mollusk/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

[nba] improve(fixes #7068)

Browse Source 
* extract more formats
* extract videos from team mini sites
* extract more metadata
This commit is contained in:
remitamine
2015-12-04 07:20:27 +01:00
parent 71d9fe7818 46cc1c65a4
commit 9c117d345f
1 changed files with 56 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

83
youtube_dl/extractor/nba.py
View File

@@ -1,63 +1,92 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
remove_end,
parse_duration, parse_duration,
int_or_none,
xpath_text,
xpath_attr,
) )
class NBAIE(InfoExtractor): class NBAIE(InfoExtractor):
_VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?:nba/)?video(?P<id>/[^?]*?)/?(?:/index\.html)?(?:\?.*)?$' _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?P<path>(?:[^/]+/)?video/(?P<id>[^?]*?))/?(?:/index\.html)?(?:\?.*)?$'
_TESTS = [{ _TESTS = [{
'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html', 'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
'md5': 'c0edcfc37607344e2ff8f13c378c88a4', 'md5': '9e7729d3010a9c71506fd1248f74e4f4',
'info_dict': { 'info_dict': {
'id': '0021200253-okc-bkn-recap.nba', 'id': '0021200253-okc-bkn-recap',
'ext': 'mp4', 'ext': 'flv',
'title': 'Thunder vs. Nets', 'title': 'Thunder vs. Nets',
'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.', 'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
'duration': 181, 'duration': 181,
'timestamp': 1354638466,
'upload_date': '20121204',
}, },
}, { }, {
'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/', 'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
'only_matching': True, 'only_matching': True,
}, { }, {
'url': 'http://watch.nba.com/nba/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba', 'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
'info_dict': { 'info_dict': {
'id': '0041400301-cle-atl-recap.nba', 'id': '0041400301-cle-atl-recap',
'ext': 'mp4', 'ext': 'mp4',
'title': 'NBA GAME TIME | Video: Hawks vs. Cavaliers Game 1', 'title': 'Hawks vs. Cavaliers Game 1',
'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d', 'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
'duration': 228, 'duration': 228,
}, 'timestamp': 1432134543,
'params': { 'upload_date': '20150520',
'skip_download': True,
} }
}] }]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) path, video_id = re.match(self._VALID_URL, url).groups()
webpage = self._download_webpage(url, video_id) video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id)
video_id = xpath_text(video_info, 'slug')
title = xpath_text(video_info, 'headline')
description = xpath_text(video_info, 'description')
duration = parse_duration(xpath_text(video_info, 'length'))
timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts'))
video_url = 'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4' thumbnails = []
for image in video_info.find('images'):
thumbnails.append({
'id': image.attrib.get('cut'),
'url': image.text,
'width': int_or_none(image.attrib.get('width')),
'height': int_or_none(image.attrib.get('height')),
})
shortened_video_id = video_id.rpartition('/')[2] formats = []
title = remove_end( for video_file in video_info.find('files').iter('file'):
self._og_search_title(webpage, default=shortened_video_id), ' : NBA.com') video_url = video_file.text
if video_url.startswith('/'):
description = self._og_search_description(webpage) continue
duration_str = self._html_search_meta( if video_url.endswith('.m3u8'):
'duration', webpage, 'duration', default=None) formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls'))
if not duration_str: elif video_url.endswith('.f4m'):
duration_str = self._html_search_regex( formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds'))
r'Duration:</b>\s*(\d+:\d+)', webpage, 'duration', fatal=False) else:
duration = parse_duration(duration_str) key = video_file.attrib.get('bitrate')
width, height, bitrate = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key).groups()
formats.append({
'format_id': key,
'url': video_url,
'width': int_or_none(width),
'height': int_or_none(height),
'tbr': int_or_none(bitrate),
})
self._sort_formats(formats)
return { return {
'id': shortened_video_id, 'id': video_id,
'url': video_url,
'title': title, 'title': title,
'description': description, 'description': description,
'duration': duration, 'duration': duration,
'timestamp': timestamp,
'thumbnails': thumbnails,
'formats': formats,
} }