[sportdeutschland] add new extractor

pull/3608/head
Philipp Hagemeister 2014-08-26 12:51:13 +02:00
parent 33ac271ba7
commit 704df56da7
3 changed files with 127 additions and 0 deletions

View File

@ -315,6 +315,7 @@ from .spankwire import SpankwireIE
from .spiegel import SpiegelIE
from .spiegeltv import SpiegeltvIE
from .spike import SpikeIE
from .sportdeutschland import SportDeutschlandIE
from .stanfordoc import StanfordOpenClassroomIE
from .steam import SteamIE
from .streamcloud import StreamcloudIE

View File

@ -636,6 +636,55 @@ class InfoExtractor(object):
return formats
def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None):
formats = [{
'format_id': 'm3u8-meta',
'url': m3u8_url,
'ext': ext,
'protocol': 'm3u8',
'preference': -1,
'resolution': 'multiple',
'format_note': 'Quality selection URL',
}]
m3u8_doc = self._download_webpage(m3u8_url, video_id)
last_info = None
kv_rex = re.compile(
r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
for line in m3u8_doc.splitlines():
if line.startswith('#EXT-X-STREAM-INF:'):
last_info = {}
for m in kv_rex.finditer(line):
v = m.group('val')
if v.startswith('"'):
v = v[1:-1]
last_info[m.group('key')] = v
elif line.startswith('#') or not line.strip():
continue
else:
tbr = int_or_none(last_info.get('BANDWIDTH'), scale=1000)
f = {
'format_id': 'm3u8-%d' % (tbr if tbr else len(formats)),
'url': line.strip(),
'tbr': tbr,
'ext': ext,
}
codecs = last_info.get('CODECS')
if codecs:
video, audio = codecs.split(',')
f['vcodec'] = video.partition('.')[0]
f['acodec'] = audio.partition('.')[0]
resolution = last_info.get('RESOLUTION')
if resolution:
width_str, height_str = resolution.split('x')
f['width'] = int(width_str)
f['height'] = int(height_str)
formats.append(f)
last_info = {}
self._sort_formats(formats)
return formats
class SearchInfoExtractor(InfoExtractor):
"""

View File

@ -0,0 +1,77 @@
# coding: utf-8
from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..utils import (
compat_urllib_request,
parse_iso8601,
)
class SportDeutschlandIE(InfoExtractor):
_VALID_URL = r'https?://sportdeutschland\.tv/(?P<sport>[^/?#]+)/(?P<id>[^?#/]+)(?:$|[?#])'
_TEST = {
'url': 'http://sportdeutschland.tv/badminton/live-li-ning-badminton-weltmeisterschaft-2014-kopenhagen',
'info_dict': {
'id': 'live-li-ning-badminton-weltmeisterschaft-2014-kopenhagen',
'ext': 'mp4',
'title': 'LIVE: Li-Ning Badminton Weltmeisterschaft 2014 Kopenhagen',
'categories': ['Badminton'],
'view_count': int,
'thumbnail': 're:^https?://.*\.jpg',
'description': 're:^Die Badminton-WM 2014 aus Kopenhagen LIVE',
'timestamp': 1409043600,
'upload_date': '20140826',
},
'params': {
'skip_download': 'Live stream',
},
}
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
sport_id = mobj.group('sport')
api_url = 'http://splink.tv/api/permalinks/%s/%s' % (
sport_id, video_id)
req = compat_urllib_request.Request(api_url, headers={
'Accept': 'application/vnd.vidibus.v2.html+json',
'Referer': url,
})
data = self._download_json(req, video_id)
categories = list(data.get('section', {}).get('tags', {}).values())
asset = data['asset']
smil_url = asset['video']
m3u8_url = smil_url.replace('.smil', '.m3u8')
formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
smil_doc = self._download_xml(
smil_url, video_id, note='Downloading SMIL metadata')
base_url = smil_doc.find('./head/meta').attrib['base']
formats.extend([{
'format_id': 'rmtp',
'url': base_url,
'play_path': n.attrib['src'],
'ext': 'flv',
'preference': -100,
'format_note': 'Seems to fail at example stream',
} for n in smil_doc.findall('./body/video')])
self._sort_formats(formats)
return {
'id': video_id,
'formats': formats,
'title': asset['title'],
'thumbnail': asset.get('image'),
'description': asset.get('teaser'),
'categories': categories,
'view_count': asset.get('views'),
'rtmp_live': asset['live'],
'timestamp': parse_iso8601(asset.get('date')),
}