Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/teamcoco.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import binascii
   5 import re
   6 import json
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_b64decode,
  11     compat_ord,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     qualities,
  16     determine_ext,
  17 )
  18
  19
  20 class TeamcocoIE(InfoExtractor):
  21     _VALID_URL = r'https?://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
  22     _TESTS = [
  23         {
  24             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
  25             'md5': '3f7746aa0dc86de18df7539903d399ea',
  26             'info_dict': {
  27                 'id': '80187',
  28                 'ext': 'mp4',
  29                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
  30                 '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                 'duration': 504,
  32                 'age_limit': 0,
  33             }
  34         }, {
  35             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
  36             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
  37             'info_dict': {
  38                 'id': '19705',
  39                 'ext': 'mp4',
  40                 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
  41                 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
  42                 'duration': 288,
  43                 'age_limit': 0,
  44             }
  45         }, {
  46             'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
  47             'info_dict': {
  48                 'id': '88748',
  49                 'ext': 'mp4',
  50                 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
  51                 'description': 'md5:15501f23f020e793aeca761205e42c24',
  52             },
  53             'params': {
  54                 'skip_download': True,  # m3u8 downloads
  55             }
  56         }, {
  57             'url': 'http://teamcoco.com/video/full-episode-mon-6-1-joel-mchale-jake-tapper-and-musical-guest-courtney-barnett?playlist=x;eyJ0eXBlIjoidGFnIiwiaWQiOjl9',
  58             'info_dict': {
  59                 'id': '89341',
  60                 'ext': 'mp4',
  61                 'title': 'Full Episode - Mon. 6/1 - Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
  62                 'description': 'Guests: Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
  63             },
  64             'params': {
  65                 'skip_download': True,  # m3u8 downloads
  66             }
  67         }
  68     ]
  69     _VIDEO_ID_REGEXES = (
  70         r'"eVar42"\s*:\s*(\d+)',
  71         r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
  72         r'"id_not"\s*:\s*(\d+)'
  73     )
  74
  75     def _real_extract(self, url):
  76         mobj = re.match(self._VALID_URL, url)
  77
  78         display_id = mobj.group('display_id')
  79         webpage, urlh = self._download_webpage_handle(url, display_id)
  80         if 'src=expired' in urlh.geturl():
  81             raise ExtractorError('This video is expired.', expected=True)
  82
  83         video_id = mobj.group('video_id')
  84         if not video_id:
  85             video_id = self._html_search_regex(
  86                 self._VIDEO_ID_REGEXES, webpage, 'video id')
  87
  88         data = None
  89
  90         preload_codes = self._html_search_regex(
  91             r'(function.+)setTimeout\(function\(\)\{playlist',
  92             webpage, 'preload codes')
  93         base64_fragments = re.findall(r'"([a-zA-Z0-9+/=]+)"', preload_codes)
  94         base64_fragments.remove('init')
  95
  96         def _check_sequence(cur_fragments):
  97             if not cur_fragments:
  98                 return
  99             for i in range(len(cur_fragments)):
 100                 cur_sequence = (''.join(cur_fragments[i:] + cur_fragments[:i])).encode('ascii')
 101                 try:
 102                     raw_data = compat_b64decode(cur_sequence)
 103                     if compat_ord(raw_data[0]) == compat_ord('{'):
 104                         return json.loads(raw_data.decode('utf-8'))
 105                 except (TypeError, binascii.Error, UnicodeDecodeError, ValueError):
 106                     continue
 107
 108         def _check_data():
 109             for i in range(len(base64_fragments) + 1):
 110                 for j in range(i, len(base64_fragments) + 1):
 111                     data = _check_sequence(base64_fragments[:i] + base64_fragments[j:])
 112                     if data:
 113                         return data
 114
 115         self.to_screen('Try to compute possible data sequence. This may take some time.')
 116         data = _check_data()
 117
 118         if not data:
 119             raise ExtractorError(
 120                 'Preload information could not be extracted', expected=True)
 121
 122         formats = []
 123         get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
 124         for filed in data['files']:
 125             if determine_ext(filed['url']) == 'm3u8':
 126                 # compat_urllib_parse.urljoin does not work here
 127                 if filed['url'].startswith('/'):
 128                     m3u8_url = 'http://ht.cdn.turner.com/tbs/big/teamcoco' + filed['url']
 129                 else:
 130                     m3u8_url = filed['url']
 131                 m3u8_formats = self._extract_m3u8_formats(
 132                     m3u8_url, video_id, ext='mp4')
 133                 for m3u8_format in m3u8_formats:
 134                     if m3u8_format not in formats:
 135                         formats.append(m3u8_format)
 136             elif determine_ext(filed['url']) == 'f4m':
 137                 # TODO Correct f4m extraction
 138                 continue
 139             else:
 140                 if filed['url'].startswith('/mp4:protected/'):
 141                     # TODO Correct extraction for these files
 142                     continue
 143                 m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
 144                 if m_format is not None:
 145                     format_id = m_format.group(1)
 146                 else:
 147                     format_id = filed['bitrate']
 148                 tbr = (
 149                     int(filed['bitrate'])
 150                     if filed['bitrate'].isdigit()
 151                     else None)
 152
 153                 formats.append({
 154                     'url': filed['url'],
 155                     'ext': 'mp4',
 156                     'tbr': tbr,
 157                     'format_id': format_id,
 158                     'quality': get_quality(format_id),
 159                 })
 160
 161         self._sort_formats(formats)
 162
 163         return {
 164             'id': video_id,
 165             'display_id': display_id,
 166             'formats': formats,
 167             'title': data['title'],
 168             'thumbnail': data.get('thumb', {}).get('href'),
 169             'description': data.get('teaser'),
 170             'duration': data.get('duration'),
 171             'age_limit': self._family_friendly_search(webpage),
 172         }