Merge remote-tracking branch 'AGSPhoenix/teamcoco-fix'

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

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index 0bd0d67c4cd6377c030e3850e0a7dbf5371252ea..fdaf62cd0fe943ba962af0b02332a4ff9c63d4e2 100644 (file)
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -9,8 +9,18 @@ from ..utils import (
 
 
 class TeamcocoIE(InfoExtractor):
-    _VALID_URL = r'http://teamcoco\.com/video/(?P<url_title>.*)'
-    _TEST = {
+    _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>\d*)?/?(?P<url_title>.*)'
+    _TESTS = [
+    {
+        'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
+        'file': '80187.mp4',
+        'md5': '3f7746aa0dc86de18df7539903d399ea',
+        'info_dict': {
+            'title': 'Conan Becomes A Mary Kay Beauty Consultant',
+            'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.'
+        }
+    },
+    {
         'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
         'file': '19705.mp4',
         'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
@@ -19,6 +29,7 @@ class TeamcocoIE(InfoExtractor):
             "title": "Louis C.K. Interview Pt. 1 11/3/11"
         }
     }
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -26,11 +37,13 @@ class TeamcocoIE(InfoExtractor):
             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, 'video id')
-
+        
+        video_id = mobj.group("video_id")
+        if video_id == '':
+            video_id = self._html_search_regex(
+                r'<article class="video" data-id="(\d+?)"',
+                webpage, 'video id')
+        
         self.report_extraction(video_id)
 
         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
@@ -38,29 +51,34 @@ class TeamcocoIE(InfoExtractor):
 
         qualities = ['500k', '480p', '1000k', '720p', '1080p']
         formats = []
-        for file in data.findall('files/file'):
-            if file.attrib.get('playmode') == 'all':
+        for filed in data.findall('files/file'):
+            if filed.attrib.get('playmode') == 'all':
                 # it just duplicates one of the entries
                 break
-            file_url = file.text
+            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 = file.attrib['bitrate']
+                format_id = filed.attrib['bitrate']
+            tbr = (
+                int(filed.attrib['bitrate'])
+                if filed.attrib['bitrate'].isdigit()
+                else None)
+
+            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,
             })
-        def sort_key(f):
-            try:
-                return qualities.index(f['format_id'])
-            except ValueError:
-                return -1
-        formats.sort(key=sort_key)
-        if not formats:
-            raise ExtractorError('Unable to extract video URL')
+
+        self._sort_formats(formats)
 
         return {
             'id': video_id,