Add support for authenticated videos

This commit is contained in:
David Wagner 2014-02-27 10:32:31 +00:00
parent 83cebb8b7a
commit 50a138d95c

View file

@ -1,25 +1,17 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import re import re
import json
import xml.etree.ElementTree import xml.etree.ElementTree
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import unified_strdate from ..utils import (
compat_urllib_parse,
compat_urllib_request,
)
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|-)+)'
_TESTS = [ _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'url': u'http://www.gdcvault.com/play/1019721/Doki-Doki-Universe-Sweet-Simple',
u'md5': u'7ce8388f544c88b7ac11c7ab1b593704', u'md5': u'7ce8388f544c88b7ac11c7ab1b593704',
@ -29,69 +21,116 @@ class GDCVaultIE(InfoExtractor):
u"title": u"Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)" u"title": u"Doki-Doki Universe: Sweet, Simple and Genuine (GDC Next 10)"
} }
}, },
{
u'url': u'http://www.gdcvault.com/play/1015683/Embracing-the-Dark-Art-of',
u'md5': u'fca91078a90f28aa5164ef6b23b78654',
u'info_dict': {
u"id": u"1015683",
u"ext": u"flv",
u"title": u"Embracing the Dark Art of Mathematical Modeling in AI"
}
},
] ]
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'(?P<root>https?://.*?/).*', 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<path>.*)', 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()
if username is None or password is None:
self.report_warning(u'It looks like ' + webpage_url + u' 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 = compat_urllib_request.Request(login_url, compat_urllib_parse.urlencode(login_form))
request.add_header('Content-Type', 'application/x-www-form-urlencoded')
login_content = self._download_webpage(request, video_id, 'Logging in')
start_page = self._download_webpage(webpage_url, video_id, 'Getting authenticated video page')
logout_content = self._download_webpage(logout_url, video_id, 'Logging out')
return start_page
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id') video_id = mobj.group('id')
webpage_url = 'http://www.gdcvault.com/play/' + video_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, video_id)
self.report_extraction(video_id) xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root', None, False)
xml_root = self._html_search_regex(r'<iframe src="(?P<xml_root>.*?)player.html.*?".*?</iframe>', start_page, 'xml root') if xml_root is None:
# Probably need to authenticate
start_page = self._login(webpage_url, video_id)
if start_page is None:
self.report_warning(u'Could not login.')
else:
# 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')
self.report_extraction(video_id)
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=(?P<xml_file>.+?\.xml).*?".*?</iframe>', start_page, 'xml filename', None, False)
if xml_name is None: if xml_name is None:
# Fallback to the older format # 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_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_decription_url = xml_root + 'xml/' + xml_name
xml_description = self._download_xml(xml_decription_url, video_id) xml_description = self._download_xml(xml_decription_url, video_id)
video_title = xml_description.find('./metadata/title').text video_title = xml_description.find('./metadata/title').text
video_details = { video_details = {
'id': video_id, 'id': video_id,
'title': video_title, 'title': video_title,
} }
video_formats = []
mp4_video = xml_description.find('./metadata/mp4video') video_formats = self._parse_mp4(xml_description)
if mp4_video is not None: if video_formats is None:
mobj = re.match(r'(?P<root>https?://.*?/).*', mp4_video.text) video_formats = self._parse_flv(xml_description)
video_root = mobj.group('root')
formats = xml_description.findall('./metadata/MBRVideos/MBRVideo')
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),
})
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': '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 [{ return [{
'id': video_id, 'id': video_id,