X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/90cd207d52d77e075e62cc256cc39dceee05bbf3..0865c28fb29a6481cd837cf8c1ef0cd134c6ef8e:/youtube_dl/extractor/teamcoco.py

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index bc48620..5793dbc 100644
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -1,71 +1,92 @@
+from __future__ import unicode_literals
+
+import base64
 import re
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-)
+from ..utils import qualities
 
 
 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"
+    _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
+    _TESTS = [
+        {
+            'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
+            'md5': '3f7746aa0dc86de18df7539903d399ea',
+            'info_dict': {
+                'id': '80187',
+                'ext': 'mp4',
+                '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.',
+                'age_limit': 0,
+            }
+        }, {
+            'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
+            'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
+            'info_dict': {
+                'id': '19705',
+                'ext': 'mp4',
+                'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
+                'title': 'Louis C.K. Interview Pt. 1 11/3/11',
+                'age_limit': 0,
+            }
         }
-    }
+    ]
+    _VIDEO_ID_REGEXES = (
+        r'"eVar42"\s*:\s*(\d+)',
+        r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
+        r'"id_not"\s*:\s*(\d+)'
+    )
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'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')
+        display_id = mobj.group('display_id')
+        webpage = self._download_webpage(url, display_id)
 
-        self.report_extraction(video_id)
+        video_id = mobj.group('video_id')
+        if not video_id:
+            video_id = self._html_search_regex(
+                self._VIDEO_ID_REGEXES, webpage, 'video id')
 
-        data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
-        data_xml = self._download_webpage(data_url, video_id, 'Downloading data webpage')
-        data = xml.etree.ElementTree.fromstring(data_xml.encode('utf-8'))
+        embed_url = 'http://teamcoco.com/embed/v/%s' % video_id
+        embed = self._download_webpage(
+            embed_url, video_id, 'Downloading embed page')
 
+        encoded_data = self._search_regex(
+            r'"preload"\s*:\s*"([^"]+)"', embed, 'encoded data')
+        data = self._parse_json(
+            base64.b64decode(encoded_data.encode('ascii')).decode('utf-8'), video_id)
 
-        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)
+        get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
+        for filed in data['files']:
+            m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
             if m_format is not None:
                 format_id = m_format.group(1)
             else:
-                format_id = file.attrib['bitrate']
+                format_id = filed['bitrate']
+            tbr = (
+                int(filed['bitrate'])
+                if filed['bitrate'].isdigit()
+                else None)
+
             formats.append({
-                'url': file_url,
+                'url': filed['url'],
                 'ext': 'mp4',
+                'tbr': tbr,
                 'format_id': format_id,
+                'quality': get_quality(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')
+
+        self._sort_formats(formats)
 
         return {
-            'id':          video_id,
+            'id': video_id,
+            'display_id': display_id,
             'formats': formats,
-            'title':       self._og_search_title(webpage),
-            'thumbnail':   self._og_search_thumbnail(webpage),
-            'description': self._og_search_description(webpage),
+            'title': data['title'],
+            'thumbnail': data.get('thumb', {}).get('href'),
+            'description': data.get('teaser'),
+            'age_limit': self._family_friendly_search(webpage),
         }