]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/teamcoco.py
56be526383b590d9b00759400a269d5164beef2f
[youtubedl] / youtube_dl / extractor / teamcoco.py
1 # -*- coding: utf-8 -*-
2 from __future__ import unicode_literals
3
4 import base64
5 import binascii
6 import re
7 import json
8
9 from .common import InfoExtractor
10 from ..utils import (
11 ExtractorError,
12 qualities,
13 )
14 from ..compat import compat_ord
15
16
17 class TeamcocoIE(InfoExtractor):
18 _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
19 _TESTS = [
20 {
21 'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
22 'md5': '3f7746aa0dc86de18df7539903d399ea',
23 'info_dict': {
24 'id': '80187',
25 'ext': 'mp4',
26 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
27 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',
28 'duration': 504,
29 'age_limit': 0,
30 }
31 }, {
32 'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
33 'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
34 'info_dict': {
35 'id': '19705',
36 'ext': 'mp4',
37 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
38 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
39 'duration': 288,
40 'age_limit': 0,
41 }
42 }, {
43 'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
44 'info_dict': {
45 'id': '88748',
46 'ext': 'mp4',
47 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
48 'description': 'md5:15501f23f020e793aeca761205e42c24',
49 },
50 'params': {
51 'skip_download': True, # m3u8 downloads
52 }
53 }
54 ]
55 _VIDEO_ID_REGEXES = (
56 r'"eVar42"\s*:\s*(\d+)',
57 r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
58 r'"id_not"\s*:\s*(\d+)'
59 )
60
61 def _real_extract(self, url):
62 mobj = re.match(self._VALID_URL, url)
63
64 display_id = mobj.group('display_id')
65 webpage, urlh = self._download_webpage_handle(url, display_id)
66 if 'src=expired' in urlh.geturl():
67 raise ExtractorError('This video is expired.', expected=True)
68
69 video_id = mobj.group('video_id')
70 if not video_id:
71 video_id = self._html_search_regex(
72 self._VIDEO_ID_REGEXES, webpage, 'video id')
73
74 data = None
75
76 preload_codes = self._html_search_regex(
77 r'(function.+)setTimeout\(function\(\)\{playlist',
78 webpage, 'preload codes')
79 base64_fragments = re.findall(r'"([a-zA-z0-9+/=]+)"', preload_codes)
80 base64_fragments.remove('init')
81
82 def _check_sequence(cur_fragments):
83 if not cur_fragments:
84 return
85 for i in range(len(cur_fragments)):
86 cur_sequence = (''.join(cur_fragments[i:] + cur_fragments[:i])).encode('ascii')
87 try:
88 raw_data = base64.b64decode(cur_sequence)
89 if compat_ord(raw_data[0]) == compat_ord('{'):
90 return json.loads(raw_data.decode('utf-8'))
91 except (TypeError, binascii.Error, UnicodeDecodeError, ValueError):
92 continue
93
94 def _check_data():
95 for i in range(len(base64_fragments) + 1):
96 for j in range(i, len(base64_fragments) + 1):
97 data = _check_sequence(base64_fragments[:i] + base64_fragments[j:])
98 if data:
99 return data
100
101 self.to_screen('Try to compute possible data sequence. This may take some time.')
102 data = _check_data()
103
104 if not data:
105 raise ExtractorError(
106 'Preload information could not be extracted', expected=True)
107
108 formats = []
109 get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
110 for filed in data['files']:
111 if filed['type'] == 'hls':
112 formats.extend(self._extract_m3u8_formats(
113 filed['url'], video_id, ext='mp4'))
114 else:
115 m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
116 if m_format is not None:
117 format_id = m_format.group(1)
118 else:
119 format_id = filed['bitrate']
120 tbr = (
121 int(filed['bitrate'])
122 if filed['bitrate'].isdigit()
123 else None)
124
125 formats.append({
126 'url': filed['url'],
127 'ext': 'mp4',
128 'tbr': tbr,
129 'format_id': format_id,
130 'quality': get_quality(format_id),
131 })
132
133 self._sort_formats(formats)
134
135 return {
136 'id': video_id,
137 'display_id': display_id,
138 'formats': formats,
139 'title': data['title'],
140 'thumbnail': data.get('thumb', {}).get('href'),
141 'description': data.get('teaser'),
142 'duration': data.get('duration'),
143 'age_limit': self._family_friendly_search(webpage),
144 }