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 (
   8     compat_urllib_parse_unquote,
   9     compat_urlparse,
  10 )
  11 from ..utils import determine_ext
  12 from .bokecc import BokeCCBaseIE
  13
  14
  15 class InfoQIE(BokeCCBaseIE):
  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         'url': 'https://www.infoq.com/presentations/Simple-Made-Easy',
  41         'md5': '0e34642d4d9ef44bf86f66f6399672db',
  42         'info_dict': {
  43             'id': 'Simple-Made-Easy',
  44             'title': 'Simple Made Easy',
  45             'ext': 'mp3',
  46             'description': 'md5:3e0e213a8bbd074796ef89ea35ada25b',
  47         },
  48         'params': {
  49             'format': 'bestaudio',
  50         },
  51     }]
  52
  53     def _extract_rtmp_video(self, webpage):
  54         # The server URL is hardcoded
  55         video_url = 'rtmpe://video.infoq.com/cfx/st/'
  56
  57         # Extract video URL
  58         encoded_id = self._search_regex(
  59             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
  60
  61         real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
  62         playpath = 'mp4:' + real_id
  63
  64         return [{
  65             'format_id': 'rtmp_video',
  66             'url': video_url,
  67             'ext': determine_ext(playpath),
  68             'play_path': playpath,
  69         }]
  70
  71     def _extract_cookies(self, webpage):
  72         policy = self._search_regex(r'InfoQConstants.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
  73         signature = self._search_regex(r'InfoQConstants.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
  74         key_pair_id = self._search_regex(r'InfoQConstants.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
  75         return 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % (
  76             policy, signature, key_pair_id)
  77
  78     def _extract_http_video(self, webpage):
  79         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
  80         return [{
  81             'format_id': 'http_video',
  82             'url': http_video_url,
  83             'http_headers': {
  84                 'Cookie': self._extract_cookies(webpage)
  85             },
  86         }]
  87
  88     def _extract_http_audio(self, webpage, video_id):
  89         fields = self._hidden_inputs(webpage)
  90         http_audio_url = fields.get('filename')
  91         if not http_audio_url:
  92             return []
  93
  94         cookies_header = {'Cookie': self._extract_cookies(webpage)}
  95
  96         # base URL is found in the Location header in the response returned by
  97         # GET https://www.infoq.com/mp3download.action?filename=... when logged in.
  98         http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url)
  99
 100         # audio file seem to be missing some times even if there is a download link
 101         # so probe URL to make sure
 102         if not self._is_valid_url(http_audio_url, video_id, headers=cookies_header):
 103             return []
 104
 105         return [{
 106             'format_id': 'http_audio',
 107             'url': http_audio_url,
 108             'vcodec': 'none',
 109             'http_headers': cookies_header,
 110         }]
 111
 112     def _real_extract(self, url):
 113         video_id = self._match_id(url)
 114         webpage = self._download_webpage(url, video_id)
 115
 116         video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
 117         video_description = self._html_search_meta('description', webpage, 'description')
 118
 119         if '/cn/' in url:
 120             # for China videos, HTTP video URL exists but always fails with 403
 121             formats = self._extract_bokecc_formats(webpage, video_id)
 122         else:
 123             formats = (
 124                 self._extract_rtmp_video(webpage) +
 125                 self._extract_http_video(webpage) +
 126                 self._extract_http_audio(webpage, video_id))
 127
 128         self._sort_formats(formats)
 129
 130         return {
 131             'id': video_id,
 132             'title': video_title,
 133             'description': video_description,
 134             'formats': formats,
 135         }