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