youtube_dl/extractor/infoq.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import base64
   6
   7 from ..compat import compat_urllib_parse_unquote
   8 from ..utils import determine_ext
   9 from .bokecc import BokeCCBaseIE
  10
  11
  12 class InfoQIE(BokeCCBaseIE):
  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': 'A-Few-of-My-Favorite-Python-Things',
  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         'url': 'http://www.infoq.com/cn/presentations/openstack-continued-delivery',
  29         'md5': '4918d0cca1497f2244572caf626687ef',
  30         'info_dict': {
  31             'id': 'openstack-continued-delivery',
  32             'title': 'OpenStack持续交付之路',
  33             'ext': 'flv',
  34             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
  35         },
  36     }]
  37
  38     def _extract_rtmp_videos(self, webpage):
  39         # The server URL is hardcoded
  40         video_url = 'rtmpe://video.infoq.com/cfx/st/'
  41
  42         # Extract video URL
  43         encoded_id = self._search_regex(
  44             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
  45
  46         real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
  47         playpath = 'mp4:' + real_id
  48
  49         return [{
  50             'format_id': 'rtmp',
  51             'url': video_url,
  52             'ext': determine_ext(playpath),
  53             'play_path': playpath,
  54         }]
  55
  56     def _extract_http_videos(self, webpage):
  57         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
  58
  59         policy = self._search_regex(r'InfoQConstants.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
  60         signature = self._search_regex(r'InfoQConstants.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
  61         key_pair_id = self._search_regex(r'InfoQConstants.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
  62
  63         return [{
  64             'format_id': 'http',
  65             'url': http_video_url,
  66             'http_headers': {
  67                 'Cookie': 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % (
  68                     policy, signature, key_pair_id),
  69             },
  70         }]
  71
  72     def _real_extract(self, url):
  73         video_id = self._match_id(url)
  74         webpage = self._download_webpage(url, video_id)
  75
  76         video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
  77         video_description = self._html_search_meta('description', webpage, 'description')
  78
  79         if '/cn/' in url:
  80             # for China videos, HTTP video URL exists but always fails with 403
  81             formats = self._extract_bokecc_formats(webpage, video_id)
  82         else:
  83             formats = self._extract_rtmp_videos(webpage) + self._extract_http_videos(webpage)
  84
  85         self._sort_formats(formats)
  86
  87         return {
  88             'id': video_id,
  89             'title': video_title,
  90             'description': video_description,
  91             'formats': formats,
  92         }