X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/1931fa5f23ac75ea766517aa1b5ebf267ad3f97d..96981814751d51f37fe7347060de3a51c982f5ec:/youtube_dl/extractor/gdcvault.py diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index de14ae1..2f555c1 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -3,20 +3,25 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from .kaltura import KalturaIE from ..utils import ( - compat_urllib_parse, - compat_urllib_request, + HEADRequest, + sanitized_Request, + smuggle_url, + urlencode_postdata, ) class GDCVaultIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P\d+)/(?P(\w|-)+)' + _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P\d+)(?:/(?P[\w-]+))?' + _NETRC_MACHINE = 'gdcvault' _TESTS = [ { 'url': 'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple', 'md5': '7ce8388f544c88b7ac11c7ab1b593704', 'info_dict': { - 'id': '1019721', + 'id': '201311826596_AWNY', + 'display_id': 'Doki-Doki-Universe-Sweet-Simple', 'ext': 'mp4', 'title': 'Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)' } @@ -24,7 +29,8 @@ class GDCVaultIE(InfoExtractor): { 'url': 'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of', 'info_dict': { - 'id': '1015683', + 'id': '201203272_1330951438328RSXR', + 'display_id': 'Embracing-the-Dark-Art-of', 'ext': 'flv', 'title': 'Embracing the Dark Art of Mathematical Modeling in AI' }, @@ -37,54 +43,69 @@ class GDCVaultIE(InfoExtractor): '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': '840376_BQRC', + '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': '12396_1299111843500GMPX', + 'ext': 'mp4', + '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': '9350_1238021887562UHXB', + 'display_id': 'An-American-engine-in-Tokyo', + 'ext': 'mp4', + 'title': 'An American Engine in Tokyo:/nThe collaboration of Epic Games and Square Enix/nFor THE LAST REMINANT', + }, + }, + { + # Kaltura Embed + 'url': 'https://www.gdcvault.com/play/1026180/Mastering-the-Apex-of-Scaling', + 'info_dict': { + 'id': '0_h1fg8j3p', + 'ext': 'mp4', + 'title': 'Mastering the Apex of Scaling Game Servers (Presented by Multiplay)', + 'timestamp': 1554401811, + 'upload_date': '20190404', + 'uploader_id': 'joe@blazestreaming.com', + }, + 'params': { + 'format': 'mp4-408', + }, + }, ] - def _parse_mp4(self, xml_description): - video_formats = [] - mp4_video = xml_description.find('./metadata/mp4video') - if mp4_video is None: - return None - - mobj = re.match(r'(?Phttps?://.*?/).*', mp4_video.text) - video_root = mobj.group('root') - formats = xml_description.findall('./metadata/MBRVideos/MBRVideo') - for format in formats: - mobj = re.match(r'mp4\:(?P.*)', format.find('streamName').text) - url = video_root + mobj.group('path') - vbr = format.find('bitrate').text - video_formats.append({ - 'url': url, - 'vbr': int(vbr), - }) - return video_formats - - def _parse_flv(self, xml_description): - video_formats = [] - akami_url = xml_description.find('./metadata/akamaiHost').text - slide_video_path = xml_description.find('./metadata/slideVideo').text - video_formats.append({ - 'url': 'rtmp://' + akami_url + '/' + slide_video_path, - '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, - 'format_note': 'speaker video', - 'quality': -1, - 'preference': -1, - 'format_id': 'speaker', - }) - return video_formats - - def _login(self, webpage_url, video_id): - (username, password) = self._get_login_info() + 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 @@ -98,69 +119,70 @@ class GDCVaultIE(InfoExtractor): 'password': password, } - request = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form)) + request = sanitized_Request(login_url, urlencode_postdata(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 def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) + video_id, name = re.match(self._VALID_URL, url).groups() + display_id = name or video_id - video_id = mobj.group('id') webpage_url = 'http://www.gdcvault.com/play/' + video_id - start_page = self._download_webpage(webpage_url, 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: - video_url = 'http://www.gdcvault.com/' + direct_url title = self._html_search_regex( - r'Session Name\s*(.*?)', + r'Session Name:?\s*(.*?)', start_page, 'title') + video_url = 'http://www.gdcvault.com' + direct_url + # resolve the url so that we can detect the correct extension + video_url = self._request_webpage( + HEADRequest(video_url), video_id).geturl() return { 'id': video_id, + 'display_id': display_id, 'url': video_url, - 'ext': 'flv', 'title': title, } - xml_root = self._html_search_regex( - r'', - start_page, 'xml root') - - xml_name = self._html_search_regex( - r'', start_page, 'xml filename') - - xml_decription_url = xml_root + 'xml/' + xml_name - xml_description = self._download_xml(xml_decription_url, video_id) - - video_title = xml_description.find('./metadata/title').text - video_formats = self._parse_mp4(xml_description) - if video_formats is None: - video_formats = self._parse_flv(xml_description) + embed_url = KalturaIE._extract_url(start_page) + if embed_url: + embed_url = smuggle_url(embed_url, {'source_url': url}) + ie_key = 'Kaltura' + else: + PLAYER_REGEX = r'', + start_page, 'xml filename') + embed_url = '%s/xml/%s' % (xml_root, xml_name) + ie_key = 'DigitallySpeaking' return { + '_type': 'url_transparent', 'id': video_id, - 'title': video_title, - 'formats': video_formats, + 'display_id': display_id, + 'url': embed_url, + 'ie_key': ie_key, }