_ Git - youtube-dl/blob - youtube_dl/extractor/teamcoco.py

   1 from __future__ import unicode_literals
   2
   3 import base64
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     qualities,
  10 )
  11
  12
  13 class TeamcocoIE(InfoExtractor):
  14     _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
  15     _TESTS = [
  16         {
  17             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
  18             'md5': '3f7746aa0dc86de18df7539903d399ea',
  19             'info_dict': {
  20                 'id': '80187',
  21                 'ext': 'mp4',
  22                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
  23                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.',
  24                 'duration': 504,
  25                 'age_limit': 0,
  26             }
  27         }, {
  28             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
  29             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
  30             'info_dict': {
  31                 'id': '19705',
  32                 'ext': 'mp4',
  33                 'description': 'Louis C.K. got starstruck by George W. Bush, so what? Part one.',
  34                 'title': 'Louis C.K. Interview Pt. 1 11/3/11',
  35                 'duration': 288,
  36                 'age_limit': 0,
  37             }
  38         }
  39     ]
  40     _VIDEO_ID_REGEXES = (
  41         r'"eVar42"\s*:\s*(\d+)',
  42         r'Ginger\.TeamCoco\.openInApp\("video",\s*"([^"]+)"',
  43         r'"id_not"\s*:\s*(\d+)'
  44     )
  45
  46     def _real_extract(self, url):
  47         mobj = re.match(self._VALID_URL, url)
  48
  49         display_id = mobj.group('display_id')
  50         webpage = self._download_webpage(url, display_id)
  51
  52         video_id = mobj.group('video_id')
  53         if not video_id:
  54             video_id = self._html_search_regex(
  55                 self._VIDEO_ID_REGEXES, webpage, 'video id')
  56
  57         preloads = re.findall(r'"preload":\s*"([^"]+)"', webpage)
  58         if not preloads:
  59             raise ExtractorError('Preload information could not be extracted')
  60         preload = max([(len(p), p) for p in preloads])[1]
  61         data = self._parse_json(
  62             base64.b64decode(preload.encode('ascii')).decode('utf-8'), video_id)
  63
  64         formats = []
  65         get_quality = qualities(['500k', '480p', '1000k', '720p', '1080p'])
  66         for filed in data['files']:
  67             if filed['type'] == 'hls':
  68                 formats.extend(self._extract_m3u8_formats(
  69                     filed['url'], video_id, ext='mp4'))
  70             else:
  71                 m_format = re.search(r'(\d+(k|p))\.mp4', filed['url'])
  72                 if m_format is not None:
  73                     format_id = m_format.group(1)
  74                 else:
  75                     format_id = filed['bitrate']
  76                 tbr = (
  77                     int(filed['bitrate'])
  78                     if filed['bitrate'].isdigit()
  79                     else None)
  80
  81                 formats.append({
  82                     'url': filed['url'],
  83                     'ext': 'mp4',
  84                     'tbr': tbr,
  85                     'format_id': format_id,
  86                     'quality': get_quality(format_id),
  87                 })
  88
  89         self._sort_formats(formats)
  90
  91         return {
  92             'id': video_id,
  93             'display_id': display_id,
  94             'formats': formats,
  95             'title': data['title'],
  96             'thumbnail': data.get('thumb', {}).get('href'),
  97             'description': data.get('teaser'),
  98             'duration': data.get('duration'),
  99             'age_limit': self._family_friendly_search(webpage),
 100         }