Raphaël G. Git Repositories - youtubedl/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 (
  11     remove_end,
  12     HEADRequest,
  13 )
  14
  15
  16 class GDCVaultIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)?'
  18     _NETRC_MACHINE = 'gdcvault'
  19     _TESTS = [
  20         {
  21             'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
  22             'md5': '7ce8388f544c88b7ac11c7ab1b593704',
  23             'info_dict': {
  24                 'id': '1019721',
  25                 'display_id': 'Doki-Doki-Universe-Sweet-Simple',
  26                 'ext': 'mp4',
  27                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
  28             }
  29         },
  30         {
  31             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
  32             'info_dict': {
  33                 'id': '1015683',
  34                 'display_id': 'Embracing-the-Dark-Art-of',
  35                 'ext': 'flv',
  36                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
  37             },
  38             'params': {
  39                 'skip_download': True,  # Requires rtmpdump
  40             }
  41         },
  42         {
  43             'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or',
  44             'md5': 'a5eb77996ef82118afbbe8e48731b98e',
  45             'info_dict': {
  46                 'id': '1015301',
  47                 'display_id': 'Thexder-Meets-Windows-95-or',
  48                 'ext': 'flv',
  49                 'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment',
  50             },
  51             'skip': 'Requires login',
  52         },
  53         {
  54             'url': 'http://gdcvault.com/play/1020791/',
  55             'only_matching': True,
  56         }
  57     ]
  58
  59     def _parse_mp4(self, xml_description):
  60         video_formats = []
  61         mp4_video = xml_description.find('./metadata/mp4video')
  62         if mp4_video is None:
  63             return None
  64
  65         mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text)
  66         video_root = mobj.group('root')
  67         formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
  68         for format in formats:
  69             mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
  70             url = video_root + mobj.group('path')
  71             vbr = format.find('bitrate').text
  72             video_formats.append({
  73                 'url': url,
  74                 'vbr': int(vbr),
  75             })
  76         return video_formats
  77
  78     def _parse_flv(self, xml_description):
  79         formats = []
  80         akamai_url = xml_description.find('./metadata/akamaiHost').text
  81         audios = xml_description.find('./metadata/audios')
  82         if audios is not None:
  83             for audio in audios:
  84                 formats.append({
  85                     'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
  86                     'play_path': remove_end(audio.get('url'), '.flv'),
  87                     'ext': 'flv',
  88                     'vcodec': 'none',
  89                     'format_id': audio.get('code'),
  90                 })
  91         slide_video_path = xml_description.find('./metadata/slideVideo').text
  92         formats.append({
  93             'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
  94             'play_path': remove_end(slide_video_path, '.flv'),
  95             'ext': 'flv',
  96             'format_note': 'slide deck video',
  97             'quality': -2,
  98             'preference': -2,
  99             'format_id': 'slides',
 100         })
 101         speaker_video_path = xml_description.find('./metadata/speakerVideo').text
 102         formats.append({
 103             'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
 104             'play_path': remove_end(speaker_video_path, '.flv'),
 105             'ext': 'flv',
 106             'format_note': 'speaker video',
 107             'quality': -1,
 108             'preference': -1,
 109             'format_id': 'speaker',
 110         })
 111         return formats
 112
 113     def _login(self, webpage_url, display_id):
 114         (username, password) = self._get_login_info()
 115         if username is None or password is None:
 116             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
 117             return None
 118
 119         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
 120         login_url = mobj.group('root_url') + 'api/login.php'
 121         logout_url = mobj.group('root_url') + 'logout'
 122
 123         login_form = {
 124             'email': username,
 125             'password': password,
 126         }
 127
 128         request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
 129         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 130         self._download_webpage(request, display_id, 'Logging in')
 131         start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page')
 132         self._download_webpage(logout_url, display_id, 'Logging out')
 133
 134         return start_page
 135
 136     def _real_extract(self, url):
 137         mobj = re.match(self._VALID_URL, url)
 138
 139         video_id = mobj.group('id')
 140         display_id = mobj.group('name') or video_id
 141
 142         webpage_url = 'http://www.gdcvault.com/play/' + video_id
 143         start_page = self._download_webpage(webpage_url, display_id)
 144
 145         direct_url = self._search_regex(
 146             r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
 147             start_page, 'url', default=None)
 148         if direct_url:
 149             title = self._html_search_regex(
 150                 r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>',
 151                 start_page, 'title')
 152             video_url = 'http://www.gdcvault.com' + direct_url
 153             # resolve the url so that we can detect the correct extension
 154             head = self._request_webpage(HEADRequest(video_url), video_id)
 155             video_url = head.geturl()
 156
 157             return {
 158                 'id': video_id,
 159                 'display_id': display_id,
 160                 'url': video_url,
 161                 'title': title,
 162             }
 163
 164         xml_root = self._html_search_regex(
 165             r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>',
 166             start_page, 'xml root', default=None)
 167         if xml_root is None:
 168             # Probably need to authenticate
 169             login_res = self._login(webpage_url, display_id)
 170             if login_res is None:
 171                 self.report_warning('Could not login.')
 172             else:
 173                 start_page = login_res
 174                 # Grab the url from the authenticated page
 175                 xml_root = self._html_search_regex(
 176                     r'<iframe src="(.*?)player.html.*?".*?</iframe>',
 177                     start_page, 'xml root')
 178
 179         xml_name = self._html_search_regex(
 180             r'<iframe src=".*?\?xml=(.+?\.xml).*?".*?</iframe>',
 181             start_page, 'xml filename', default=None)
 182         if xml_name is None:
 183             # Fallback to the older format
 184             xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')
 185
 186         xml_description_url = xml_root + 'xml/' + xml_name
 187         xml_description = self._download_xml(xml_description_url, display_id)
 188
 189         video_title = xml_description.find('./metadata/title').text
 190         video_formats = self._parse_mp4(xml_description)
 191         if video_formats is None:
 192             video_formats = self._parse_flv(xml_description)
 193
 194         return {
 195             'id': video_id,
 196             'display_id': display_id,
 197             'title': video_title,
 198             'formats': video_formats,
 199         }