_ Git - youtube-dl/blob - youtube_dl/extractor/nba.py

   1 from __future__ import unicode_literals
   2
   3 import os.path
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     parse_duration,
  10     remove_start,
  11     xpath_text,
  12     xpath_attr,
  13 )
  14
  15
  16 class NBAIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?P<path>(?:[^/]+/)+(?P<id>[^?]*?))/?(?:/index\.html)?(?:\?.*)?$'
  18     _TESTS = [{
  19         'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
  20         'md5': '9e7729d3010a9c71506fd1248f74e4f4',
  21         'info_dict': {
  22             'id': '0021200253-okc-bkn-recap',
  23             'ext': 'mp4',
  24             'title': 'Thunder vs. Nets',
  25             'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
  26             'duration': 181,
  27             'timestamp': 1354638466,
  28             'upload_date': '20121204',
  29         },
  30         'params': {
  31             # m3u8 download
  32             'skip_download': True,
  33         },
  34     }, {
  35         'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
  36         'only_matching': True,
  37     }, {
  38         'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
  39         'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
  40         'info_dict': {
  41             'id': '0041400301-cle-atl-recap',
  42             'ext': 'mp4',
  43             'title': 'Hawks vs. Cavaliers Game 1',
  44             'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
  45             'duration': 228,
  46             'timestamp': 1432134543,
  47             'upload_date': '20150520',
  48         }
  49     }, {
  50         'url': 'http://www.nba.com/clippers/news/doc-rivers-were-not-trading-blake',
  51         'info_dict': {
  52             'id': '1455672027478-Doc_Feb16_720',
  53             'ext': 'mp4',
  54             'title': 'Practice: Doc Rivers - 2/16/16',
  55             'description': 'Head Coach Doc Rivers addresses the media following practice.',
  56             'upload_date': '20160217',
  57             'timestamp': 1455672000,
  58         },
  59         'params': {
  60             # m3u8 download
  61             'skip_download': True,
  62         },
  63     }]
  64
  65     def _real_extract(self, url):
  66         path, video_id = re.match(self._VALID_URL, url).groups()
  67         if path.startswith('nba/'):
  68             path = path[3:]
  69
  70         if 'video/' not in path:
  71             webpage = self._download_webpage(url, video_id)
  72             path = remove_start(self._search_regex(r'data-videoid="([^"]+)"', webpage, 'video id'), '/')
  73             # See prepareContentId() of pkgCvp.js
  74             if path.startswith('video/teams'):
  75                 path = 'video/channels/proxy/' + path[6:]
  76
  77         video_info = self._download_xml('http://www.nba.com/%s.xml' % path, video_id)
  78         video_id = os.path.splitext(xpath_text(video_info, 'slug'))[0]
  79         title = xpath_text(video_info, 'headline')
  80         description = xpath_text(video_info, 'description')
  81         duration = parse_duration(xpath_text(video_info, 'length'))
  82         timestamp = int_or_none(xpath_attr(video_info, 'dateCreated', 'uts'))
  83
  84         thumbnails = []
  85         for image in video_info.find('images'):
  86             thumbnails.append({
  87                 'id': image.attrib.get('cut'),
  88                 'url': image.text,
  89                 'width': int_or_none(image.attrib.get('width')),
  90                 'height': int_or_none(image.attrib.get('height')),
  91             })
  92
  93         formats = []
  94         for video_file in video_info.findall('.//file'):
  95             video_url = video_file.text
  96             if video_url.startswith('/'):
  97                 continue
  98             if video_url.endswith('.m3u8'):
  99                 formats.extend(self._extract_m3u8_formats(video_url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
 100             elif video_url.endswith('.f4m'):
 101                 formats.extend(self._extract_f4m_formats(video_url + '?hdcore=3.4.1.1', video_id, f4m_id='hds', fatal=False))
 102             else:
 103                 key = video_file.attrib.get('bitrate')
 104                 format_info = {
 105                     'format_id': key,
 106                     'url': video_url,
 107                 }
 108                 mobj = re.search(r'(\d+)x(\d+)(?:_(\d+))?', key)
 109                 if mobj:
 110                     format_info.update({
 111                         'width': int(mobj.group(1)),
 112                         'height': int(mobj.group(2)),
 113                         'tbr': int_or_none(mobj.group(3)),
 114                     })
 115                 formats.append(format_info)
 116         self._sort_formats(formats)
 117
 118         return {
 119             'id': video_id,
 120             'title': title,
 121             'description': description,
 122             'duration': duration,
 123             'timestamp': timestamp,
 124             'thumbnails': thumbnails,
 125             'formats': formats,
 126         }