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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6
   7
   8 class TeamcocoIE(InfoExtractor):
   9     _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>[0-9]+)?/?(?P<display_id>.*)'
  10     _TESTS = [
  11         {
  12             'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
  13             'md5': '3f7746aa0dc86de18df7539903d399ea',
  14             'info_dict': {
  15                 'id': '80187',
  16                 'ext': 'mp4',
  17                 'title': 'Conan Becomes A Mary Kay Beauty Consultant',
  18                 'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.'
  19             }
  20         }, {
  21             'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
  22             'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
  23             'info_dict': {
  24                 'id': '19705',
  25                 'ext': 'mp4',
  26                 "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one.",
  27                 "title": "Louis C.K. Interview Pt. 1 11/3/11"
  28             }
  29         }
  30     ]
  31
  32     def _real_extract(self, url):
  33         mobj = re.match(self._VALID_URL, url)
  34
  35         display_id = mobj.group('display_id')
  36         webpage = self._download_webpage(url, display_id)
  37
  38         video_id = mobj.group("video_id")
  39         if not video_id:
  40             video_id = self._html_search_regex(
  41                 r'<div\s+class="player".*?data-id="(\d+?)"',
  42                 webpage, 'video id')
  43
  44         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
  45         data = self._download_xml(
  46             data_url, display_id, 'Downloading data webpage')
  47
  48         qualities = ['500k', '480p', '1000k', '720p', '1080p']
  49         formats = []
  50         for filed in data.findall('files/file'):
  51             if filed.attrib.get('playmode') == 'all':
  52                 # it just duplicates one of the entries
  53                 break
  54             file_url = filed.text
  55             m_format = re.search(r'(\d+(k|p))\.mp4', file_url)
  56             if m_format is not None:
  57                 format_id = m_format.group(1)
  58             else:
  59                 format_id = filed.attrib['bitrate']
  60             tbr = (
  61                 int(filed.attrib['bitrate'])
  62                 if filed.attrib['bitrate'].isdigit()
  63                 else None)
  64
  65             try:
  66                 quality = qualities.index(format_id)
  67             except ValueError:
  68                 quality = -1
  69             formats.append({
  70                 'url': file_url,
  71                 'ext': 'mp4',
  72                 'tbr': tbr,
  73                 'format_id': format_id,
  74                 'quality': quality,
  75             })
  76
  77         self._sort_formats(formats)
  78
  79         return {
  80             'id': video_id,
  81             'display_id': display_id,
  82             'formats': formats,
  83             'title': self._og_search_title(webpage),
  84             'thumbnail': self._og_search_thumbnail(webpage),
  85             'description': self._og_search_description(webpage),
  86         }