1
mirror of https://github.com/yt-dlp/yt-dlp synced 2024-11-30 21:29:36 +01:00
yt-dlp/youtube_dl/extractor/nba.py

134 lines
5.0 KiB
Python
Raw Normal View History

from __future__ import unicode_literals
2013-06-23 21:18:00 +02:00
from .common import InfoExtractor
2014-09-04 15:06:14 +02:00
from ..utils import (
parse_duration,
parse_iso8601,
int_or_none,
2014-09-04 15:06:14 +02:00
)
2013-06-23 21:18:00 +02:00
class NBABaseIE(InfoExtractor):
def _get_formats(self, video_id):
2015-10-03 10:47:19 +02:00
formats = self._extract_m3u8_formats(
'http://nbavod-f.akamaihd.net/i/nba/big%s_,640x360_664m,768x432_996,768x432_1404,960x540_2104,1280x720,.mp4.csmil/master.m3u8' % video_id,
video_id,
m3u8_id='hls')
formats.extend(self._extract_f4m_formats(
'http://nbavod-f.akamaihd.net/z/nba/big%s_,640x360_664m,768x432_996,768x432_1404,960x540_2104,1280x720,.mp4.csmil/manifest.f4m?hdcore=3.4.1.1' % video_id,
video_id,
f4m_id='hds'))
base_url = 'http://nba.cdn.turner.com/nba/big%s' % video_id
2015-10-03 10:47:19 +02:00
formats.extend([{
'url': base_url + '_nba_ipad.mp4',
'width': 400,
'height': 224,
'format_id': '224p',
'preference': 1,
},{
'url': base_url + '_nba_android_high.mp4',
'width': 480,
'height': 320,
'format_id': '320p',
2015-10-03 10:47:19 +02:00
'preference': 2,
},{
'url': base_url + '_nba_576x324.mp4',
'width': 576,
'height': 324,
'format_id': '324p',
'preference': 3,
},{
'url': base_url + '_640x360_664b.mp4',
'width': 640,
'height': 360,
'format_id': '360p',
2015-10-03 10:47:19 +02:00
'preference': 4,
},{
'url': base_url + '_768x432_1404.mp4',
'width': 768,
'height': 432,
'format_id': '432p',
2015-10-03 10:47:19 +02:00
'preference': 5,
},{
'url': base_url + '_960x540_2104.mp4',
'width': 960,
'height': 540,
'format_id': '540p',
'preference': 6,
},{
'url': base_url + '_1280x720.mp4',
'width': 1280,
'height': 720,
'format_id': '720p',
2015-10-03 10:47:19 +02:00
'preference': 7,
}])
self._sort_formats(formats)
return formats
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
ret = self._extract_metadata(webpage, video_id)
ret['id'] = video_id.rpartition('/')[2]
ret['formats'] = self._get_formats(video_id)
return ret
class NBAIE(NBABaseIE):
IE_NAME = 'nba'
_VALID_URL = r'https?://(?:www\.)?nba\.com/(?:nba/)?video(?P<id>/[^?]*?)/?(?:/index\.html)?(?:\?.*)?$'
2014-12-06 11:26:17 +01:00
_TESTS = [{
'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
'md5': '9d902940d2a127af3f7f9d2f3dc79c96',
'info_dict': {
2014-03-29 14:54:45 +01:00
'id': '0021200253-okc-bkn-recap.nba',
'ext': 'mp4',
'title': 'Thunder vs. Nets',
2014-09-04 15:06:14 +02:00
'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
'duration': 181,
'timestamp': 1354680189,
'upload_date': '20121205',
},
2014-12-06 11:26:17 +01:00
}, {
'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
'only_matching': True,
}]
def _extract_metadata(self, webpage, video_id):
return {
'title': self._html_search_meta('name', webpage),
'description': self._html_search_meta('description', webpage),
'duration': parse_duration(self._html_search_meta('duration', webpage)),
'thumbnail': self._html_search_meta('thumbnailUrl', webpage),
'timestamp': parse_iso8601(self._html_search_meta('uploadDate', webpage))
}
class NBAWatchIE(NBABaseIE):
IE_NAME = 'nba:watch'
_VALID_URL = r'https?://watch.nba\.com/(?:nba/)?video(?P<id>/[^?]*?)/?(?:/index\.html)?(?:\?.*)?$'
_TESTS = [{
'url': 'http://watch.nba.com/nba/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
'info_dict': {
'id': '0041400301-cle-atl-recap.nba',
'ext': 'mp4',
'title': 'Hawks vs. Cavaliers Game 1',
'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
'duration': 228,
'timestamp': 1432094400,
'upload_date': '20150520',
}
2014-12-06 11:26:17 +01:00
}]
2013-06-23 21:18:00 +02:00
def _extract_metadata(self, webpage, video_id):
program_id = self._search_regex(r'var\s+programId\s*=\s*(\d+);', webpage, 'program id')
metadata = self._download_json(
'http://smbsolr.cdnak.neulion.com/solr_nbav6/nba/nba/mlt/?wt=json&fl=name,description,image,runtime,releaseDate&q=sequence%3A' + program_id, video_id)['match']['docs'][0]
return {
'title': metadata['name'],
'description': metadata.get('description'),
'duration': int_or_none(metadata.get('runtime')),
'thumbnail': metadata.get('image'),
'timestamp': parse_iso8601(metadata.get('releaseDate'))
2013-06-23 21:18:00 +02:00
}