aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorremitamine <remitamine@gmail.com>2015-12-04 07:20:27 +0100
committerremitamine <remitamine@gmail.com>2015-12-04 07:20:27 +0100
commit9c117d345ff05da2d4b107881cc387657208b76e (patch)
tree636a770c85684c357575d63d8154e7f6b6b141c2
parent71d9fe7818b8c766c3b1190017746d8e51db3081 (diff)
parent46cc1c65a4b5c57555f2549534788ad004c46c03 (diff)
downloadyoutube-dl-9c117d345ff05da2d4b107881cc387657208b76e.tar.xz
[nba] improve(fixes #7068)
* extract more formats * extract videos from team mini sites * extract more metadata
-rw-r--r--youtube_dl/extractor/nba.py83
1 files changed, 56 insertions, 27 deletions
diff --git a/youtube_dl/extractor/nba.py b/youtube_dl/extractor/nba.py
index 944096e1c..77a3b49ef 100644
--- a/youtube_dl/extractor/nba.py
+++ b/youtube_dl/extractor/nba.py
@@ -1,63 +1,92 @@
from __future__ import unicode_literals
+import re
+
from .common import InfoExtractor
from ..utils import (
- remove_end,
parse_duration,
+ int_or_none,
+ xpath_text,
+ xpath_attr,
)
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 = [{
'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
- 'md5': 'c0edcfc37607344e2ff8f13c378c88a4',
+ 'md5': '9e7729d3010a9c71506fd1248f74e4f4',
'info_dict': {
- 'id': '0021200253-okc-bkn-recap.nba',
- 'ext': 'mp4',
+ 'id': '0021200253-okc-bkn-recap',
+ 'ext': 'flv',
'title': 'Thunder vs. Nets',
'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
'duration': 181,
+ 'timestamp': 1354638466,
+ 'upload_date': '20121204',
},
}, {
'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
'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': {
- 'id': '0041400301-cle-atl-recap.nba',
+ 'id': '0041400301-cle-atl-recap',
'ext': 'mp4',
- 'title': 'NBA GAME TIME | Video: Hawks vs. Cavaliers Game 1',
+ 'title': 'Hawks vs. Cavaliers Game 1',
'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
'duration': 228,
- },
- 'params': {
- 'skip_download': True,
+ 'timestamp': 1432134543,
+ 'upload_date': '20150520',
}
}]
def _real_extract(self, url):
- video_id = self._match_id(url)
- webpage = self._download_webpage(url, video_id)
-
- video_url = 'http://ht-mobile.cdn.turner.com/nba/big' + video_id + '_nba_1280x720.mp4'
+ path, video_id = re.match(self._VALID_URL, url).groups()
+ 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'))
- shortened_video_id = video_id.rpartition('/')[2]
- title = remove_end(
- self._og_search_title(webpage, default=shortened_video_id), ' : NBA.com')
+ 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')),
+ })
- description = self._og_search_description(webpage)
- duration_str = self._html_search_meta(
- 'duration', webpage, 'duration', default=None)
- if not duration_str:
- duration_str = self._html_search_regex(
- r'Duration:</b>\s*(\d+:\d+)', webpage, 'duration', fatal=False)
- duration = parse_duration(duration_str)
+ formats = []
+ for video_file in video_info.find('files').iter('file'):
+ video_url = video_file.text
+ if video_url.startswith('/'):
+ continue
+ if video_url.endswith('.m3u8'):
+ formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls'))
+ elif video_url.endswith('.f4m'):
+ formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds'))
+ else:
+ 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 {
- 'id': shortened_video_id,
- 'url': video_url,
+ 'id': video_id,
'title': title,
'description': description,
'duration': duration,
+ 'timestamp': timestamp,
+ 'thumbnails': thumbnails,
+ 'formats': formats,
}