X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fteamcoco.py;h=95d58ddd0429e84abeef9f6dc3dea74f011318f9;hb=322915014f0378e2675a2a17cd67fe89a6e6a7d7;hp=e85d452a31277af6a298296bbe07958fec8aac86;hpb=641eb10d34629c177445dd38e9d3f1f73c4cb92f;p=youtube-dl

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index e85d452a3..95d58ddd0 100644
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -1,8 +1,16 @@
+# -*- coding: utf-8 -*-
 from __future__ import unicode_literals
 
+import base64
+import binascii
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    qualities,
+)
+from ..compat import compat_ord
 
 
 class TeamcocoIE(InfoExtractor):
@@ -16,6 +24,7 @@ class TeamcocoIE(InfoExtractor):
                 '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.',
+                'duration': 504,
                 'age_limit': 0,
             }
         }, {
@@ -24,12 +33,29 @@ 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',
+                'duration': 288,
                 'age_limit': 0,
             }
+        }, {
+            'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
+            'info_dict': {
+                'id': '88748',
+                'ext': 'mp4',
+                'title': 'Timothy Olyphant Raises A Toast To “Justified”',
+                'description': 'md5:15501f23f020e793aeca761205e42c24',
+            },
+            'params': {
+                'skip_download': True,  # m3u8 downloads
+            }
         }
     ]
+    _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)
@@ -37,44 +63,71 @@ 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(
-                r'<div\s+class="player".*?data-id="(\d+?)"',
-                webpage, 'video id')
+                self._VIDEO_ID_REGEXES, webpage, 'video id')
+
+        data = preload = None
+        preloads = re.findall(r'"preload":\s*"([^"]+)"', webpage)
+        if preloads:
+            preload = max([(len(p), p) for p in preloads])[1]
+
+        if not preload:
+            preload = ''.join(re.findall(r'this\.push\("([^"]+)"\);', webpage))
 
-        data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
-        data = self._download_xml(
-            data_url, display_id, 'Downloading data webpage')
+        if not preload:
+            preload = self._html_search_regex([
+                r'player,\[?"([^"]+)"\]?', r'player.init\(\[?"([^"]+)"\]?\)'
+            ], webpage.replace('","', ''), 'preload data', default=None)
+
+        if not preload:
+            preload_codes = self._html_search_regex(
+                r'(function.+)setTimeout\(function\(\)\{playlist',
+                webpage, 'preload codes')
+            base64_fragments = re.findall(r'"([a-zA-z0-9+/=]+)"', preload_codes)
+            base64_fragments.remove('init')
+            for i in range(len(base64_fragments)):
+                cur_sequence = (''.join(base64_fragments[i:] + base64_fragments[:i])).encode('ascii')
+                try:
+                    raw_data = base64.b64decode(cur_sequence)
+                except (TypeError, binascii.Error):
+                    continue
+                if compat_ord(raw_data[0]) == compat_ord('{'):
+                    data = self._parse_json(raw_data.decode('utf-8'), video_id, fatal=False)
+
+        if not preload and not data:
+            raise ExtractorError(
+                'Preload information could not be extracted', expected=True)
+
+        if not data:
+            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)
-
-            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,
-            })
+                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)
+
+                formats.append({
+                    'url': filed['url'],
+                    'ext': 'mp4',
+                    'tbr': tbr,
+                    'format_id': format_id,
+                    'quality': get_quality(format_id),
+                })
 
         self._sort_formats(formats)
 
@@ -82,8 +135,9 @@ 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'),
+            'duration': data.get('duration'),
             'age_limit': self._family_friendly_search(webpage),
         }