Add an extractor for syfy.com

It uses theplatfrom.com, which has been updated to work with f4m manifests
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-12-23 16:41:35 +01:00
parent 17968e444c
commit a97bcd80ba
3 changed files with 73 additions and 24 deletions

View File

@ -203,6 +203,7 @@ from .stanfordoc import StanfordOpenClassroomIE
from .statigram import StatigramIE from .statigram import StatigramIE
from .steam import SteamIE from .steam import SteamIE
from .streamcloud import StreamcloudIE from .streamcloud import StreamcloudIE
from .syfy import SyfyIE
from .sztvhu import SztvHuIE from .sztvhu import SztvHuIE
from .teamcoco import TeamcocoIE from .teamcoco import TeamcocoIE
from .techtalks import TechTalksIE from .techtalks import TechTalksIE

View File

@ -0,0 +1,30 @@
from __future__ import unicode_literals
import re
from .common import InfoExtractor
class SyfyIE(InfoExtractor):
_VALID_URL = r'https?://www\.syfy\.com/videos/.+?vid:(?P<id>\d+)'
_TEST = {
'url': 'http://www.syfy.com/videos/Robot%20Combat%20League/Behind%20the%20Scenes/vid:2631458',
'info_dict': {
'id': 'NmqMrGnXvmO1',
'ext': 'flv',
'title': 'George Lucas has Advice for his Daughter',
'description': 'Listen to what insights George Lucas give his daughter Amanda.',
},
'params': {
# f4m download
'skip_download': True,
},
'add_ie': ['ThePlatform'],
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
webpage = self._download_webpage(url, video_id)
return self.url_result(self._og_search_video_url(webpage))

View File

@ -11,7 +11,10 @@ _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language
class ThePlatformIE(InfoExtractor): class ThePlatformIE(InfoExtractor):
_VALID_URL = r'(?:https?://link\.theplatform\.com/s/[^/]+/|theplatform:)(?P<id>[^/\?]+)' _VALID_URL = r'''(?x)
(?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
(?P<config>[^/\?]+/(?:swf|config)/select/)?
|theplatform:)(?P<id>[^/\?&]+)'''
_TEST = { _TEST = {
# from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/ # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
@ -29,9 +32,7 @@ class ThePlatformIE(InfoExtractor):
}, },
} }
def _get_info(self, video_id): def _get_info(self, video_id, smil_url):
smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
'format=smil&mbr=true'.format(video_id))
meta = self._download_xml(smil_url, video_id) meta = self._download_xml(smil_url, video_id)
try: try:
@ -50,6 +51,15 @@ class ThePlatformIE(InfoExtractor):
head = meta.find(_x('smil:head')) head = meta.find(_x('smil:head'))
body = meta.find(_x('smil:body')) body = meta.find(_x('smil:body'))
f4m_node = body.find(_x('smil:seq/smil:video'))
if f4m_node is not None:
formats = [{
'ext': 'flv',
# the parameters are from syfy.com, other sites may use others
'url': f4m_node.attrib['src'] + '?g=UXWGVKRWHFSP&hdcore=3.0.3',
}]
else:
base_url = head.find(_x('smil:meta')).attrib['base'] base_url = head.find(_x('smil:meta')).attrib['base']
switch = body.find(_x('smil:switch')) switch = body.find(_x('smil:switch'))
formats = [] formats = []
@ -68,7 +78,6 @@ class ThePlatformIE(InfoExtractor):
'height': height, 'height': height,
'vbr': vbr, 'vbr': vbr,
}) })
self._sort_formats(formats) self._sort_formats(formats)
return { return {
@ -83,4 +92,13 @@ class ThePlatformIE(InfoExtractor):
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')
return self._get_info(video_id) if mobj.group('config'):
config_url = url+ '&form=json'
config_url = config_url.replace('swf/', 'config/')
config_json = self._download_webpage(config_url, video_id, u'Downloading config')
config = json.loads(config_json)
smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4'
else:
smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
'format=smil&mbr=true'.format(video_id))
return self._get_info(video_id, smil_url)