yt_dlp/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_str,
   9     compat_urllib_parse_unquote,
  10 )
  11 from ..utils import (
  12     int_or_none,
  13     merge_dicts,
  14     OnDemandPagedList,
  15     parse_duration,
  16     parse_iso8601,
  17     parse_qs,
  18     try_get,
  19     update_url_query,
  20     urljoin,
  21 )
  22
  23
  24 class NBACVPBaseIE(TurnerBaseIE):
  25     def _extract_nba_cvp_info(self, path, video_id, fatal=False):
  26         return self._extract_cvp_info(
  27             'http://secure.nba.com/%s' % path, video_id, {
  28                 'default': {
  29                     'media_src': 'http://nba.cdn.turner.com/nba/big',
  30                 },
  31                 'm3u8': {
  32                     'media_src': 'http://nbavod-f.akamaihd.net',
  33                 },
  34             }, fatal=fatal)
  35
  36
  37 class NBAWatchBaseIE(NBACVPBaseIE):
  38     _VALID_URL_BASE = r'https?://(?:(?:www\.)?nba\.com(?:/watch)?|watch\.nba\.com)/'
  39
  40     def _extract_video(self, filter_key, filter_value):
  41         video = self._download_json(
  42             'https://neulionscnbav2-a.akamaihd.net/solr/nbad_program/usersearch',
  43             filter_value, query={
  44                 'fl': 'description,image,name,pid,releaseDate,runtime,tags,seoName',
  45                 'q': filter_key + ':' + filter_value,
  46                 'wt': 'json',
  47             })['response']['docs'][0]
  48
  49         video_id = str(video['pid'])
  50         title = video['name']
  51
  52         formats = []
  53         m3u8_url = (self._download_json(
  54             'https://watch.nba.com/service/publishpoint', video_id, query={
  55                 'type': 'video',
  56                 'format': 'json',
  57                 'id': video_id,
  58             }, headers={
  59                 'User-Agent': 'Mozilla/5.0 (iPhone; CPU iPhone OS 11_0_1 like Mac OS X) AppleWebKit/604.1.38 (KHTML, like Gecko) Version/11.0 Mobile/15A402 Safari/604.1',
  60             }, fatal=False) or {}).get('path')
  61         if m3u8_url:
  62             m3u8_formats = self._extract_m3u8_formats(
  63                 re.sub(r'_(?:pc|iphone)\.', '.', m3u8_url), video_id, 'mp4',
  64                 'm3u8_native', m3u8_id='hls', fatal=False)
  65             formats.extend(m3u8_formats)
  66             for f in m3u8_formats:
  67                 http_f = f.copy()
  68                 http_f.update({
  69                     'format_id': http_f['format_id'].replace('hls-', 'http-'),
  70                     'protocol': 'http',
  71                     'url': http_f['url'].replace('.m3u8', ''),
  72                 })
  73                 formats.append(http_f)
  74
  75         info = {
  76             'id': video_id,
  77             'title': title,
  78             'thumbnail': urljoin('https://nbadsdmt.akamaized.net/media/nba/nba/thumbs/', video.get('image')),
  79             'description': video.get('description'),
  80             'duration': int_or_none(video.get('runtime')),
  81             'timestamp': parse_iso8601(video.get('releaseDate')),
  82             'tags': video.get('tags'),
  83         }
  84
  85         seo_name = video.get('seoName')
  86         if seo_name and re.search(r'\d{4}/\d{2}/\d{2}/', seo_name):
  87             base_path = ''
  88             if seo_name.startswith('teams/'):
  89                 base_path += seo_name.split('/')[1] + '/'
  90             base_path += 'video/'
  91             cvp_info = self._extract_nba_cvp_info(
  92                 base_path + seo_name + '.xml', video_id, False)
  93             if cvp_info:
  94                 formats.extend(cvp_info['formats'])
  95                 info = merge_dicts(info, cvp_info)
  96
  97         self._sort_formats(formats)
  98         info['formats'] = formats
  99         return info
 100
 101
 102 class NBAWatchEmbedIE(NBAWatchBaseIE):
 103     IENAME = 'nba:watch:embed'
 104     _VALID_URL = NBAWatchBaseIE._VALID_URL_BASE + r'embed\?.*?\bid=(?P<id>\d+)'
 105     _TESTS = [{
 106         'url': 'http://watch.nba.com/embed?id=659395',
 107         'md5': 'b7e3f9946595f4ca0a13903ce5edd120',
 108         'info_dict': {
 109             'id': '659395',
 110             'ext': 'mp4',
 111             'title': 'Mix clip: More than 7 points of  Joe Ingles, Luc Mbah a Moute, Blake Griffin and 6 more in Utah Jazz vs. the Clippers, 4/15/2017',
 112             'description': 'Mix clip: More than 7 points of  Joe Ingles, Luc Mbah a Moute, Blake Griffin and 6 more in Utah Jazz vs. the Clippers, 4/15/2017',
