Merge tag 'upstream/2015.11.10'

[youtubedl] / youtube_dl / extractor / gdcvault.py

diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py
index 89d5994eef02623222b73fa04e0287a264b01333..a6834db433de249ed2d0f9ffa7c07c3ff5becc16 100644 (file)
--- a/youtube_dl/extractor/gdcvault.py
+++ b/youtube_dl/extractor/gdcvault.py
@@ -3,19 +3,26 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..utils import (
+from ..compat import (

     compat_urllib_parse,
     compat_urllib_request,
 )
     compat_urllib_parse,
     compat_urllib_request,
 )

+from ..utils import (
+    remove_end,
+    HEADRequest,
+)
+

 
 class GDCVaultIE(InfoExtractor):
 
 class GDCVaultIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P<id>\d+)/(?P<name>(\w|-)+)'
+    _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',
     _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)'
             }
                 'ext': 'mp4',
                 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)'
             }


@@ -24,6 +31,7 @@ class GDCVaultIE(InfoExtractor):
             'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
             'info_dict': {
                 'id': '1015683',
             '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'
             },
                 'ext': 'flv',
                 'title': 'Embracing the Dark Art of Mathematical Modeling in AI'
             },


@@ -31,6 +39,21 @@ class GDCVaultIE(InfoExtractor):
                 'skip_download': True,  # Requires rtmpdump
             }
         },
                 '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,
+        }

     ]
 
     def _parse_mp4(self, xml_description):
     ]
 
     def _parse_mp4(self, xml_description):


@@ -53,27 +76,41 @@ class GDCVaultIE(InfoExtractor):
         return video_formats
 
     def _parse_flv(self, xml_description):
         return video_formats
 
     def _parse_flv(self, xml_description):

-        video_formats = []
-        akami_url = xml_description.find('./metadata/akamaiHost').text
+        formats = []
+        akamai_url = xml_description.find('./metadata/akamaiHost').text
+        audios = xml_description.find('./metadata/audios')
+        if audios is not None:
+            for audio in audios:
+                formats.append({
+                    'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
+                    'play_path': remove_end(audio.get('url'), '.flv'),
+                    'ext': 'flv',
+                    'vcodec': 'none',
+                    'format_id': audio.get('code'),
+                })

         slide_video_path = xml_description.find('./metadata/slideVideo').text
         slide_video_path = xml_description.find('./metadata/slideVideo').text

-        video_formats.append({
-            'url': 'rtmp://' + akami_url + '/' + slide_video_path,
+        formats.append({
+            'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
+            'play_path': remove_end(slide_video_path, '.flv'),
+            'ext': 'flv',

             'format_note': 'slide deck video',
             'quality': -2,
             'preference': -2,
             'format_id': 'slides',
         })
         speaker_video_path = xml_description.find('./metadata/speakerVideo').text
             'format_note': 'slide deck video',
             'quality': -2,
             'preference': -2,
             'format_id': 'slides',
         })
         speaker_video_path = xml_description.find('./metadata/speakerVideo').text

-        video_formats.append({
-            'url': 'rtmp://' + akami_url + '/' + speaker_video_path,
+        formats.append({
+            'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url,
+            'play_path': remove_end(speaker_video_path, '.flv'),
+            'ext': 'flv',

             'format_note': 'speaker video',
             'quality': -1,
             'preference': -1,
             'format_id': 'speaker',
         })
             'format_note': 'speaker video',
             'quality': -1,
             'preference': -1,
             'format_id': 'speaker',
         })

-        return video_formats
+        return formats

 
 

-    def _login(self, webpage_url, video_id):
+    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.')
         (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.')


@@ -90,9 +127,9 @@ class GDCVaultIE(InfoExtractor):
 
         request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')
 
         request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
         request.add_header('Content-Type', 'application/x-www-form-urlencoded')

-        self._download_webpage(request, video_id, 'Logging in')
-        start_page = self._download_webpage(webpage_url, video_id, 'Getting authenticated video page')
-        self._download_webpage(logout_url, video_id, 'Logging out')
+        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
 
 
         return start_page
 


@@ -100,27 +137,54 @@ class GDCVaultIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
 
         video_id = mobj.group('id')
         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

 
 

-        xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root', None, False)
+        webpage_url = 'http://www.gdcvault.com/play/' + video_id
+        start_page = self._download_webpage(webpage_url, display_id)
+
+        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()
+
+            return {
+                'id': video_id,
+                'display_id': display_id,
+                'url': video_url,
+                'title': title,
+            }

 
 

+        xml_root = self._html_search_regex(
+            r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>',
+            start_page, 'xml root', default=None)

         if xml_root is None:
             # Probably need to authenticate
         if xml_root is None:
             # Probably need to authenticate

-            start_page = self._login(webpage_url, video_id)
-            if start_page is None:
+            login_res = self._login(webpage_url, display_id)
+            if login_res is None:

                 self.report_warning('Could not login.')
             else:
                 self.report_warning('Could not login.')
             else:

+                start_page = login_res

                 # Grab the url from the authenticated page
                 # Grab the url from the authenticated page

-                xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root')
+                xml_root = self._html_search_regex(
+                    r'<iframe src="(.*?)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', None, False)
+        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')
 
         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')
 

-        xml_decription_url = xml_root + 'xml/' + xml_name
-        xml_description = self._download_xml(xml_decription_url, video_id)
+        xml_description_url = xml_root + 'xml/' + xml_name
+        xml_description = self._download_xml(xml_description_url, display_id)

 
         video_title = xml_description.find('./metadata/title').text
         video_formats = self._parse_mp4(xml_description)
 
         video_title = xml_description.find('./metadata/title').text
         video_formats = self._parse_mp4(xml_description)


@@ -129,6 +193,7 @@ class GDCVaultIE(InfoExtractor):
 
         return {
             'id': video_id,
 
         return {
             'id': video_id,

+            'display_id': display_id,

             'title': video_title,
             'formats': video_formats,
         }
             'title': video_title,
             'formats': video_formats,
         }