X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fgdcvault.py;h=8806dc48a945e7dea5760498f5b2ef570fb419a4;hb=a2b6f946f17ba231131166703a8702dc52f7be62;hp=a1e4c32212343a8413276bd48143e7ec1c03380b;hpb=9e68f9fdf1efa6b6afe141feb27d9626e40c3056;p=youtube-dl

diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py
index a1e4c3221..8806dc48a 100644
--- a/youtube_dl/extractor/gdcvault.py
+++ b/youtube_dl/extractor/gdcvault.py
@@ -1,61 +1,174 @@
 from __future__ import unicode_literals
 
 import re
-import json
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
-from ..utils import unified_strdate
+from ..utils import (
+    HEADRequest,
+    sanitized_Request,
+    urlencode_postdata,
+)
 
 
 class GDCVaultIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)'
-
-    _TEST = {
-        u'url': u'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
-        u'md5': u'7ce8388f544c88b7ac11c7ab1b593704',
-        u'info_dict': {
-            u"id": u"1019721",
-            u"ext": u"mp4",
-            u"title": u"Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)"
+    _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)?'
+    _NETRC_MACHINE = 'gdcvault'
+    _TESTS = [
+        {
+            'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
+            'md5': '7ce8388f544c88b7ac11c7ab1b593704',
+            'info_dict': {
+                'id': '1019721',
+                'display_id': 'Doki-Doki-Universe-Sweet-Simple',
+                'ext': 'mp4',
+                'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
+            }
+        },
+        {
+            'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
+            'info_dict': {
+                'id': '1015683',
+                'display_id': 'Embracing-the-Dark-Art-of',
+                'ext': 'flv',
+                'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
+            },
+            'params': {
+                'skip_download': True,  # Requires rtmpdump
+            }
+        },
+        {
+            'url': 'http://www.gdcvault.com/play/1015301/Thexder-Meets-Windows-95-or',
+            'md5': 'a5eb77996ef82118afbbe8e48731b98e',
+            'info_dict': {
+                'id': '1015301',
+                'display_id': 'Thexder-Meets-Windows-95-or',
+                'ext': 'flv',
+                'title': 'Thexder Meets Windows 95, or Writing Great Games in the Windows 95 Environment',
+            },
+            'skip': 'Requires login',
+        },
+        {
+            'url': 'http://gdcvault.com/play/1020791/',
+            'only_matching': True,
+        },
+        {
+            # Hard-coded hostname
+            'url': 'http://gdcvault.com/play/1023460/Tenacious-Design-and-The-Interface',
+            'md5': 'a8efb6c31ed06ca8739294960b2dbabd',
+            'info_dict': {
+                'id': '1023460',
+                'ext': 'mp4',
+                'display_id': 'Tenacious-Design-and-The-Interface',
+                'title': 'Tenacious Design and The Interface of \'Destiny\'',
+            },
+        },
+        {
+            # Multiple audios
+            'url': 'http://www.gdcvault.com/play/1014631/Classic-Game-Postmortem-PAC',
+            'info_dict': {
+                'id': '1014631',
+                'ext': 'flv',
+                'title': 'How to Create a Good Game - From My Experience of Designing Pac-Man',
+            },
+            'params': {
+                'skip_download': True,  # Requires rtmpdump
+                'format': 'jp',  # The japanese audio
+            }
+        },
+        {
+            # gdc-player.html
+            'url': 'http://www.gdcvault.com/play/1435/An-American-engine-in-Tokyo',
+            'info_dict': {
+                'id': '1435',
+                'display_id': 'An-American-engine-in-Tokyo',
+                'ext': 'flv',
+                'title': 'An American Engine in Tokyo:/nThe collaboration of Epic Games and Square Enix/nFor THE LAST REMINANT',
+            },
+            'params': {
+                'skip_download': True,  # Requires rtmpdump
+            },
+        },
+    ]
+
+    def _login(self, webpage_url, display_id):
+        username, password = self._get_login_info()
+        if username is None or password is None:
+            self.report_warning('It looks like ' + webpage_url + ' requires a login. Try specifying a username and password and try again.')
+            return None
+
+        mobj = re.match(r'(?P<root_url>https?://.*?/).*', webpage_url)
+        login_url = mobj.group('root_url') + 'api/login.php'
+        logout_url = mobj.group('root_url') + 'logout'
+
+        login_form = {
+            'email': username,
+            'password': password,
         }
-    }
+
+        request = sanitized_Request(login_url, urlencode_postdata(login_form))
+        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        self._download_webpage(request, display_id, 'Logging in')
+        start_page = self._download_webpage(webpage_url, display_id, 'Getting authenticated video page')
+        self._download_webpage(logout_url, display_id, 'Logging out')
+
+        return start_page
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
         video_id = mobj.group('id')
-        webpage_url = 'http://www.gdcvault.com/play/' + video_id
-
-        start_page = self._download_webpage(webpage_url, video_id)
+        display_id = mobj.group('name') or video_id
 