 113             'timestamp': 1492228800,
 114             'upload_date': '20170415',
 115         },
 116     }]
 117
 118     def _real_extract(self, url):
 119         video_id = self._match_id(url)
 120         return self._extract_video('pid', video_id)
 121
 122
 123 class NBAWatchIE(NBAWatchBaseIE):
 124     IE_NAME = 'nba:watch'
 125     _VALID_URL = NBAWatchBaseIE._VALID_URL_BASE + r'(?:nba/)?video/(?P<id>.+?(?=/index\.html)|(?:[^/]+/)*[^/?#&]+)'
 126     _TESTS = [{
 127         'url': 'http://www.nba.com/video/games/nets/2012/12/04/0021200253-okc-bkn-recap.nba/index.html',
 128         'md5': '9d902940d2a127af3f7f9d2f3dc79c96',
 129         'info_dict': {
 130             'id': '70946',
 131             'ext': 'mp4',
 132             'title': 'Thunder vs. Nets',
 133             'description': 'Kevin Durant scores 32 points and dishes out six assists as the Thunder beat the Nets in Brooklyn.',
 134             'duration': 181,
 135             'timestamp': 1354597200,
 136             'upload_date': '20121204',
 137         },
 138     }, {
 139         'url': 'http://www.nba.com/video/games/hornets/2014/12/05/0021400276-nyk-cha-play5.nba/',
 140         'only_matching': True,
 141     }, {
 142         'url': 'http://watch.nba.com/video/channels/playoffs/2015/05/20/0041400301-cle-atl-recap.nba',
 143         'md5': 'b2b39b81cf28615ae0c3360a3f9668c4',
 144         'info_dict': {
 145             'id': '330865',
 146             'ext': 'mp4',
 147             'title': 'Hawks vs. Cavaliers Game 1',
 148             'description': 'md5:8094c3498d35a9bd6b1a8c396a071b4d',
 149             'duration': 228,
 150             'timestamp': 1432094400,
 151             'upload_date': '20150521',
 152         },
 153     }, {
 154         'url': 'http://watch.nba.com/nba/video/channels/nba_tv/2015/06/11/YT_go_big_go_home_Game4_061115',
 155         'only_matching': True,
 156     }, {
 157         # only CVP mp4 format available
 158         'url': 'https://watch.nba.com/video/teams/cavaliers/2012/10/15/sloan121015mov-2249106',
 159         'only_matching': True,
 160     }, {
 161         'url': 'https://watch.nba.com/video/top-100-dunks-from-the-2019-20-season?plsrc=nba&collection=2019-20-season-highlights',
 162         'only_matching': True,
 163     }]
 164
 165     def _real_extract(self, url):
 166         display_id = self._match_id(url)
 167         collection_id = parse_qs(url).get('collection', [None])[0]
 168         if collection_id:
 169             if self.get_param('noplaylist'):
 170                 self.to_screen('Downloading just video %s because of --no-playlist' % display_id)
 171             else:
 172                 self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % collection_id)
 173                 return self.url_result(
 174                     'https://www.nba.com/watch/list/collection/' + collection_id,
 175                     NBAWatchCollectionIE.ie_key(), collection_id)
 176         return self._extract_video('seoName', display_id)
 177
 178
 179 class NBAWatchCollectionIE(NBAWatchBaseIE):
 180     IE_NAME = 'nba:watch:collection'
 181     _VALID_URL = NBAWatchBaseIE._VALID_URL_BASE + r'list/collection/(?P<id>[^/?#&]+)'
 182     _TESTS = [{
 183         'url': 'https://watch.nba.com/list/collection/season-preview-2020',
 184         'info_dict': {
 185             'id': 'season-preview-2020',
 186         },
 187         'playlist_mincount': 43,
 188     }]
 189     _PAGE_SIZE = 100
 190
 191     def _fetch_page(self, collection_id, page):
 192         page += 1
 193         videos = self._download_json(
 194             'https://content-api-prod.nba.com/public/1/endeavor/video-list/collection/' + collection_id,
 195             collection_id, 'Downloading page %d JSON metadata' % page, query={
 196                 'count': self._PAGE_SIZE,
 197                 'page': page,
 198             })['results']['videos']
 199         for video in videos:
 200             program = video.get('program') or {}
 201             seo_name = program.get('seoName') or program.get('slug')
 202             if not seo_name:
 203                 continue
 204             yield {
 205                 '_type': 'url',
 206                 'id': program.get('id'),
 207                 'title': program.get('title') or video.get('title'),
 208                 'url': 'https://www.nba.com/watch/video/' + seo_name,
 209                 'thumbnail': video.get('image'),
 210                 'description': program.get('description') or video.get('description'),
 211                 'duration': parse_duration(program.get('runtimeHours')),
