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