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 ..utils import (
   7     compat_urllib_parse,
   8     compat_urllib_request,
   9 )
  10
  11 class GDCVaultIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)'
  13     _TESTS = [
  14         {
  15             'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
  16             'md5': '7ce8388f544c88b7ac11c7ab1b593704',
  17             'info_dict': {
  18                 'id': '1019721',
  19                 'ext': 'mp4',
  20                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
  21             }
  22         },
  23         {
  24             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
  25             'info_dict': {
  26                 'id': '1015683',
  27                 'ext': 'flv',
  28                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
  29             },
  30             'params': {
  31                 'skip_download': True,  # Requires rtmpdump
  32             }
  33         },
  34     ]
  35
  36     def _parse_mp4(self, xml_description):
  37         video_formats = []
  38         mp4_video = xml_description.find('./metadata/mp4video')
  39         if mp4_video is None:
  40             return None
  41
  42         mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text)
  43         video_root = mobj.group('root')
  44         formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
  45         for format in formats:
  46             mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
  47             url = video_root + mobj.group('path')
  48             vbr = format.find('bitrate').text
  49             video_formats.append({
  50                 'url': url,
  51                 'vbr': int(vbr),
  52             })
  53         return video_formats
  54
  55     def _parse_flv(self, xml_description):
  56         video_formats = []
  57         akami_url = xml_description.find('./metadata/akamaiHost').text
  58         slide_video_path = xml_description.find('./metadata/slideVideo').text
  59         video_formats.append({
  60             'url': 'rtmp://' + akami_url + '/' + slide_video_path,
  61             'format_note': 'slide deck video',
  62             'quality': -2,
  63             'preference': -2,
  64             'format_id': 'slides',
  65         })
  66         speaker_video_path = xml_description.find('./metadata/speakerVideo').text
  67         video_formats.append({
  68             'url': 'rtmp://' + akami_url + '/' + speaker_video_path,
  69             'format_note': 'speaker video',
  70             'quality': -1,
  71             'preference': -1,
  72             'format_id': 'speaker',
  73         })
  74         return video_formats
  75
  76     def _login(self, webpage_url, video_id):
  77         (username, password) = self._get_login_info()
  78         if username is None or password is None:
  79             self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
  80             return None
  81
  82         mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
  83         login_url = mobj.group('root_url') + 'api/login.php'
  84         logout_url = mobj.group('root_url') + 'logout'
  85
  86         login_form = {
  87             'email': username,
  88             'password': password,
  89         }
  90
  91         request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
  92         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
  93         self._download_webpage(request, video_id, 'Logging in')
  94         start_page = self._download_webpage(webpage_url, video_id, 'Getting authenticated video page')
  95         self._download_webpage(logout_url, video_id, 'Logging out')
  96
  97         return start_page
  98
  99     def _real_extract(self, url):
 100         mobj = re.match(self._VALID_URL, url)
 101
 102         video_id = mobj.group('id')
 103         webpage_url = 'http://www.gdcvault.com/play/' + video_id
 104         start_page = self._download_webpage(webpage_url, video_id)
 105
 106         xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root', None, False)
 107
 108         if xml_root is None:
 109             # Probably need to authenticate
 110             start_page = self._login(webpage_url, video_id)
 111             if start_page is None:
 112                 self.report_warning('Could not login.')
 113             else:
 114                 # Grab the url from the authenticated page
 115                 xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root')
 116
 117         xml_name = self._html_search_regex(r'<iframe src=".*?\?xml=(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename', None, False)
 118         if xml_name is None:
 119             # Fallback to the older format
 120             xml_name = self._html_search_regex(r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')
 121
 122         xml_decription_url = xml_root + 'xml/' + xml_name
 123         xml_description = self._download_xml(xml_decription_url, video_id)
 124
 125         video_title = xml_description.find('./metadata/title').text
 126         video_formats = self._parse_mp4(xml_description)
 127         if video_formats is None:
 128             video_formats = self._parse_flv(xml_description)
 129
 130         return {
 131             'id': video_id,
 132             'title': video_title,
 133             'formats': video_formats,
 134         }