_ Git - youtube-dl/blob - 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             'url': 'http://gdcvault.com/play/1020791/',
  49             'only_matching': True,
  50         }
  51     ]
  52
  53     def _parse_mp4(self, xml_description):
  54         video_formats = []
  55         mp4_video = xml_description.find('./metadata/mp4video')
  56         if mp4_video is None:
  57             return None
  58
  59         mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text)
  60         video_root = mobj.group('root')
  61         formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
  62         for format in formats:
  63             mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
  64             url = video_root + mobj.group('path')
  65             vbr = format.find('bitrate').text
  66             video_formats.append({
  67                 'url': url,
  68                 'vbr': int(vbr),
  69             })
  70         return video_formats
  71
  72     def _parse_flv(self, xml_description):
  73         video_formats = []
  74         akamai_url = xml_description.find('./metadata/akamaiHost').text
  75         slide_video_path = xml_description.find('./metadata/slideVideo').text
  76         video_formats.append({
  77             'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
  78             'play_path': remove_end(slide_video_path, '.flv'),
  79             'ext': 'flv',
  80             'format_note': 'slide deck video',
  81             'quality': -2,
  82             'preference': -2,
  83             'format_id': 'slides',
  84         })
  85         speaker_video_path = xml_description.find('./metadata/speakerVideo').text
  86         video_formats.append({
  87             'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
  88             'play_path': remove_end(speaker_video_path, '.flv'),
  89             'ext': 'flv',
  90             'format_note': 'speaker video',
  91             'quality': -1,
  92             'preference': -1,
  93             'format_id': 'speaker',
  94         })
  95         return video_formats
  96
  97     def _login(self, webpage_url, video_id):
  98         (username, password) = self._get_login_info()
  99         if username is None or password is None:
 100             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
 101             return None
 102
 103         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
 104         login_url = mobj.group('root_url') + 'api/login.php'
 105         logout_url = mobj.group('root_url') + 'logout'
 106
 107         login_form = {
 108             'email': username,
 109             'password': password,
 110         }
 111
 112         request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
 113         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 114         self._download_webpage(request, video_id, 'Logging in')
 115         start_page = self._download_webpage(webpage_url, video_id, 'Getting authenticated video page')
 116         self._download_webpage(logout_url, video_id, 'Logging out')
 117
 118         return start_page
 119
 120     def _real_extract(self, url):
 121         mobj = re.match(self._VALID_URL, url)
 122
 123         video_id = mobj.group('id')
 124         webpage_url = 'http://www.gdcvault.com/play/' + video_id
 125         start_page = self._download_webpage(webpage_url, video_id)
 126
 127         direct_url = self._search_regex(
 128             r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
 129             start_page, 'url', default=None)
 130         if direct_url:
 131             video_url = 'http://www.gdcvault.com/' + direct_url
 132             title = self._html_search_regex(
 133                 r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>',
 134                 start_page, 'title')
 135
 136             return {
 137                 'id': video_id,
 138                 'url': video_url,
 139                 'ext': 'flv',
 140                 'title': title,
 141             }
 142
 143         xml_root = self._html_search_regex(
 144             r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>',
 145             start_page, 'xml root', default=None)
 146         if xml_root is None:
 147             # Probably need to authenticate
 148             login_res = self._login(webpage_url, video_id)
 149             if login_res is None:
 150                 self.report_warning('Could not login.')
 151             else:
 152                 start_page = login_res
 153                 # Grab the url from the authenticated page
 154                 xml_root = self._html_search_regex(
 155                     r'<iframe src="(.*?)player.html.*?".*?</iframe>',
 156                     start_page, 'xml root')
 157
 158         xml_name = self._html_search_regex(
 159             r'<iframe src=".*?\?xml=(.+?\.xml).*?".*?</iframe>',
 160             start_page, 'xml filename', default=None)
 161         if xml_name is None:
 162             # Fallback to the older format
 163             xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')
 164
 165         xml_decription_url = xml_root + 'xml/' + xml_name
 166         xml_description = self._download_xml(xml_decription_url, video_id)
 167
 168         video_title = xml_description.find('./metadata/title').text
 169         video_formats = self._parse_mp4(xml_description)
 170         if video_formats is None:
 171             video_formats = self._parse_flv(xml_description)
 172
 173         return {
 174             'id': video_id,
 175             'title': video_title,
 176             'formats': video_formats,
 177         }