yt_dlp/extractor/nfl.py

   1 import base64
   2 import json
   3 import re
   4 import time
   5 import uuid
   6
   7 from .anvato import AnvatoIE
   8 from .common import InfoExtractor
   9 from ..utils import (
  10     ExtractorError,
  11     clean_html,
  12     determine_ext,
  13     get_element_by_class,
  14     traverse_obj,
  15     urlencode_postdata,
  16 )
  17
  18
  19 class NFLBaseIE(InfoExtractor):
  20     _VALID_URL_BASE = r'''(?x)
  21                     https?://
  22                         (?P<host>
  23                             (?:www\.)?
  24                             (?:
  25                                 (?:
  26                                     nfl|
  27                                     buffalobills|
  28                                     miamidolphins|
  29                                     patriots|
  30                                     newyorkjets|
  31                                     baltimoreravens|
  32                                     bengals|
  33                                     clevelandbrowns|
  34                                     steelers|
  35                                     houstontexans|
  36                                     colts|
  37                                     jaguars|
  38                                     (?:titansonline|tennesseetitans)|
  39                                     denverbroncos|
  40                                     (?:kc)?chiefs|
  41                                     raiders|
  42                                     chargers|
  43                                     dallascowboys|
  44                                     giants|
  45                                     philadelphiaeagles|
  46                                     (?:redskins|washingtonfootball)|
  47                                     chicagobears|
  48                                     detroitlions|
  49                                     packers|
  50                                     vikings|
  51                                     atlantafalcons|
  52                                     panthers|
  53                                     neworleanssaints|
  54                                     buccaneers|
  55                                     azcardinals|
  56                                     (?:stlouis|the)rams|
  57                                     49ers|
  58                                     seahawks
  59                                 )\.com|
  60                                 .+?\.clubs\.nfl\.com
  61                             )
  62                         )/
  63                     '''
  64     _VIDEO_CONFIG_REGEX = r'<script[^>]+id="[^"]*video-config-[0-9a-f]{8}-(?:[0-9a-f]{4}-){3}[0-9a-f]{12}[^"]*"[^>]*>\s*({.+});?\s*</script>'
  65     _ANVATO_PREFIX = 'anvato:GXvEgwyJeWem8KCYXfeoHWknwP48Mboj:'
  66
  67     _CLIENT_DATA = {
  68         'clientKey': '4cFUW6DmwJpzT9L7LrG3qRAcABG5s04g',
  69         'clientSecret': 'CZuvCL49d9OwfGsR',
  70         'deviceId': str(uuid.uuid4()),
  71         'deviceInfo': base64.b64encode(json.dumps({
  72             'model': 'desktop',
  73             'version': 'Chrome',
  74             'osName': 'Windows',
  75             'osVersion': '10.0',
  76         }, separators=(',', ':')).encode()).decode(),
  77         'networkType': 'other',
  78         'nflClaimGroupsToAdd': [],
  79         'nflClaimGroupsToRemove': [],
  80     }
  81     _ACCOUNT_INFO = {}
  82     _API_KEY = None
  83
  84     _TOKEN = None
  85     _TOKEN_EXPIRY = 0
  86
  87     def _get_account_info(self, url, slug):
  88         if not self._API_KEY:
  89             webpage = self._download_webpage(url, slug, fatal=False) or ''
  90             self._API_KEY = self._search_regex(
  91                 r'window\.gigyaApiKey\s*=\s*["\'](\w+)["\'];', webpage, 'API key',
