]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/teamcoco.py
Update changelog.
[youtubedl] / youtube_dl / extractor / teamcoco.py
index 1dd5e1b685e7aa99804d51d99a594945f30961a6..9dcffead04d5466c14c6f2ff60995ecfb5435e6d 100644 (file)
@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -9,47 +11,66 @@ 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'27b6f7527da5acf534b15f21b032656e',
-        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"
+        'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
+        'file': '19705.mp4',
+        'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
+        'info_dict': {
+            "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one.",
+            "title": "Louis C.K. Interview Pt. 1 11/3/11"
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)
         url_title = mobj.group('url_title')
         webpage = self._download_webpage(url, url_title)
 
-        video_id = self._html_search_regex(r'<article class="video" data-id="(\d+?)"',
-            webpage, u'video id')
+        video_id = self._html_search_regex(
+            r'<article class="video" data-id="(\d+?)"',
+            webpage, 'video id')
 
         self.report_extraction(video_id)
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
-            webpage, u'title')
+        data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
+        data = self._download_xml(data_url, video_id, 'Downloading data webpage')
 
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)"',
-            webpage, u'thumbnail', fatal=False)
+        qualities = ['500k', '480p', '1000k', '720p', '1080p']
+        formats = []
+        for filed in data.findall('files/file'):
+            if filed.attrib.get('playmode') == 'all':
+                # it just duplicates one of the entries
+                break
+            file_url = filed.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 = filed.attrib['bitrate']
+            tbr = (
+                int(filed.attrib['bitrate'])
+                if filed.attrib['bitrate'].isdigit()
+                else None)
 
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(.*?)"',
-            webpage, u'description', fatal=False)
+            try:
+                quality = qualities.index(format_id)
+            except ValueError:
+                quality = -1
+            formats.append({
+                'url': file_url,
+                'ext': 'mp4',
+                'tbr': tbr,
+                'format_id': format_id,
+                'quality': quality,
+            })
 
-        data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
-        data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
-
-        video_url = self._html_search_regex(r'<file type="high".*?>(.*?)</file>',
-            data, u'video URL')
-
-        return [{
-            'id':          video_id,
-            'url':         video_url,
-            'ext':         'mp4',
-            'title':       video_title,
-            'thumbnail':   thumbnail,
-            'description': video_description,
-        }]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'title': self._og_search_title(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': self._og_search_description(webpage),
+        }