youtube_dl/extractor/nba.py

   1 from __future__ import unicode_literals
   2
   3 import functools
   4 import re
   5
   6 from .turner import TurnerBaseIE
   7 from ..compat import (
   8     compat_urllib_parse_urlencode,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     OnDemandPagedList,
  13     remove_start,
  14 )
  15
  16
  17 class NBAIE(TurnerBaseIE):
  18     _VALID_URL = r'https?://(?:watch\.|www\.)?nba\.com/(?P<path>(?:[^/]+/)+(?P<id>[^?]*?))/?(?:/index\.html)?(?:\?.*)?$'
  19     _TESTS = [{
  20         'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
  21         'md5': '9e7729d3010a9c71506fd1248f74e4f4',
  22         'info_dict': {
  23             'id': '0021200253-okc-bkn-recap',
  24             'ext': 'mp4',
  25             'title': 'Thunder vs. Nets',
  26             'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
  27             'duration': 181,
  28             'timestamp': 1354638466,
  29             'upload_date': '20121204',
  30         },
  31         'params': {
  32             # m3u8 download
  33             'skip_download': True,
  34         },
  35     }, {
  36         'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
  40         'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
  41         'info_dict': {
  42             'id': 'channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
  43             'ext': 'mp4',
  44             'title': 'Hawks vs. Cavaliers Game 1',
  45             'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
  46             'duration': 228,
  47             'timestamp': 1432134543,
  48             'upload_date': '20150520',
  49         },
  50         'expected_warnings': ['Unable to download f4m manifest'],
  51     }, {
  52         'url': 'http://www.nba.com/clippers/news/doc-rivers-were-not-trading-blake',
  53         'info_dict': {
  54             'id': 'teams/clippers/2016/02/17/1455672027478-Doc_Feb16_720.mov-297324',
  55             'ext': 'mp4',
  56             'title': 'Practice: Doc Rivers - 2/16/16',
  57             'description': 'Head Coach Doc Rivers addresses the media following practice.',
  58             'upload_date': '20160216',
  59             'timestamp': 1455672000,
  60         },
  61         'params': {
  62             # m3u8 download
  63             'skip_download': True,
  64         },
  65         'expected_warnings': ['Unable to download f4m manifest'],
  66     }, {
  67         'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#',
  68         'info_dict': {
  69             'id': 'timberwolves',
  70             'title': 'Shootaround Access - Dec. 12 | Andrew Wiggins',
  71         },
  72         'playlist_count': 30,
  73         'params': {
  74             # Download the whole playlist takes too long time
  75             'playlist_items': '1-30',
  76         },
  77     }, {
  78         'url': 'http://www.nba.com/timberwolves/wiggins-shootaround#',
  79         'info_dict': {
  80             'id': 'teams/timberwolves/2014/12/12/Wigginsmp4-3462601',
  81             'ext': 'mp4',
  82             'title': 'Shootaround Access - Dec. 12 | Andrew Wiggins',
  83             'description': 'Wolves rookie Andrew Wiggins addresses the media after Friday\'s shootaround.',
  84             'upload_date': '20141212',
  85             'timestamp': 1418418600,
  86         },
  87         'params': {
  88             'noplaylist': True,
  89             # m3u8 download
  90             'skip_download': True,
  91         },
  92         'expected_warnings': ['Unable to download f4m manifest'],
  93     }]
  94
  95     _PAGE_SIZE = 30
  96
  97     def _fetch_page(self, team, video_id, page):
  98         search_url = 'http://searchapp2.nba.com/nba-search/query.jsp?' + compat_urllib_parse_urlencode({
  99             'type': 'teamvideo',
 100             'start': page * self._PAGE_SIZE + 1,
 101             'npp': (page + 1) * self._PAGE_SIZE + 1,
 102             'sort': 'recent',
 103             'output': 'json',
 104             'site': team,
 105         })
 106         results = self._download_json(
 107             search_url, video_id, note='Download page %d of playlist data' % page)['results'][0]
 108         for item in results:
 109             yield self.url_result(compat_urlparse.urljoin('http://www.nba.com/', item['url']))
 110
 111     def _extract_playlist(self, orig_path, video_id, webpage):
 112         team = orig_path.split('/')[0]
 113
 114         if self._downloader.params.get('noplaylist'):
 115             self.to_screen('Downloading just video because of --no-playlist')
 116             video_path = self._search_regex(
 117                 r'nbaVideoCore\.firstVideo\s*=\s*\'([^\']+)\';', webpage, 'video path')
 118             video_url = 'http://www.nba.com/%s/video/%s' % (team, video_path)
 119             return self.url_result(video_url)
 120
 121         self.to_screen('Downloading playlist - add --no-playlist to just download video')
 122         playlist_title = self._og_search_title(webpage, fatal=False)
 123         entries = OnDemandPagedList(
 124             functools.partial(self._fetch_page, team, video_id),
 125             self._PAGE_SIZE)
 126
 127         return self.playlist_result(entries, team, playlist_title)
 128
 129     def _real_extract(self, url):
 130         path, video_id = re.match(self._VALID_URL, url).groups()
 131         orig_path = path
 132         if path.startswith('nba/'):
 133             path = path[3:]
 134
 135         if 'video/' not in path:
 136             webpage = self._download_webpage(url, video_id)
 137             path = remove_start(self._search_regex(r'data-videoid="([^"]+)"', webpage, 'video id'), '/')
 138
 139             if path == '{{id}}':
 140                 return self._extract_playlist(orig_path, video_id, webpage)
 141
 142             # See prepareContentId() of pkgCvp.js
 143             if path.startswith('video/teams'):
 144                 path = 'video/channels/proxy/' + path[6:]
 145
 146         return self._extract_cvp_info(
 147             'http://www.nba.com/%s.xml' % path, video_id, {
 148                 'default': {
 149                     'media_src': 'http://nba.cdn.turner.com/nba/big',
 150                 },
 151                 'm3u8': {
 152                     'media_src': 'http://nbavod-f.akamaihd.net',
 153                 },
 154             })