  92                 fatal=False) or '3_Qa8TkWpIB8ESCBT8tY2TukbVKgO5F6BJVc7N1oComdwFzI7H2L9NOWdm11i_BY9f'
  93
  94         cookies = self._get_cookies('https://auth-id.nfl.com/')
  95         login_token = traverse_obj(cookies, (
  96             (f'glt_{self._API_KEY}', lambda k, _: k.startswith('glt_')), {lambda x: x.value}), get_all=False)
  97         if not login_token:
  98             self.raise_login_required()
  99         if 'ucid' not in cookies:
 100             raise ExtractorError(
 101                 'Required cookies for the auth-id.nfl.com domain were not found among passed cookies. '
 102                 'If using --cookies, these cookies must be exported along with .nfl.com cookies, '
 103                 'or else try using --cookies-from-browser instead', expected=True)
 104
 105         account = self._download_json(
 106             'https://auth-id.nfl.com/accounts.getAccountInfo', slug,
 107             note='Downloading account info', data=urlencode_postdata({
 108                 'include': 'profile,data',
 109                 'lang': 'en',
 110                 'APIKey': self._API_KEY,
 111                 'sdk': 'js_latest',
 112                 'login_token': login_token,
 113                 'authMode': 'cookie',
 114                 'pageURL': url,
 115                 'sdkBuild': traverse_obj(cookies, (
 116                     'gig_canary_ver', {lambda x: x.value.partition('-')[0]}), default='15170'),
 117                 'format': 'json',
 118             }), headers={'Content-Type': 'application/x-www-form-urlencoded'})
 119
 120         self._ACCOUNT_INFO = traverse_obj(account, {
 121             'signatureTimestamp': 'signatureTimestamp',
 122             'uid': 'UID',
 123             'uidSignature': 'UIDSignature',
 124         })
 125
 126         if len(self._ACCOUNT_INFO) != 3:
 127             raise ExtractorError('Failed to retrieve account info with provided cookies', expected=True)
 128
 129     def _get_auth_token(self, url, slug):
 130         if self._TOKEN and self._TOKEN_EXPIRY > int(time.time() + 30):
 131             return
 132
 133         if not self._ACCOUNT_INFO:
 134             self._get_account_info(url, slug)
 135
 136         token = self._download_json(
 137             'https://api.nfl.com/identity/v3/token%s' % (
 138                 '/refresh' if self._ACCOUNT_INFO.get('refreshToken') else ''),
 139             slug, headers={'Content-Type': 'application/json'}, note='Downloading access token',
 140             data=json.dumps({**self._CLIENT_DATA, **self._ACCOUNT_INFO}, separators=(',', ':')).encode())
 141
 142         self._TOKEN = token['accessToken']
 143         self._TOKEN_EXPIRY = token['expiresIn']
 144         self._ACCOUNT_INFO['refreshToken'] = token['refreshToken']
 145
 146     def _parse_video_config(self, video_config, display_id):
 147         video_config = self._parse_json(video_config, display_id)
 148         item = video_config['playlist'][0]
 149         mcp_id = item.get('mcpID')
 150         if mcp_id:
 151             info = self.url_result(f'{self._ANVATO_PREFIX}{mcp_id}', AnvatoIE, mcp_id)
 152         else:
 153             media_id = item.get('id') or item['entityId']
 154             title = item.get('title')
 155             item_url = item['url']
 156             info = {'id': media_id}
 157             ext = determine_ext(item_url)
 158             if ext == 'm3u8':
 159                 info['formats'] = self._extract_m3u8_formats(item_url, media_id, 'mp4')
 160             else:
 161                 info['url'] = item_url
 162                 if item.get('audio') is True:
 163                     info['vcodec'] = 'none'
 164             is_live = video_config.get('live') is True
 165             thumbnails = None
 166             image_url = item.get(item.get('imageSrc')) or item.get(item.get('posterImage'))
 167             if image_url:
