youtube_dl/extractor/teamcoco.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8 )
   9
  10
  11 class TeamcocoIE(InfoExtractor):
  12     _VALID_URL = r'http://teamcoco\.com/video/(?P<video_id>\d*)?/?(?P<url_title>.*)'
  13     _TESTS = [
  14     {
  15         'url': 'http://teamcoco.com/video/80187/conan-becomes-a-mary-kay-beauty-consultant',
  16         'file': '80187.mp4',
  17         'md5': '3f7746aa0dc86de18df7539903d399ea',
  18         'info_dict': {
  19             'title': 'Conan Becomes A Mary Kay Beauty Consultant',
  20             'description': 'Mary Kay is perhaps the most trusted name in female beauty, so of course Conan is a natural choice to sell their products.'
  21         }
  22     },
  23     {
  24         'url': 'http://teamcoco.com/video/louis-ck-interview-george-w-bush',
  25         'file': '19705.mp4',
  26         'md5': 'cde9ba0fa3506f5f017ce11ead928f9a',
  27         'info_dict': {
  28             "description": "Louis C.K. got starstruck by George W. Bush, so what? Part one.",
  29             "title": "Louis C.K. Interview Pt. 1 11/3/11"
  30         }
  31     }
  32     ]
  33
  34     def _real_extract(self, url):
  35         mobj = re.match(self._VALID_URL, url)
  36         if mobj is None:
  37             raise ExtractorError('Invalid URL: %s' % url)
  38         url_title = mobj.group('url_title')
  39         webpage = self._download_webpage(url, url_title)
  40
  41         video_id = mobj.group("video_id")
  42         if video_id == '':
  43             video_id = self._html_search_regex(
  44                 r'<article class="video" data-id="(\d+?)"',
  45                 webpage, 'video id')
  46
  47         self.report_extraction(video_id)
  48
  49         data_url = 'http://teamcoco.com/cvp/2.0/%s.xml' % video_id
  50         data = self._download_xml(data_url, video_id, 'Downloading data webpage')
  51
  52         qualities = ['500k', '480p', '1000k', '720p', '1080p']
  53         formats = []
  54         for filed in data.findall('files/file'):
  55             if filed.attrib.get('playmode') == 'all':
  56                 # it just duplicates one of the entries
  57                 break
  58             file_url = filed.text
  59             m_format = re.search(r'(\d+(k|p))\.mp4', file_url)
  60             if m_format is not None:
  61                 format_id = m_format.group(1)
  62             else:
  63                 format_id = filed.attrib['bitrate']
  64             tbr = (
  65                 int(filed.attrib['bitrate'])
  66                 if filed.attrib['bitrate'].isdigit()
  67                 else None)
  68
  69             try:
  70                 quality = qualities.index(format_id)
  71             except ValueError:
  72                 quality = -1
  73             formats.append({
  74                 'url': file_url,
  75                 'ext': 'mp4',
  76                 'tbr': tbr,
  77                 'format_id': format_id,
  78                 'quality': quality,
  79             })
  80
  81         self._sort_formats(formats)
  82
  83         return {
  84             'id': video_id,
  85             'formats': formats,
  86             'title': self._og_search_title(webpage),
  87             'thumbnail': self._og_search_thumbnail(webpage),
  88             'description': self._og_search_description(webpage),
  89         }