-        self.report_extraction(video_id)
+        webpage_url = 'http://www.gdcvault.com/play/' + video_id
+        start_page = self._download_webpage(webpage_url, display_id)
 
-        xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root')
-        xml_name = self._html_search_regex(r'<iframe src=".*?\?xml=(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename')
-        xml_decription_url = xml_root + 'xml/' + xml_name
+        direct_url = self._search_regex(
+            r's1\.addVariable\("file",\s*encodeURIComponent\("(/[^"]+)"\)\);',
+            start_page, 'url', default=None)
+        if direct_url:
+            title = self._html_search_regex(
+                r'<td><strong>Session Name</strong></td>\s*<td>(.*?)</td>',
+                start_page, 'title')
+            video_url = 'http://www.gdcvault.com' + direct_url
+            # resolve the url so that we can detect the correct extension
+            head = self._request_webpage(HEADRequest(video_url), video_id)
+            video_url = head.geturl()
 
-        xml_description = self._download_xml(xml_decription_url, video_id)
+            return {
+                'id': video_id,
+                'display_id': display_id,
+                'url': video_url,
+                'title': title,
+            }
 
-        video_title = xml_description.find('./metadata/title').text
+        PLAYER_REGEX = r'<iframe src="(?P<xml_root>.+?)/(?:gdc-)?player.*?\.html.*?".*?</iframe>'
 
-        mp4_video = xml_description.find('./metadata/mp4video').text
-        mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video)
-        video_root = mobj.group('root')
+        xml_root = self._html_search_regex(
+            PLAYER_REGEX, start_page, 'xml root', default=None)
+        if xml_root is None:
+            # Probably need to authenticate
+            login_res = self._login(webpage_url, display_id)
+            if login_res is None:
+                self.report_warning('Could not login.')
+            else:
+                start_page = login_res
+                # Grab the url from the authenticated page
+                xml_root = self._html_search_regex(
+                    PLAYER_REGEX, start_page, 'xml root')
 
-        formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
-        video_formats = []
-        for format in formats:
-            mobj = re.match(r'mp4\:(?P<path>.*)', format.find('streamName').text)
-            url = video_root + mobj.group('path')
-            vbr = format.find('bitrate').text
-            video_formats.append({
-                'url': url,
-                'vbr': int(vbr),
-            })
+        xml_name = self._html_search_regex(
+            r'<iframe src=".*?\?xml=(.+?\.xml).*?".*?</iframe>',
+            start_page, 'xml filename', default=None)
+        if xml_name is None:
+            # Fallback to the older format
+            xml_name = self._html_search_regex(
+                r'<iframe src=".*?\?xmlURL=xml/(?P<xml_file>.+?\.xml).*?".*?</iframe>',
+                start_page, 'xml filename')
 
-        return [{
+        return {
+            '_type': 'url_transparent',
             'id': video_id,
-            'formats': video_formats,
-            'title': video_title,
-        }]
+            'display_id': display_id,
+            'url': '%s/xml/%s' % (xml_root, xml_name),
+            'ie_key': 'DigitallySpeaking',
+        }