 168                 thumbnails = [{
 169                     'url': image_url,
 170                     'ext': determine_ext(image_url, 'jpg'),
 171                 }]
 172             info.update({
 173                 'title': title,
 174                 'is_live': is_live,
 175                 'description': clean_html(item.get('description')),
 176                 'thumbnails': thumbnails,
 177             })
 178         return info
 179
 180
 181 class NFLIE(NFLBaseIE):
 182     IE_NAME = 'nfl.com'
 183     _VALID_URL = NFLBaseIE._VALID_URL_BASE + r'(?:videos?|listen|audio)/(?P<id>[^/#?&]+)'
 184     _TESTS = [{
 185         'url': 'https://www.nfl.com/videos/baker-mayfield-s-game-changing-plays-from-3-td-game-week-14',
 186         'info_dict': {
 187             'id': '899441',
 188             'ext': 'mp4',
 189             'title': "Baker Mayfield's game-changing plays from 3-TD game Week 14",
 190             'description': 'md5:85e05a3cc163f8c344340f220521136d',
 191             'upload_date': '20201215',
 192             'timestamp': 1608009755,
 193             'thumbnail': r're:^https?://.*\.jpg$',
 194             'uploader': 'NFL',
 195             'tags': 'count:6',
 196             'duration': 157,
 197             'categories': 'count:3',
 198         }
 199     }, {
 200         'url': 'https://www.chiefs.com/listen/patrick-mahomes-travis-kelce-react-to-win-over-dolphins-the-breakdown',
 201         'md5': '6886b32c24b463038c760ceb55a34566',
 202         'info_dict': {
 203             'id': 'd87e8790-3e14-11eb-8ceb-ff05c2867f99',
 204             'ext': 'mp3',
 205             'title': 'Patrick Mahomes, Travis Kelce React to Win Over Dolphins | The Breakdown',
 206             'description': 'md5:12ada8ee70e6762658c30e223e095075',
 207         },
 208         'skip': 'HTTP Error 404: Not Found',
 209     }, {
 210         'url': 'https://www.buffalobills.com/video/buffalo-bills-military-recognition-week-14',
 211         'only_matching': True,
 212     }, {
 213         'url': 'https://www.raiders.com/audio/instant-reactions-raiders-week-14-loss-to-indianapolis-colts-espn-jason-fitz',
 214         'only_matching': True,
 215     }]
 216
 217     def _real_extract(self, url):
 218         display_id = self._match_id(url)
 219         webpage = self._download_webpage(url, display_id)
 220         return self._parse_video_config(self._search_regex(
 221             self._VIDEO_CONFIG_REGEX, webpage, 'video config'), display_id)
 222
 223
 224 class NFLArticleIE(NFLBaseIE):
 225     IE_NAME = 'nfl.com:article'
 226     _VALID_URL = NFLBaseIE._VALID_URL_BASE + r'news/(?P<id>[^/#?&]+)'
 227     _TEST = {
 228         'url': 'https://www.buffalobills.com/news/the-only-thing-we-ve-earned-is-the-noise-bills-coaches-discuss-handling-rising-e',
 229         'info_dict': {
 230             'id': 'the-only-thing-we-ve-earned-is-the-noise-bills-coaches-discuss-handling-rising-e',
 231             'title': "'The only thing we've earned is the noise' | Bills coaches discuss handling rising expectations",
 232         },
 233         'playlist_count': 4,
 234     }
 235
 236     def _real_extract(self, url):
 237         display_id = self._match_id(url)
 238         webpage = self._download_webpage(url, display_id)
 239         entries = []
 240         for video_config in re.findall(self._VIDEO_CONFIG_REGEX, webpage):
 241             entries.append(self._parse_video_config(video_config, display_id))
 242         title = clean_html(get_element_by_class(
 243             'nfl-c-article__title', webpage)) or self._html_search_meta(
 244             ['og:title', 'twitter:title'], webpage)
 245         return self.playlist_result(entries, display_id, title)
 246
 247
 248 class NFLPlusReplayIE(NFLBaseIE):
 249     IE_NAME = 'nfl.com:plus:replay'
 250     _VALID_URL = r'https?://(?:www\.)?nfl\.com/plus/games/(?P<slug>[\w-]+)(?:/(?P<id>\d+))?'
