Raphaël G. Git Repositories - youtubedl/blob - 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     determine_ext,
  14 )
  15 from ..compat import compat_ord
  16
  17
  18 class TeamcocoIE(InfoExtractor):
  19     _VALID_URL = r'https?://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
  20     _TESTS = [
  21         {
  22             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
  23             'md5': '3f7746aa0dc86de18df7539903d399ea',
  24             'info_dict': {
  25                 'id': '80187',
  26                 'ext': 'mp4',
  27                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
  28                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',
  29                 'duration': 504,
  30                 'age_limit': 0,
  31             }
  32         }, {
  33             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
  34             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
  35             'info_dict': {
  36                 'id': '19705',
  37                 'ext': 'mp4',
  38                 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
  39                 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
  40                 'duration': 288,
  41                 'age_limit': 0,
  42             }
  43         }, {
  44             'url': 'http://teamcoco.com/video/timothy-olyphant-drinking-whiskey',
  45             'info_dict': {
  46                 'id': '88748',
  47                 'ext': 'mp4',
  48                 'title': 'Timothy Olyphant Raises A Toast To “Justified”',
  49                 'description': 'md5:15501f23f020e793aeca761205e42c24',
  50             },
  51             'params': {
  52                 'skip_download': True,  # m3u8 downloads
  53             }
  54         }, {
  55             'url': 'http://teamcoco.com/video/full-episode-mon-6-1-joel-mchale-jake-tapper-and-musical-guest-courtney-barnett?playlist=x;eyJ0eXBlIjoidGFnIiwiaWQiOjl9',
  56             'info_dict': {
  57                 'id': '89341',
  58                 'ext': 'mp4',
  59                 'title': 'Full Episode - Mon. 6/1 - Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
  60                 'description': 'Guests: Joel McHale, Jake Tapper, And Musical Guest Courtney Barnett',
  61             },
  62             'params': {
  63                 'skip_download': True,  # m3u8 downloads
  64             }
  65         }
  66     ]
  67     _VIDEO_ID_REGEXES = (
  68         r'"eVar42"\s*:\s*(\d+)',
  69         r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
  70         r'"id_not"\s*:\s*(\d+)'
  71     )
  72
  73     def _real_extract(self, url):
  74         mobj = re.match(self._VALID_URL, url)
  75
  76         display_id = mobj.group('display_id')
  77         webpage, urlh = self._download_webpage_handle(url, display_id)
  78         if 'src=expired' in urlh.geturl():
  79             raise ExtractorError('This video is expired.', expected=True)
  80
  81         video_id = mobj.group('video_id')
  82         if not video_id:
  83             video_id = self._html_search_regex(
  84                 self._VIDEO_ID_REGEXES, webpage, 'video id')
  85
  86         data = None
  87
  88         preload_codes = self._html_search_regex(
  89             r'(function.+)setTimeout\(function\(\)\{playlist',
  90             webpage, 'preload codes')
  91         base64_fragments = re.findall(r'"([a-zA-Z0-9+/=]+)"', preload_codes)
  92         base64_fragments.remove('init')
  93
  94         def _check_sequence(cur_fragments):
  95             if not cur_fragments:
  96                 return
  97             for i in range(len(cur_fragments)):
  98                 cur_sequence = (''.join(cur_fragments[i:] + cur_fragments[:i])).encode('ascii')
  99                 try:
 100                     raw_data = base64.b64decode(cur_sequence)
 101                     if compat_ord(raw_data[0]) == compat_ord('{'):
 102                         return json.loads(raw_data.decode('utf-8'))
 103                 except (TypeError, binascii.Error, UnicodeDecodeError, ValueError):
 104                     continue
 105
 106         def _check_data():
 107             for i in range(len(base64_fragments) + 1):
 108                 for j in range(i, len(base64_fragments) + 1):
 109                     data = _check_sequence(base64_fragments[:i] + base64_fragments[j:])
 110                     if data:
 111                         return data
 112
 113         self.to_screen('Try to compute possible data sequence. This may take some time.')
 114         data = _check_data()
 115
 116         if not data:
 117             raise ExtractorError(
 118                 'Preload information could not be extracted', expected=True)
 119
 120         formats = []
 121         get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
 122         for filed in data['files']:
 123             if determine_ext(filed['url']) == 'm3u8':
 124                 # compat_urllib_parse.urljoin does not work here
 125                 if filed['url'].startswith('/'):
 126                     m3u8_url = 'http://ht.cdn.turner.com/tbs/big/teamcoco' + filed['url']
 127                 else:
 128                     m3u8_url = filed['url']
 129                 m3u8_formats = self._extract_m3u8_formats(
 130                     m3u8_url, video_id, ext='mp4')
 131                 for m3u8_format in m3u8_formats:
 132                     if m3u8_format not in formats:
 133                         formats.append(m3u8_format)
 134             elif determine_ext(filed['url']) == 'f4m':
 135                 # TODO Correct f4m extraction
 136                 continue
 137             else:
 138                 if filed['url'].startswith('/mp4:protected/'):
 139                     # TODO Correct extraction for these files
 140                     continue
 141                 m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
 142                 if m_format is not None:
 143                     format_id = m_format.group(1)
 144                 else:
 145                     format_id = filed['bitrate']
 146                 tbr = (
 147                     int(filed['bitrate'])
 148                     if filed['bitrate'].isdigit()
 149                     else None)
 150
 151                 formats.append({
 152                     'url': filed['url'],
 153                     'ext': 'mp4',
 154                     'tbr': tbr,
 155                     'format_id': format_id,
 156                     'quality': get_quality(format_id),
 157                 })
 158
 159         self._sort_formats(formats)
 160
 161         return {
 162             'id': video_id,
 163             'display_id': display_id,
 164             'formats': formats,
 165             'title': data['title'],
 166             'thumbnail': data.get('thumb', {}).get('href'),
 167             'description': data.get('teaser'),
 168             'duration': data.get('duration'),
 169             'age_limit': self._family_friendly_search(webpage),
 170         }