youtube_dl/extractor/infoq.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 from ..compat import (
   6     compat_b64decode,
   7     compat_urllib_parse_unquote,
   8     compat_urlparse,
   9 )
  10 from ..utils import (
  11     determine_ext,
  12     update_url_query,
  13 )
  14 from .bokecc import BokeCCBaseIE
  15
  16
  17 class InfoQIE(BokeCCBaseIE):
  18     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
  19
  20     _TESTS = [{
  21         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
  22         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
  23         'info_dict': {
  24             'id': 'A-Few-of-My-Favorite-Python-Things',
  25             'ext': 'mp4',
  26             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
  27             'title': 'A Few of My Favorite [Python] Things',
  28         },
  29     }, {
  30         'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
  31         'only_matching': True,
  32     }, {
  33         'url': 'http://www.infoq.com/cn/presentations/openstack-continued-delivery',
  34         'md5': '4918d0cca1497f2244572caf626687ef',
  35         'info_dict': {
  36             'id': 'openstack-continued-delivery',
  37             'title': 'OpenStack持续交付之路',
  38             'ext': 'flv',
  39             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
  40         },
  41     }, {
  42         'url': 'https://www.infoq.com/presentations/Simple-Made-Easy',
  43         'md5': '0e34642d4d9ef44bf86f66f6399672db',
  44         'info_dict': {
  45             'id': 'Simple-Made-Easy',
  46             'title': 'Simple Made Easy',
  47             'ext': 'mp3',
  48             'description': 'md5:3e0e213a8bbd074796ef89ea35ada25b',
  49         },
  50         'params': {
  51             'format': 'bestaudio',
  52         },
  53     }]
  54
  55     def _extract_rtmp_video(self, webpage):
  56         # The server URL is hardcoded
  57         video_url = 'rtmpe://video.infoq.com/cfx/st/'
  58
  59         # Extract video URL
  60         encoded_id = self._search_regex(
  61             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
  62
  63         real_id = compat_urllib_parse_unquote(compat_b64decode(encoded_id).decode('utf-8'))
  64         playpath = 'mp4:' + real_id
  65
  66         return [{
  67             'format_id': 'rtmp_video',
  68             'url': video_url,
  69             'ext': determine_ext(playpath),
  70             'play_path': playpath,
  71         }]
  72
  73     def _extract_cf_auth(self, webpage):
  74         policy = self._search_regex(r'InfoQConstants\.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
  75         signature = self._search_regex(r'InfoQConstants\.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
  76         key_pair_id = self._search_regex(r'InfoQConstants\.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
  77         return {
  78             'Policy': policy,
  79             'Signature': signature,
  80             'Key-Pair-Id': key_pair_id,
  81         }
  82
  83     def _extract_http_video(self, webpage):
  84         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
  85         http_video_url = update_url_query(http_video_url, self._extract_cf_auth(webpage))
  86         return [{
  87             'format_id': 'http_video',
  88             'url': http_video_url,
  89         }]
  90
  91     def _extract_http_audio(self, webpage, video_id):
  92         fields = self._hidden_inputs(webpage)
  93         http_audio_url = fields.get('filename')
  94         if not http_audio_url:
  95             return []
  96
  97         # base URL is found in the Location header in the response returned by
  98         # GET https://www.infoq.com/mp3download.action?filename=... when logged in.
  99         http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url)
 100         http_audio_url = update_url_query(http_audio_url, self._extract_cf_auth(webpage))
 101
 102         # audio file seem to be missing some times even if there is a download link
 103         # so probe URL to make sure
 104         if not self._is_valid_url(http_audio_url, video_id):
 105             return []
 106
 107         return [{
 108             'format_id': 'http_audio',
 109             'url': http_audio_url,
 110             'vcodec': 'none',
 111         }]
 112
 113     def _real_extract(self, url):
 114         video_id = self._match_id(url)
 115         webpage = self._download_webpage(url, video_id)
 116
 117         video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
 118         video_description = self._html_search_meta('description', webpage, 'description')
 119
 120         if '/cn/' in url:
 121             # for China videos, HTTP video URL exists but always fails with 403
 122             formats = self._extract_bokecc_formats(webpage, video_id)
 123         else:
 124             formats = (
 125                 self._extract_rtmp_video(webpage) +
 126                 self._extract_http_video(webpage) +
 127                 self._extract_http_audio(webpage, video_id))
 128
 129         self._sort_formats(formats)
 130
 131         return {
 132             'id': video_id,
 133             'title': video_title,
 134             'description': video_description,
 135             'formats': formats,
 136         }