[teamcoco] Extract m3u8 URLs

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

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index a73da1c9c0d92657bd90f302b03e9fa8404c2dcf..41677503c7a3f5d1530d4a4d8b8e2fc8631e28a2 100644 (file)
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -1,8 +1,13 @@
 from __future__ import unicode_literals
 
+import base64
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    qualities,
+)
 
 
 class TeamcocoIE(InfoExtractor):
@@ -24,8 +29,8 @@ class TeamcocoIE(InfoExtractor):
             '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",
+                '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,
             }
         }
@@ -42,43 +47,42 @@ class TeamcocoIE(InfoExtractor):
         display_id = mobj.group('display_id')
         webpage = self._download_webpage(url, display_id)
 
-        video_id = mobj.group("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 = self._download_xml(
-            data_url, display_id, 'Downloading data webpage')
+        preloads = re.findall(r'"preload":\s*"([^"]+)"', webpage)
+        if not preloads:
+            raise ExtractorError('Preload information could not be extracted')
+        preload = max([(len(p), p) for p in preloads])[1]
+        data = self._parse_json(
+            base64.b64decode(preload.encode('ascii')).decode('utf-8'), video_id)
 
-        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)
+        get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
+        for filed in data['files']:
+            if filed['type'] == 'hls':
+                formats.extend(self._extract_m3u8_formats(
+                    filed['url'], video_id, ext='mp4'))
             else:
-                format_id = filed.attrib['bitrate']
-            tbr = (
-                int(filed.attrib['bitrate'])
-                if filed.attrib['bitrate'].isdigit()
-                else None)
+                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 = filed['bitrate']
+                tbr = (
+                    int(filed['bitrate'])
+                    if filed['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,
-            })
+                formats.append({
+                    'url': filed['url'],
+                    'ext': 'mp4',
+                    'tbr': tbr,
+                    'format_id': format_id,
+                    'quality': get_quality(format_id),
+                })
 
         self._sort_formats(formats)
 
@@ -86,8 +90,8 @@ class TeamcocoIE(InfoExtractor):
             '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),
         }