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

   1 from __future__ import unicode_literals
   2
   3 import base64
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_urllib_parse_unquote
   7
   8
   9 class InfoQIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
  11
  12     _TESTS = [{
  13         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
  14         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
  15         'info_dict': {
  16             'id': '12-jan-pythonthings',
  17             'ext': 'mp4',
  18             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
  19             'title': 'A Few of My Favorite [Python] Things',
  20         },
  21     }, {
  22         'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
  23         'only_matching': True,
  24     }]
  25
  26     def _real_extract(self, url):
  27         video_id = self._match_id(url)
  28         webpage = self._download_webpage(url, video_id)
  29
  30         video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
  31         video_description = self._html_search_meta('description', webpage, 'description')
  32
  33         # The server URL is hardcoded
  34         video_url = 'rtmpe://video.infoq.com/cfx/st/'
  35
  36         # Extract video URL
  37         encoded_id = self._search_regex(
  38             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id')
  39         real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
  40         playpath = 'mp4:' + real_id
  41
  42         video_filename = playpath.split('/')[-1]
  43         video_id, extension = video_filename.split('.')
  44
  45         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
  46
  47         policy = self._search_regex(r'InfoQConstants.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
  48         signature = self._search_regex(r'InfoQConstants.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
  49         key_pair_id = self._search_regex(r'InfoQConstants.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
  50
  51         formats = [{
  52             'format_id': 'rtmp',
  53             'url': video_url,
  54             'ext': extension,
  55             'play_path': playpath,
  56         }, {
  57             'format_id': 'http',
  58             'url': http_video_url,
  59             'http_headers': {
  60                 'Cookie': 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % (
  61                     policy, signature, key_pair_id),
  62             },
  63         }]
  64         self._sort_formats(formats)
  65
  66         return {
  67             'id': video_id,
  68             'title': video_title,
  69             'description': video_description,
  70             'formats': formats,
  71         }