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