]> jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/nba.py
[nba] use xpath utils
[yt-dlp.git] / 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': 'flv',
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 }, {
29 'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
30 'only_matching': True,
31 }, {
32 'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
33 'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
34 'info_dict': {
35 'id': '0041400301-cle-atl-recap',
36 'ext': 'mp4',
37 'title': 'Hawks vs. Cavaliers Game 1',
38 'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
39 'duration': 228,
40 'timestamp': 1432134543,
41 'upload_date': '20150520',
42 }
43 }]
44
45 def _real_extract(self, url):
46 path, video_id = re.match(self._VALID_URL, url).groups()
47 video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id)
48 video_id = xpath_text(video_info, 'slug')
49 title = xpath_text(video_info, 'headline')
50 description = xpath_text(video_info, 'description')
51 duration = parse_duration(xpath_text(video_info, 'length'))
52 timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts'))
53
54 thumbnails = []
55 for image in video_info.find('images'):
56 thumbnails.append({
57 'id': image.attrib.get('cut'),
58 'url': image.text,
59 'width': int_or_none(image.attrib.get('width')),
60 'height': int_or_none(image.attrib.get('height')),
61 })
62
63 formats = []
64 for video_file in video_info.find('files').iter('file'):
65 video_url = video_file.text
66 if video_url.startswith('/'):
67 continue
68 if video_url.endswith('.m3u8'):
69 formats.extend(self._extract_m3u8_formats(video_url, video_id, m3u8_id='hls'))
70 elif video_url.endswith('.f4m'):
71 formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds'))
72 else:
73 key = video_file.attrib.get('bitrate')
74 width, height, bitrate = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key).groups()
75 formats.append({
76 'format_id': key,
77 'url': video_url,
78 'width': int_or_none(width),
79 'height': int_or_none(height),
80 'tbr': int_or_none(bitrate),
81 })
82 self._sort_formats(formats)
83
84 return {
85 'id': video_id,
86 'title': title,
87 'description': description,
88 'duration': duration,
89 'timestamp': timestamp,
90 'thumbnails': thumbnails,
91 'formats': formats,
92 }