Merge branch 'master' of github.com:rg3/youtube-dl

This commit is contained in:
Philipp Hagemeister 2013-12-05 14:29:14 +01:00
commit ef2fac6f4a
6 changed files with 98 additions and 6 deletions

View File

@ -33,6 +33,7 @@ def __init__(self, *args, **kwargs):
INFO_JSON_FILE = TEST_ID + '.info.json'
DESCRIPTION_FILE = TEST_ID + '.mp4.description'
EXPECTED_DESCRIPTION = u'''test chars: "'/\ä↭𝕐
test URL: https://github.com/rg3/youtube-dl/issues/1892
This is a test video for youtube-dl.

View File

@ -145,6 +145,7 @@
from .techtalks import TechTalksIE
from .ted import TEDIE
from .tf1 import TF1IE
from .theplatform import ThePlatformIE
from .thisav import ThisAVIE
from .toutv import TouTvIE
from .traileraddict import TrailerAddictIE

View File

@ -11,7 +11,7 @@ class HotNewHipHopIE(InfoExtractor):
u'file': u'1435540.mp3',
u'md5': u'2c2cd2f76ef11a9b3b581e8b232f3d96',
u'info_dict': {
u"title": u"Freddie Gibbs - Lay It Down"
u"title": u'Freddie Gibbs "Lay It Down"'
}
}

View File

@ -69,6 +69,21 @@ class MetacafeIE(InfoExtractor):
u'age_limit': 18,
},
},
# cbs video
{
u'url': u'http://www.metacafe.com/watch/cb-0rOxMBabDXN6/samsung_galaxy_note_2_samsungs_next_generation_phablet/',
u'info_dict': {
u'id': u'0rOxMBabDXN6',
u'ext': u'flv',
u'title': u'Samsung Galaxy Note 2: Samsung\'s next-generation phablet',
u'description': u'md5:54d49fac53d26d5a0aaeccd061ada09d',
u'duration': 129,
},
u'params': {
# rtmp download
u'skip_download': True,
},
},
]
@ -106,10 +121,16 @@ def _real_extract(self, url):
video_id = mobj.group(1)
# Check if video comes from YouTube
mobj2 = re.match(r'^yt-(.*)$', video_id)
if mobj2 is not None:
return [self.url_result('http://www.youtube.com/watch?v=%s' % mobj2.group(1), 'Youtube')]
# the video may come from an external site
m_external = re.match('^(\w{2})-(.*)$', video_id)
if m_external is not None:
prefix, ext_id = m_external.groups()
# Check if video comes from YouTube
if prefix == 'yt':
return self.url_result('http://www.youtube.com/watch?v=%s' % ext_id, 'Youtube')
# CBS videos use theplatform.com
if prefix == 'cb':
return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
# Retrieve video webpage to extract further information
req = compat_urllib_request.Request('http://www.metacafe.com/watch/%s/' % video_id)

View File

@ -0,0 +1,69 @@
import re
import json
from .common import InfoExtractor
from ..utils import (
xpath_with_ns,
find_xpath_attr,
)
_x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
class ThePlatformIE(InfoExtractor):
_VALID_URL = r'(?:https?://link\.theplatform\.com/s/[^/]+/|theplatform:)(?P<id>[^/\?]+)'
_TEST = {
# from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
u'url': u'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
u'info_dict': {
u'id': u'e9I_cZgTgIPd',
u'ext': u'flv',
u'title': u'Blackberry\'s big, bold Z30',
u'description': u'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
u'duration': 247,
},
u'params': {
# rtmp download
u'skip_download': True,
},
}
def _get_info(self, video_id):
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)
info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
info_json = self._download_webpage(info_url, video_id)
info = json.loads(info_json)
head = meta.find(_x('smil:head'))
body = meta.find(_x('smil:body'))
base_url = head.find(_x('smil:meta')).attrib['base']
switch = body.find(_x('smil:switch'))
formats = []
for f in switch.findall(_x('smil:video')):
attr = f.attrib
formats.append({
'url': base_url,
'play_path': 'mp4:' + attr['src'],
'ext': 'flv',
'width': int(attr['width']),
'height': int(attr['height']),
'vbr': int(attr['system-bitrate']),
})
formats.sort(key=lambda f: (f['height'], f['width'], f['vbr']))
return {
'id': video_id,
'title': info['title'],
'formats': formats,
'description': info['description'],
'thumbnail': info['defaultThumbnailUrl'],
'duration': info['duration']//1000,
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
return self._get_info(video_id)

View File

@ -15,7 +15,7 @@ class VevoIE(InfoExtractor):
Accepts urls from vevo.com or in the format 'vevo:{id}'
(currently used by MTVIE)
"""
_VALID_URL = r'((http://www\.vevo\.com/watch/.*?/.*?/)|(vevo:))(?P<id>.*?)(\?|$)'
_VALID_URL = r'((http://www\.vevo\.com/watch/(?:[^/]+/[^/]+/)?)|(vevo:))(?P<id>.*?)(\?|$)'
_TESTS = [{
u'url': u'http://www.vevo.com/watch/hurts/somebody-to-die-for/GB1101300280',
u'file': u'GB1101300280.mp4',