Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/infoq.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 import base64
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_urllib_parse_unquote,
  10     compat_parse_qs,
  11 )
  12 from ..utils import determine_ext
  13
  14
  15 class InfoQIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
  17
  18     _TESTS = [{
  19         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
  20         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
  21         'info_dict': {
  22             'id': 'A-Few-of-My-Favorite-Python-Things',
  23             'ext': 'mp4',
  24             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
  25             'title': 'A Few of My Favorite [Python] Things',
  26         },
  27     }, {
  28         'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
  29         'only_matching': True,
  30     }, {
  31         'url': 'http://www.infoq.com/cn/presentations/openstack-continued-delivery',
  32         'md5': '4918d0cca1497f2244572caf626687ef',
  33         'info_dict': {
  34             'id': 'openstack-continued-delivery',
  35             'title': 'OpenStack持续交付之路',
  36             'ext': 'flv',
  37             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
  38         },
  39     }]
  40
  41     def _extract_bokecc_videos(self, webpage, video_id):
  42         # TODO: bokecc.com is a Chinese video cloud platform
  43         # It should have an independent extractor but I don't have other
  44         # examples using bokecc
  45         player_params_str = self._html_search_regex(
  46             r'<script[^>]+src="http://p\.bokecc\.com/player\?([^"]+)',
  47             webpage, 'player params', default=None)
  48
  49         player_params = compat_parse_qs(player_params_str)
  50
  51         info_xml = self._download_xml(
  52             'http://p.bokecc.com/servlet/playinfo?uid=%s&vid=%s&m=1' % (
  53                 player_params['siteid'][0], player_params['vid'][0]), video_id)
  54
  55         return [{
  56             'format_id': 'bokecc',
  57             'url': quality.find('./copy').attrib['playurl'],
  58             'preference': int(quality.attrib['value']),
  59         } for quality in info_xml.findall('./video/quality')]
  60
  61     def _extract_rtmp_videos(self, webpage):
  62         # The server URL is hardcoded
  63         video_url = 'rtmpe://video.infoq.com/cfx/st/'
  64
  65         # Extract video URL
  66         encoded_id = self._search_regex(
  67             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
  68
  69         real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
  70         playpath = 'mp4:' + real_id
  71
  72         return [{
  73             'format_id': 'rtmp',
  74             'url': video_url,
  75             'ext': determine_ext(playpath),
  76             'play_path': playpath,
  77         }]
  78
  79     def _extract_http_videos(self, webpage):
  80         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
  81
  82         policy = self._search_regex(r'InfoQConstants.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
  83         signature = self._search_regex(r'InfoQConstants.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
  84         key_pair_id = self._search_regex(r'InfoQConstants.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
  85
  86         return [{
  87             'format_id': 'http',
  88             'url': http_video_url,
  89             'http_headers': {
  90                 'Cookie': 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % (
  91                     policy, signature, key_pair_id),
  92             },
  93         }]
  94
  95     def _real_extract(self, url):
  96         video_id = self._match_id(url)
  97         webpage = self._download_webpage(url, video_id)
  98
  99         video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
 100         video_description = self._html_search_meta('description', webpage, 'description')
 101
 102         if '/cn/' in url:
 103             # for China videos, HTTP video URL exists but always fails with 403
 104             formats = self._extract_bokecc_videos(webpage, video_id)
 105         else:
 106             formats = self._extract_rtmp_videos(webpage) + self._extract_http_videos(webpage)
 107
 108         self._sort_formats(formats)
 109
 110         return {
 111             'id': video_id,
 112             'title': video_title,
 113             'description': video_description,
 114             'formats': formats,
 115         }