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