youtube_dl/extractor/gdcvault.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import xml.etree.ElementTree
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_urllib_parse,
   9     compat_urllib_request,
  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
  37     def _parse_mp4(self, xml_description):
  38         video_formats = []
  39         mp4_video = xml_description.find('./metadata/mp4video')
  40         if mp4_video is None:
  41             return None
  42
  43         mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text)
  44         video_root = mobj.group('root')
  45         formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
  46         for format in formats:
  47             mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
  48             url = video_root + mobj.group('path')
  49             vbr = format.find('bitrate').text
  50             video_formats.append({
  51                 'url': url,
  52                 'vbr': int(vbr),
  53             })
  54         return video_formats
  55
  56     def _parse_flv(self, xml_description):
  57         video_formats = []
  58         akami_url = xml_description.find('./metadata/akamaiHost').text
  59         slide_video_path = xml_description.find('./metadata/slideVideo').text
  60         video_formats.append({
  61                 'url': 'rtmp://' + akami_url + '/' + slide_video_path,
  62                 'format_note': 'slide deck video',
  63                 'quality': -2,
  64                 'preference': -2,
  65                 'format_id': 'slides',
  66             })
  67         speaker_video_path = xml_description.find('./metadata/speakerVideo').text
  68         video_formats.append({
  69                 'url': 'rtmp://' + akami_url + '/' + speaker_video_path,
  70                 'format_note': 'speaker video',
  71                 'quality': -1,
  72                 'preference': -1,
  73                 'format_id': 'speaker',
  74             })
  75         return video_formats
  76
  77     def _login(self, webpage_url, video_id):
  78         (username, password) = self._get_login_info()
  79         if username is None or password is None:
  80             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
  81             return None
  82
  83         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
  84         login_url = mobj.group('root_url') + 'api/login.php'
  85         logout_url = mobj.group('root_url') + 'logout'
  86
  87         login_form = {
  88             'email': username,
  89             'password': password,
  90         }
  91
  92         request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
  93         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  94         login_content = self._download_webpage(request, video_id, 'Logging in')
  95         start_page = self._download_webpage(webpage_url, video_id, 'Getting authenticated video page')
  96         logout_content = self._download_webpage(logout_url, video_id, 'Logging out')
  97
  98         return start_page
  99
 100     def _real_extract(self, url):
 101         mobj = re.match(self._VALID_URL, url)
 102
 103         video_id = mobj.group('id')
 104         webpage_url = 'http://www.gdcvault.com/play/' + video_id
 105         start_page = self._download_webpage(webpage_url, video_id)
 106
 107         xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root', None, False)
 108
 109         if xml_root is None:
 110             # Probably need to authenticate
 111             start_page = self._login(webpage_url, video_id)
 112             if start_page is None:
 113                 self.report_warning('Could not login.')
 114             else:
 115                 # Grab the url from the authenticated page
 116                 xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root')
 117
 118         xml_name = self._html_search_regex(r'<iframe src=".*?\?xml=(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename', None, False)
 119         if xml_name is None:
 120             # Fallback to the older format
 121             xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')
 122
 123         xml_decription_url = xml_root + 'xml/' + xml_name
 124         xml_description = self._download_xml(xml_decription_url, video_id)
 125
 126         video_title = xml_description.find('./metadata/title').text
 127         video_formats = self._parse_mp4(xml_description)
 128         if video_formats is None:
 129             video_formats = self._parse_flv(xml_description)
 130
 131         return {
 132             'id': video_id,
 133             'title': video_title,
 134             'formats': video_formats,
 135         }