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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import (
   7     InfoExtractor,
   8     ExtractorError,
   9 )
  10
  11 from ..utils import (
  12     int_or_none,
  13 )
  14
  15 class ScreenJunkiesIE(InfoExtractor):
  16     _VALID_URL = r'http://www.screenjunkies.com/video/(.+-(?P<id>\d+)|.+)'
  17
  18     _TESTS = [{
  19         'url': 'http://www.screenjunkies.com/video/best-quentin-tarantino-movie-2841915',
  20         'info_dict': {
  21             'id': '2841915',
  22             'ext': 'mp4',
  23             'title': 'Best Quentin Tarantino Movie',
  24         },
  25     }, {
  26         'url': 'http://www.screenjunkies.com/video/honest-trailers-the-dark-knight',
  27         'info_dict': {
  28             'id': '2348808',
  29             'ext': 'mp4',
  30             'title': "Honest Trailers - 'The Dark Knight'",
  31         },
  32     }, {
  33         'url': 'http://www.screenjunkies.com/video/knocking-dead-ep-1-the-show-so-far-3003285',
  34         'only_matching': True,
  35     }]
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39
  40         if not video_id: # Older urls didn't have the id in them, but we can grab it manually
  41             webpage = self._download_webpage(url, url)
  42             video_id = self._html_search_regex(r'src="/embed/(\d+)"', webpage, 'video id')
  43
  44         webpage = self._download_webpage('http://www.screenjunkies.com/embed/%s' %video_id, video_id)
  45         video_vars_str = self._html_search_regex(r'embedVars = (\{.+\})\s*</script>', webpage, 'video variables', flags=re.S)
  46         video_vars = self._parse_json(video_vars_str, video_id)
  47
  48         # TODO: Figure out required cookies
  49         if video_vars['subscriptionLevel'] > 0:
  50             raise ExtractorError('This video requires ScreenJunkiesPlus', expected=True)
  51
  52         formats = []
  53         for f in video_vars['media']:
  54             if not f['mediaPurpose'] == 'play':
  55                 continue
  56
  57             formats.append({
  58                 'url': f['uri'],
  59                 'width': int_or_none(f.get('width')),
  60                 'height': int_or_none(f.get('height')),
  61                 'tbr': int_or_none(f.get('bitRate')),
  62                 'format': 'mp4',
  63             })
  64         self._sort_formats(formats)
  65
  66         return {
  67             'id': video_id,
  68             'title': video_vars['contentName'],
  69             'formats': formats,
  70             'duration': int_or_none(video_vars.get('videoLengthInSeconds')),
  71             'thumbnail': video_vars.get('thumbUri'),
  72             'tags': video_vars['tags'].split(',') if 'tags' in video_vars else [],
  73         }
  74