From 83cebb8b7adeff0474b32f12d651837c8353ab59 Mon Sep 17 00:00:00 2001 From: David Wagner Date: Thu, 27 Feb 2014 00:20:34 +0000 Subject: [PATCH] Add support for FLV videos with speaker decks --- youtube_dl/extractor/gdcvault.py | 87 +++++++++++++++++++++++--------- 1 file changed, 63 insertions(+), 24 deletions(-) diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index a1e4c3221..63527bc85 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -10,16 +10,26 @@ from ..utils import unified_strdate class GDCVaultIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?gdcvault\.com/play/(?P\d+)/(?P(\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)" - } - } + _TESTS = [ + { + u'url': u'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of', + u'md5': u'05763e5edd1a74776999a12b02ee1c4e', + u'info_dict': { + u"id": u"1015683", + u"ext": u"flv", + u"title": u"Embracing the Dark Art of Mathematical Modeling in AI" + } + }, + { + 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)" + } + }, + ] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) @@ -32,30 +42,59 @@ class GDCVaultIE(InfoExtractor): self.report_extraction(video_id) xml_root = self._html_search_regex(r'', start_page, 'xml filename') + 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 - mp4_video = xml_description.find('./metadata/mp4video').text - mobj = re.match(r'(?Phttps?://.*?/).*', mp4_video) - video_root = mobj.group('root') - - formats = xml_description.findall('./metadata/MBRVideos/MBRVideo') + video_details = { + 'id': video_id, + 'title': video_title, + } video_formats = [] - 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 + + mp4_video = xml_description.find('./metadata/mp4video') + if mp4_video is not 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), + }) + video_details['formats'] = video_formats + else: + # Fallback to flv + akami_url = xml_description.find('./metadata/akamaiHost').text + slide_video_path = xml_description.find('./metadata/slideVideo').text video_formats.append({ - 'url': url, - 'vbr': int(vbr), - }) + '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 [{ 'id': video_id, - 'formats': video_formats, 'title': video_title, + 'formats': video_formats, }]