Merge branch 'master' into rtmpdump

[youtube-dl] / youtube_dl / extractor / teamcoco.py

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index 092ac74d8ea4969ecfc047b88ffd6e5d099f8ca3..bc48620f0b992366e866181b1dad22aeb2e5d0a6 100644 (file)
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -1,4 +1,5 @@
 import re
+import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..utils import (
@@ -8,6 +9,15 @@ from ..utils import (
 
 class TeamcocoIE(InfoExtractor):
     _VALID_URL = r'http://teamcoco\.com/video/(?P<url_title>.*)'
+    _TEST = {
+        u'url': u'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
+        u'file': u'19705.mp4',
+        u'md5': u'cde9ba0fa3506f5f017ce11ead928f9a',
+        u'info_dict': {
+            u"description": u"Louis C.K. got starstruck by George W. Bush, so what? Part one.", 
+            u"title": u"Louis C.K. Interview Pt. 1 11/3/11"
+        }
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -21,26 +31,41 @@ class TeamcocoIE(InfoExtractor):
 
         self.report_extraction(video_id)
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
-            webpage, u'title')
-
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)"',
-            webpage, u'thumbnail', fatal=False)
-
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(.*?)"',
-            webpage, u'description', fatal=False)
-
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
-        data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
+        data_xml = self._download_webpage(data_url, video_id, 'Downloading data webpage')
+        data = xml.etree.ElementTree.fromstring(data_xml.encode('utf-8'))
+
 
-        video_url = self._html_search_regex(r'<file type="high".*?>(.*?)</file>',
-            data, u'video URL')
+        qualities = ['500k', '480p', '1000k', '720p', '1080p']
+        formats = []
+        for file in data.findall('files/file'):
+            if file.attrib.get('playmode') == 'all':
+                # it just duplicates one of the entries
+                break
+            file_url = file.text
+            m_format = re.search(r'(\d+(k|p))\.mp4', file_url)
+            if m_format is not None:
+                format_id = m_format.group(1)
+            else:
+                format_id = file.attrib['bitrate']
+            formats.append({
+                'url': file_url,
+                'ext': 'mp4',
+                'format_id': format_id,
+            })
+        def sort_key(f):
+            try:
+                return qualities.index(f['format_id'])
+            except ValueError:
+                return -1
+        formats.sort(key=sort_key)
+        if not formats:
+            raise RegexNotFoundError(u'Unable to extract video URL')
 
-        return [{
+        return {
             'id':          video_id,
-            'url':         video_url,
-            'ext':         'mp4',
-            'title':       video_title,
-            'thumbnail':   thumbnail,
-            'description': video_description,
-        }]
+            'formats': formats,
+            'title':       self._og_search_title(webpage),
+            'thumbnail':   self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage),
+        }