 212                 'timestamp': parse_iso8601(video.get('releaseDate')),
 213             }
 214
 215     def _real_extract(self, url):
 216         collection_id = self._match_id(url)
 217         entries = OnDemandPagedList(
 218             functools.partial(self._fetch_page, collection_id),
 219             self._PAGE_SIZE)
 220         return self.playlist_result(entries, collection_id)
 221
 222
 223 class NBABaseIE(NBACVPBaseIE):
 224     _VALID_URL_BASE = r'''(?x)
 225         https?://(?:www\.)?nba\.com/
 226             (?P<team>
 227                 blazers|
 228                 bucks|
 229                 bulls|
 230                 cavaliers|
 231                 celtics|
 232                 clippers|
 233                 grizzlies|
 234                 hawks|
 235                 heat|
 236                 hornets|
 237                 jazz|
 238                 kings|
 239                 knicks|
 240                 lakers|
 241                 magic|
 242                 mavericks|
 243                 nets|
 244                 nuggets|
 245                 pacers|
 246                 pelicans|
 247                 pistons|
 248                 raptors|
 249                 rockets|
 250                 sixers|
 251                 spurs|
 252                 suns|
 253                 thunder|
 254                 timberwolves|
 255                 warriors|
 256                 wizards
 257             )
 258         (?:/play\#)?/'''
 259     _CHANNEL_PATH_REGEX = r'video/channel|series'
 260
 261     def _embed_url_result(self, team, content_id):
 262         return self.url_result(update_url_query(
 263             'https://secure.nba.com/assets/amp/include/video/iframe.html', {
 264                 'contentId': content_id,
 265                 'team': team,
 266             }), NBAEmbedIE.ie_key())
 267
 268     def _call_api(self, team, content_id, query, resource):
 269         return self._download_json(
 270             'https://api.nba.net/2/%s/video,imported_video,wsc/' % team,
 271             content_id, 'Download %s JSON metadata' % resource,
 272             query=query, headers={
 273                 'accessToken': 'internal|bb88df6b4c2244e78822812cecf1ee1b',
 274             })['response']['result']
 275
 276     def _extract_video(self, video, team, extract_all=True):
 277         video_id = compat_str(video['nid'])
 278         team = video['brand']
 279
 280         info = {
 281             'id': video_id,
 282             'title': video.get('title') or video.get('headline') or video['shortHeadline'],
 283             'description': video.get('description'),
 284             'timestamp': parse_iso8601(video.get('published')),
 285         }
 286
 287         subtitles = {}
 288         captions = try_get(video, lambda x: x['videoCaptions']['sidecars'], dict) or {}
 289         for caption_url in captions.values():
 290             subtitles.setdefault('en', []).append({'url': caption_url})
 291
 292         formats = []
 293         mp4_url = video.get('mp4')
 294         if mp4_url:
 295             formats.append({
 296                 'url': mp4_url,
 297             })
 298
 299         if extract_all:
 300             source_url = video.get('videoSource')
 301             if source_url and not source_url.startswith('s3://') and self._is_valid_url(source_url, video_id, 'source'):
 302                 formats.append({
 303                     'format_id': 'source',
 304                     'url': source_url,
 305                     'quality': 1,
 306                 })
 307
 308             m3u8_url = video.get('m3u8')
 309             if m3u8_url:
 310                 if '.akamaihd.net/i/' in m3u8_url:
 311                     formats.extend(self._extract_akamai_formats(
 312                         m3u8_url, video_id, {'http': 'pmd.cdn.turner.com'}))
 313                 else:
 314                     formats.extend(self._extract_m3u8_formats(
 315                         m3u8_url, video_id, 'mp4',
 316                         'm3u8_native', m3u8_id='hls', fatal=False))
 317
 318             content_xml = video.get('contentXml')
 319             if team and content_xml:
 320                 cvp_info = self._extract_nba_cvp_info(
 321                     team + content_xml, video_id, fatal=False)
 322                 if cvp_info:
 323                     formats.extend(cvp_info['formats'])
 324                     subtitles = self._merge_subtitles(subtitles, cvp_info['subtitles'])
 325                     info = merge_dicts(info, cvp_info)
 326
 327             self._sort_formats(formats)
