEST = {
-        u'url': u'http://www.break.com/video/when-girls-act-like-guys-2468056',
-        u'file': u'2468056.mp4',
-        u'md5': u'a3513fb1547fba4fb6cfac1bffc6c46b',
-        u'info_dict': {
-            u"title": u"When Girls Act Like D-Bags"
+    _VALID_URL = r'http://(?:www\.)?break\.com/video/(?:[^/]+/)*.+-(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056',
+        'info_dict': {
+            'id': '2468056',
+            'ext': 'mp4',
+            'title': 'When Girls Act Like D-Bags',
         }
-    }
+    }, {
+        'url': 'http://www.break.com/video/ugc/baby-flex-2773063',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1).split("-")[-1]
-        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 = ({.*?});', webpage,
-                                       u'info json', flags=re.DOTALL)
-        info = json.loads(info_json)
-        video_url = info['videoUri']
-        m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url)
-        if m_youtube is not None:
-            return self.url_result(m_youtube.group(1), 'Youtube')
-        final_url = video_url + '?' + info['AuthToken']
-        return [{
-            'id':        video_id,
-            'url':       final_url,
-            'ext':       determine_ext(final_url),
-            'title':     info['contentName'],
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            'http://www.break.com/embed/%s' % video_id, video_id)
+        info = json.loads(self._search_regex(
+            r'var embedVars = ({.*})\s*?</script>',
+            webpage, 'info json', flags=re.DOTALL))
+
+        youtube_id = info.get('youtubeId')
+        if youtube_id:
+            return self.url_result(youtube_id, 'Youtube')
+
+        formats = [{
+            'url': media['uri'] + '?' + info['AuthToken'],
+            'tbr': media['bitRate'],
+            'width': media['width'],
+            'height': media['height'],
+        } for media in info['media'] if media.get('mediaPurpose') == 'play']
+
+        if not formats:
+            formats.append({
+                'url': info['videoUri']
+            })
+
+        self._sort_formats(formats)
+
+        duration = int_or_none(info.get('videoLengthInSeconds'))
+        age_limit = parse_age_limit(info.get('audienceRating'))
+
+        return {
+            'id': video_id,
+            'title': info['contentName'],
             'thumbnail': info['thumbUri'],
-        }]
+            'duration': duration,
+            'age_limit': age_limit,
+            'formats': formats,
+        }