youtube_dl/extractor/gdcvault.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_parse,
   8     compat_urllib_request,
   9 )
  10 from ..utils import remove_end
  11
  12
  13 class GDCVaultIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)'
  15     _NETRC_MACHINE = 'gdcvault'
  16     _TESTS = [
  17         {
  18             'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
  19             'md5': '7ce8388f544c88b7ac11c7ab1b593704',
  20             'info_dict': {
  21                 'id': '1019721',
  22                 'ext': 'mp4',
  23                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
  24             }
  25         },
  26         {
  27             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
  28             'info_dict': {
  29                 'id': '1015683',
  30                 'ext': 'flv',
  31                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
  32             },
  33             'params': {
  34                 'skip_download': True,  # Requires rtmpdump
  35             }
  36         },
  37         {
  38             'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or',
  39             'md5': 'a5eb77996ef82118afbbe8e48731b98e',
  40             'info_dict': {
  41                 'id': '1015301',
  42                 'ext': 'flv',
  43                 'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment',
  44             },
  45             'skip': 'Requires login',
  46         }
  47     ]
  48
  49     def _parse_mp4(self, xml_description):
  50         video_formats = []
  51         mp4_video = xml_description.find('./metadata/mp4video')
  52         if mp4_video is None:
  53             return None
  54
  55         mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text)
  56         video_root = mobj.group('root')
  57         formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
  58         for format in formats:
  59             mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
  60             url = video_root + mobj.group('path')
  61             vbr = format.find('bitrate').text
  62             video_formats.append({
  63                 'url': url,
  64                 'vbr': int(vbr),
  65             })
  66         return video_formats
  67
  68     def _parse_flv(self, xml_description):
  69         video_formats = []
  70         akamai_url = xml_description.find('./metadata/akamaiHost').text
  71         slide_video_path = xml_description.find('./metadata/slideVideo').text
  72         video_formats.append({
  73             'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
  74             'play_path': remove_end(slide_video_path, '.flv'),
  75             'ext': 'flv',
  76             'format_note': 'slide deck video',
  77             'quality': -2,
  78             'preference': -2,
  79             'format_id': 'slides',
  80         })
  81         speaker_video_path = xml_description.find('./metadata/speakerVideo').text
  82         video_formats.append({
  83             'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
  84             'play_path': remove_end(speaker_video_path, '.flv'),
  85             'ext': 'flv',
  86             'format_note': 'speaker video',
  87             'quality': -1,
  88             'preference': -1,
  89             'format_id': 'speaker',
  90         })
  91         return video_formats
  92
  93     def _login(self, webpage_url, video_id):
  94         (username, password) = self._get_login_info()
  95         if username is None or password is None:
  96             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
  97             return None
  98
  99         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
 100         login_url = mobj.group('root_url') + 'api/login.php'
 101         logout_url = mobj.group('root_url') + 'logout'
 102
 103         login_form = {
 104             'email': username,
 105             'password': password,
 106         }
 107
 108         request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
 109         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 110         self._download_webpage(request, video_id, 'Logging in')
 111         start_page = self._download_webpage(webpage_url, video_id, 'Getting authenticated video page')
 112         self._download_webpage(logout_url, video_id, 'Logging out')
 113
 114         return start_page
 115
 116     def _real_extract(self, url):
 117         mobj = re.match(self._VALID_URL, url)
 118
 119         video_id = mobj.group('id')
 120         webpage_url = 'http://www.gdcvault.com/play/' + video_id
 121         start_page = self._download_webpage(webpage_url, video_id)
 122
 123         direct_url = self._search_regex(
 124             r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
 125             start_page, 'url', default=None)
 126         if direct_url:
 127             video_url = 'http://www.gdcvault.com/' + direct_url
 128             title = self._html_search_regex(
 129                 r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>',
 130                 start_page, 'title')
 131
 132             return {
 133                 'id': video_id,
 134                 'url': video_url,
 135                 'ext': 'flv',
 136                 'title': title,
 137             }
 138
 139         xml_root = self._html_search_regex(
 140             r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>',
 141             start_page, 'xml root', default=None)
 142         if xml_root is None:
 143             # Probably need to authenticate
 144             login_res = self._login(webpage_url, video_id)
 145             if login_res is None:
 146                 self.report_warning('Could not login.')
 147             else:
 148                 start_page = login_res
 149                 # Grab the url from the authenticated page
 150                 xml_root = self._html_search_regex(
 151                     r'<iframe src="(.*?)player.html.*?".*?</iframe>',
 152                     start_page, 'xml root')
 153
 154         xml_name = self._html_search_regex(
 155             r'<iframe src=".*?\?xml=(.+?\.xml).*?".*?</iframe>',
 156             start_page, 'xml filename', default=None)
 157         if xml_name is None:
 158             # Fallback to the older format
 159             xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')
 160
 161         xml_decription_url = xml_root + 'xml/' + xml_name
 162         xml_description = self._download_xml(xml_decription_url, video_id)
 163
 164         video_title = xml_description.find('./metadata/title').text
 165         video_formats = self._parse_mp4(xml_description)
 166         if video_formats is None:
 167             video_formats = self._parse_flv(xml_description)
 168
 169         return {
 170             'id': video_id,
 171             'title': video_title,
 172             'formats': video_formats,
 173         }