 328         else:
 329             info.update(self._embed_url_result(team, video['videoId']))
 330
 331         info.update({
 332             'formats': formats,
 333             'subtitles': subtitles,
 334         })
 335
 336         return info
 337
 338     def _real_extract(self, url):
 339         team, display_id = self._match_valid_url(url).groups()
 340         if '/play#/' in url:
 341             display_id = compat_urllib_parse_unquote(display_id)
 342         else:
 343             webpage = self._download_webpage(url, display_id)
 344             display_id = self._search_regex(
 345                 self._CONTENT_ID_REGEX + r'\s*:\s*"([^"]+)"', webpage, 'video id')
 346         return self._extract_url_results(team, display_id)
 347
 348
 349 class NBAEmbedIE(NBABaseIE):
 350     IENAME = 'nba:embed'
 351     _VALID_URL = r'https?://secure\.nba\.com/assets/amp/include/video/(?:topI|i)frame\.html\?.*?\bcontentId=(?P<id>[^?#&]+)'
 352     _TESTS = [{
 353         'url': 'https://secure.nba.com/assets/amp/include/video/topIframe.html?contentId=teams/bulls/2020/12/04/3478774/1607105587854-20201204_SCHEDULE_RELEASE_FINAL_DRUPAL-3478774&team=bulls&adFree=false&profile=71&videoPlayerName=TAMPCVP&baseUrl=&videoAdsection=nba.com_mobile_web_teamsites_chicagobulls&ampEnv=',
 354         'only_matching': True,
 355     }, {
 356         'url': 'https://secure.nba.com/assets/amp/include/video/iframe.html?contentId=2016/10/29/0021600027boschaplay7&adFree=false&profile=71&team=&videoPlayerName=LAMPCVP',
 357         'only_matching': True,
 358     }]
 359
 360     def _real_extract(self, url):
 361         qs = parse_qs(url)
 362         content_id = qs['contentId'][0]
 363         team = qs.get('team', [None])[0]
 364         if not team:
 365             return self.url_result(
 366                 'https://watch.nba.com/video/' + content_id, NBAWatchIE.ie_key())
 367         video = self._call_api(team, content_id, {'videoid': content_id}, 'video')[0]
 368         return self._extract_video(video, team)
 369
 370
 371 class NBAIE(NBABaseIE):
 372     IENAME = 'nba'
 373     _VALID_URL = NBABaseIE._VALID_URL_BASE + '(?!%s)video/(?P<id>(?:[^/]+/)*[^/?#&]+)' % NBABaseIE._CHANNEL_PATH_REGEX
 374     _TESTS = [{
 375         'url': 'https://www.nba.com/bulls/video/teams/bulls/2020/12/04/3478774/1607105587854-20201204schedulereleasefinaldrupal-3478774',
 376         'info_dict': {
 377             'id': '45039',
 378             'ext': 'mp4',
 379             'title': 'AND WE BACK.',
 380             'description': 'Part 1 of our 2020-21 schedule is here! Watch our games on NBC Sports Chicago.',
 381             'duration': 94,
 382             'timestamp': 1607112000,
 383             'upload_date': '20201218',
 384         },
 385     }, {
 386         'url': 'https://www.nba.com/bucks/play#/video/teams%2Fbucks%2F2020%2F12%2F17%2F64860%2F1608252863446-Op_Dream_16x9-64860',
 387         'only_matching': True,
 388     }, {
 389         'url': 'https://www.nba.com/bucks/play#/video/wsc%2Fteams%2F2787C911AA1ACD154B5377F7577CCC7134B2A4B0',
 390         'only_matching': True,
 391     }]
 392     _CONTENT_ID_REGEX = r'videoID'
 393
 394     def _extract_url_results(self, team, content_id):
 395         return self._embed_url_result(team, content_id)
 396
 397
 398 class NBAChannelIE(NBABaseIE):
 399     IENAME = 'nba:channel'
 400     _VALID_URL = NBABaseIE._VALID_URL_BASE + '(?:%s)/(?P<id>[^/?#&]+)' % NBABaseIE._CHANNEL_PATH_REGEX
 401     _TESTS = [{
 402         'url': 'https://www.nba.com/blazers/video/channel/summer_league',
 403         'info_dict': {
 404             'title': 'Summer League',
 405         },
 406         'playlist_mincount': 138,
 407     }, {
 408         'url': 'https://www.nba.com/bucks/play#/series/On%20This%20Date',
 409         'only_matching': True,
 410     }]
 411     _CONTENT_ID_REGEX = r'videoSubCategory'
 412     _PAGE_SIZE = 100
 413
 414     def _fetch_page(self, team, channel, page):
 415         results = self._call_api(team, channel, {
 416             'channels': channel,
 417             'count': self._PAGE_SIZE,
 418             'offset': page * self._PAGE_SIZE,
 419         }, 'page %d' % (page + 1))
 420         for video in results:
 421             yield self._extract_video(video, team, False)
 422
 423     def _extract_url_results(self, team, content_id):
 424         entries = OnDemandPagedList(
 425             functools.partial(self._fetch_page, team, content_id),
 426             self._PAGE_SIZE)
 427         return self.playlist_result(entries, playlist_title=content_id)