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 ..compat import (
   8     compat_urllib_parse_unquote,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     update_url_query,
  14 )
  15 from .bokecc import BokeCCBaseIE
  16
  17
  18 class InfoQIE(BokeCCBaseIE):
  19     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
  20
  21     _TESTS = [{
  22         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
  23         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
  24         'info_dict': {
  25             'id': 'A-Few-of-My-Favorite-Python-Things',
  26             'ext': 'mp4',
  27             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
  28             'title': 'A Few of My Favorite [Python] Things',
  29         },
  30     }, {
  31         'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
  32         'only_matching': True,
  33     }, {
  34         'url': 'http://www.infoq.com/cn/presentations/openstack-continued-delivery',
  35         'md5': '4918d0cca1497f2244572caf626687ef',
  36         'info_dict': {
  37             'id': 'openstack-continued-delivery',
  38             'title': 'OpenStack持续交付之路',
  39             'ext': 'flv',
  40             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
  41         },
  42     }, {
  43         'url': 'https://www.infoq.com/presentations/Simple-Made-Easy',
  44         'md5': '0e34642d4d9ef44bf86f66f6399672db',
  45         'info_dict': {
  46             'id': 'Simple-Made-Easy',
  47             'title': 'Simple Made Easy',
  48             'ext': 'mp3',
  49             'description': 'md5:3e0e213a8bbd074796ef89ea35ada25b',
  50         },
  51         'params': {
  52             'format': 'bestaudio',
  53         },
  54     }]
  55
  56     def _extract_rtmp_video(self, webpage):
  57         # The server URL is hardcoded
  58         video_url = 'rtmpe://video.infoq.com/cfx/st/'
  59
  60         # Extract video URL
  61         encoded_id = self._search_regex(
  62             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
  63
  64         real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
  65         playpath = 'mp4:' + real_id
  66
  67         return [{
  68             'format_id': 'rtmp_video',
  69             'url': video_url,
  70             'ext': determine_ext(playpath),
  71             'play_path': playpath,
  72         }]
  73
  74     def _extract_cf_auth(self, webpage):
  75         policy = self._search_regex(r'InfoQConstants\.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
  76         signature = self._search_regex(r'InfoQConstants\.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
  77         key_pair_id = self._search_regex(r'InfoQConstants\.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
  78         return {
  79             'Policy': policy,
  80             'Signature': signature,
  81             'Key-Pair-Id': key_pair_id,
  82         }
  83
  84     def _extract_http_video(self, webpage):
  85         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
  86         http_video_url = update_url_query(http_video_url, self._extract_cf_auth(webpage))
  87         return [{
  88             'format_id': 'http_video',
  89             'url': http_video_url,
  90         }]
  91
  92     def _extract_http_audio(self, webpage, video_id):
  93         fields = self._hidden_inputs(webpage)
  94         http_audio_url = fields.get('filename')
  95         if not http_audio_url:
  96             return []
  97
  98         # base URL is found in the Location header in the response returned by
  99         # GET https://www.infoq.com/mp3download.action?filename=... when logged in.
 100         http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url)
 101         http_audio_url = update_url_query(http_audio_url, self._extract_cf_auth(webpage))
 102
 103         # audio file seem to be missing some times even if there is a download link
 104         # so probe URL to make sure
 105         if not self._is_valid_url(http_audio_url, video_id):
 106             return []
 107
 108         return [{
 109             'format_id': 'http_audio',
 110             'url': http_audio_url,
 111             'vcodec': 'none',
 112         }]
 113
 114     def _real_extract(self, url):
 115         video_id = self._match_id(url)
 116         webpage = self._download_webpage(url, video_id)
 117
 118         video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
 119         video_description = self._html_search_meta('description', webpage, 'description')
 120
 121         if '/cn/' in url:
 122             # for China videos, HTTP video URL exists but always fails with 403
 123             formats = self._extract_bokecc_formats(webpage, video_id)
 124         else:
 125             formats = (
 126                 self._extract_rtmp_video(webpage) +
 127                 self._extract_http_video(webpage) +
 128                 self._extract_http_audio(webpage, video_id))
 129
 130         self._sort_formats(formats)
 131
 132         return {
 133             'id': video_id,
 134             'title': video_title,
 135             'description': video_description,
 136             'formats': formats,
 137         }