1 # -*- coding: utf-8 -*-
2 from __future__ import unicode_literals
8 from .common import InfoExtractor
13 from ..compat import compat_ord
16 class TeamcocoIE(InfoExtractor):
17 _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
20 'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
21 'md5': '3f7746aa0dc86de18df7539903d399ea',
25 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
26 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',
31 'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
32 'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
36 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
37 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
42 'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
46 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
47 'description': 'md5:15501f23f020e793aeca761205e42c24',
50 'skip_download': True, # m3u8 downloads
55 r'"eVar42"\s*:\s*(\d+)',
56 r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
57 r'"id_not"\s*:\s*(\d+)'
60 def _real_extract(self, url):
61 mobj = re.match(self._VALID_URL, url)
63 display_id = mobj.group('display_id')
64 webpage = self._download_webpage(url, display_id)
66 video_id = mobj.group('video_id')
68 video_id = self._html_search_regex(
69 self._VIDEO_ID_REGEXES, webpage, 'video id')
72 preloads = re.findall(r'"preload":\s*"([^"]+)"', webpage)
74 preload = max([(len(p), p) for p in preloads])[1]
77 preload = ''.join(re.findall(r'this\.push\("([^"]+)"\);', webpage))
80 preload = self._html_search_regex([
81 r'player,\[?"([^"]+)"\]?', r'player.init\(\[?"([^"]+)"\]?\)'
82 ], webpage.replace('","', ''), 'preload data', default=None)
85 preload_codes = self._html_search_regex(
86 r'(function.+)setTimeout\(function\(\)\{playlist',
87 webpage, 'preload codes')
88 base64_fragments = re.findall(r'"([a-zA-z0-9+/=]+)"', preload_codes)
89 base64_fragments.remove('init')
90 for i in range(len(base64_fragments)):
91 cur_sequence = (''.join(base64_fragments[i:] + base64_fragments[:i])).encode('ascii')
93 raw_data = base64.b64decode(cur_sequence)
94 except (TypeError, binascii.Error):
96 if compat_ord(raw_data[0]) == compat_ord('{'):
97 data = self._parse_json(raw_data.decode('utf-8'), video_id, fatal=False)
99 if not preload and not data:
100 raise ExtractorError(
101 'Preload information could not be extracted', expected=True)
104 data = self._parse_json(
105 base64.b64decode(preload.encode('ascii')).decode('utf-8'), video_id)
108 get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
109 for filed in data['files']:
110 if filed['type'] == 'hls':
111 formats.extend(self._extract_m3u8_formats(
112 filed['url'], video_id, ext='mp4'))
114 m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
115 if m_format is not None:
116 format_id = m_format.group(1)
118 format_id = filed['bitrate']
120 int(filed['bitrate'])
121 if filed['bitrate'].isdigit()
128 'format_id': format_id,
129 'quality': get_quality(format_id),
132 self._sort_formats(formats)
136 'display_id': display_id,
138 'title': data['title'],
139 'thumbnail': data.get('thumb', {}).get('href'),
140 'description': data.get('teaser'),
141 'duration': data.get('duration'),
142 'age_limit': self._family_friendly_search(webpage),