]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/nba.py
Merge tag 'upstream/2016.02.22'
[youtubedl] / youtube_dl / extractor / nba.py
1 from __future__ import unicode_literals
2
3 import re
4
5 from .common import InfoExtractor
6 from ..utils import (
7 parse_duration,
8 int_or_none,
9 xpath_text,
10 xpath_attr,
11 )
12
13
14 class NBAIE(InfoExtractor):
15 _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?P<path>(?:[^/]+/)?video/(?P<id>[^?]*?))/?(?:/index\.html)?(?:\?.*)?$'
16 _TESTS = [{
17 'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
18 'md5': '9e7729d3010a9c71506fd1248f74e4f4',
19 'info_dict': {
20 'id': '0021200253-okc-bkn-recap',
21 'ext': 'mp4',
22 'title': 'Thunder vs. Nets',
23 'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
24 'duration': 181,
25 'timestamp': 1354638466,
26 'upload_date': '20121204',
27 },
28 'params': {
29 # m3u8 download
30 'skip_download': True,
31 },
32 }, {
33 'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
34 'only_matching': True,
35 }, {
36 'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
37 'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
38 'info_dict': {
39 'id': '0041400301-cle-atl-recap',
40 'ext': 'mp4',
41 'title': 'Hawks vs. Cavaliers Game 1',
42 'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
43 'duration': 228,
44 'timestamp': 1432134543,
45 'upload_date': '20150520',
46 }
47 }]
48
49 def _real_extract(self, url):
50 path, video_id = re.match(self._VALID_URL, url).groups()
51 if path.startswith('nba/'):
52 path = path[3:]
53 video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id)
54 video_id = xpath_text(video_info, 'slug')
55 title = xpath_text(video_info, 'headline')
56 description = xpath_text(video_info, 'description')
57 duration = parse_duration(xpath_text(video_info, 'length'))
58 timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts'))
59
60 thumbnails = []
61 for image in video_info.find('images'):
62 thumbnails.append({
63 'id': image.attrib.get('cut'),
64 'url': image.text,
65 'width': int_or_none(image.attrib.get('width')),
66 'height': int_or_none(image.attrib.get('height')),
67 })
68
69 formats = []
70 for video_file in video_info.findall('.//file'):
71 video_url = video_file.text
72 if video_url.startswith('/'):
73 continue
74 if video_url.endswith('.m3u8'):
75 formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
76 elif video_url.endswith('.f4m'):
77 formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds', fatal=False))
78 else:
79 key = video_file.attrib.get('bitrate')
80 format_info = {
81 'format_id': key,
82 'url': video_url,
83 }
84 mobj = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key)
85 if mobj:
86 format_info.update({
87 'width': int(mobj.group(1)),
88 'height': int(mobj.group(2)),
89 'tbr': int_or_none(mobj.group(3)),
90 })
91 formats.append(format_info)
92 self._sort_formats(formats)
93
94 return {
95 'id': video_id,
96 'title': title,
97 'description': description,
98 'duration': duration,
99 'timestamp': timestamp,
100 'thumbnails': thumbnails,
101 'formats': formats,
102 }