X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fteamcoco.py;h=95d58ddd0429e84abeef9f6dc3dea74f011318f9;hb=322915014f0378e2675a2a17cd67fe89a6e6a7d7;hp=1dd5e1b685e7aa99804d51d99a594945f30961a6;hpb=6f5ac90cf314e8e1a7c5c416fb51e3a7b8483e9c;p=youtube-dl

diff --git a/youtube_dl/extractor/teamcoco.py b/youtube_dl/extractor/teamcoco.py
index 1dd5e1b68..95d58ddd0 100644
--- a/youtube_dl/extractor/teamcoco.py
+++ b/youtube_dl/extractor/teamcoco.py
@@ -1,55 +1,143 @@
+# -*- 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):
-    _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"
+    _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.',
+                'duration': 504,
+                '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',
+                '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)
-        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)
+
+        video_id = mobj.group('video_id')
+        if not video_id:
+            video_id = self._html_search_regex(
+                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]
 
-        self.report_extraction(video_id)
+        if not preload:
+            preload = ''.join(re.findall(r'this\.push\("([^"]+)"\);', webpage))
 
-        video_title = self._html_search_regex(r'<meta property="og:title" content="(.+?)"',
-            webpage, u'title')
+        if not preload:
+            preload = self._html_search_regex([
+                r'player,\[?"([^"]+)"\]?', r'player.init\(\[?"([^"]+)"\]?\)'
+            ], webpage.replace('","', ''), 'preload data', default=None)
 
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)"',
-            webpage, u'thumbnail', fatal=False)
+        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)
 
-        video_description = self._html_search_regex(r'<meta property="og:description" content="(.*?)"',
-            webpage, u'description', fatal=False)
+        if not preload and not data:
+            raise ExtractorError(
+                'Preload information could not be extracted', expected=True)
 
-        data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
-        data = self._download_webpage(data_url, video_id, 'Downloading data webpage')
+        if not data:
+            data = self._parse_json(
+                base64.b64decode(preload.encode('ascii')).decode('utf-8'), video_id)
 
-        video_url = self._html_search_regex(r'<file type="high".*?>(.*?)</file>',
-            data, u'video URL')
+        formats = []
+        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:
+                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)
 
-        return [{
-            'id':          video_id,
-            'url':         video_url,
-            'ext':         'mp4',
-            'title':       video_title,
-            'thumbnail':   thumbnail,
-            'description': video_description,
-        }]
+                formats.append({
+                    'url': filed['url'],
+                    'ext': 'mp4',
+                    'tbr': tbr,
+                    'format_id': format_id,
+                    'quality': get_quality(format_id),
+                })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'formats': formats,
+            'title': data['title'],
+            'thumbnail': data.get('thumb', {}).get('href'),
+            'description': data.get('teaser'),
+            'duration': data.get('duration'),
+            'age_limit': self._family_friendly_search(webpage),
+        }