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