 251     _TESTS = [{
 252         'url': 'https://www.nfl.com/plus/games/giants-at-vikings-2022-post-1/1572108',
 253         'info_dict': {
 254             'id': '1572108',
 255             'ext': 'mp4',
 256             'title': 'New York Giants at Minnesota Vikings',
 257             'description': 'New York Giants play the Minnesota Vikings at U.S. Bank Stadium on January 15, 2023',
 258             'uploader': 'NFL',
 259             'upload_date': '20230116',
 260             'timestamp': 1673864520,
 261             'duration': 7157,
 262             'categories': ['Game Highlights'],
 263             'tags': ['Minnesota Vikings', 'New York Giants', 'Minnesota Vikings vs. New York Giants'],
 264             'thumbnail': r're:^https?://.*\.jpg',
 265         },
 266         'params': {'skip_download': 'm3u8'},
 267     }, {
 268         'note': 'Subscription required',
 269         'url': 'https://www.nfl.com/plus/games/giants-at-vikings-2022-post-1',
 270         'playlist_count': 4,
 271         'info_dict': {
 272             'id': 'giants-at-vikings-2022-post-1',
 273         },
 274     }, {
 275         'note': 'Subscription required',
 276         'url': 'https://www.nfl.com/plus/games/giants-at-patriots-2011-pre-4',
 277         'playlist_count': 2,
 278         'info_dict': {
 279             'id': 'giants-at-patriots-2011-pre-4',
 280         },
 281     }, {
 282         'note': 'Subscription required',
 283         'url': 'https://www.nfl.com/plus/games/giants-at-patriots-2011-pre-4',
 284         'info_dict': {
 285             'id': '950701',
 286             'ext': 'mp4',
 287             'title': 'Giants @ Patriots',
 288             'description': 'Giants at Patriots on September 01, 2011',
 289             'uploader': 'NFL',
 290             'upload_date': '20210724',
 291             'timestamp': 1627085874,
 292             'duration': 1532,
 293             'categories': ['Game Highlights'],
 294             'tags': ['play-by-play'],
 295             'thumbnail': r're:^https?://.*\.jpg',
 296         },
 297         'params': {
 298             'skip_download': 'm3u8',
 299             'extractor_args': {'nflplusreplay': {'type': ['condensed_game']}},
 300         },
 301     }]
 302
 303     _REPLAY_TYPES = {
 304         'full_game': 'Full Game',
 305         'full_game_spanish': 'Full Game - Spanish',
 306         'condensed_game': 'Condensed Game',
 307         'all_22': 'All-22',
 308     }
 309
 310     def _real_extract(self, url):
 311         slug, video_id = self._match_valid_url(url).group('slug', 'id')
 312         requested_types = self._configuration_arg('type', ['all'])
 313         if 'all' in requested_types:
 314             requested_types = list(self._REPLAY_TYPES.keys())
 315         requested_types = traverse_obj(self._REPLAY_TYPES, (None, requested_types))
 316
 317         if not video_id:
 318             self._get_auth_token(url, slug)
 319             headers = {'Authorization': f'Bearer {self._TOKEN}'}
 320             game_id = self._download_json(
 321                 f'https://api.nfl.com/football/v2/games/externalId/slug/{slug}', slug,
 322                 'Downloading game ID', query={'withExternalIds': 'true'}, headers=headers)['id']
 323             replays = self._download_json(
 324                 'https://api.nfl.com/content/v1/videos/replays', slug, 'Downloading replays JSON',
 325                 query={'gameId': game_id}, headers=headers)
 326             if len(requested_types) == 1:
 327                 video_id = traverse_obj(replays, (
 328                     'items', lambda _, v: v['subType'] == requested_types[0], 'mcpPlaybackId'), get_all=False)
 329
 330         if video_id:
 331             return self.url_result(f'{self._ANVATO_PREFIX}{video_id}', AnvatoIE, video_id)
 332
 333         def entries():
 334             for replay in traverse_obj(
 335                 replays, ('items', lambda _, v: v['mcpPlaybackId'] and v['subType'] in requested_types)
 336             ):
 337                 video_id = replay['mcpPlaybackId']
 338                 yield self.url_result(f'{self._ANVATO_PREFIX}{video_id}', AnvatoIE, video_id)
 339
 340         return self.playlist_result(entries(), slug)
 341
 342
 343 class NFLPlusEpisodeIE(NFLBaseIE):
 344     IE_NAME = 'nfl.com:plus:episode'
 345     _VALID_URL = r'https?://(?:www\.)?nfl\.com/plus/episodes/(?P<id>[\w-]+)'
 346     _TESTS = [{
 347         'note': 'Subscription required',
 348         'url': 'https://www.nfl.com/plus/episodes/kurt-s-qb-insider-conference-championships',
 349         'info_dict': {
 350             'id': '1576832',
 351             'ext': 'mp4',
 352             'title': 'Conference Championships',
 353             'description': 'md5:944f7fab56f7a37430bf8473f5473857',
 354             'uploader': 'NFL',
 355             'upload_date': '20230127',
 356             'timestamp': 1674782760,
 357             'duration': 730,
 358             'categories': ['Analysis'],
 359             'tags': ['Cincinnati Bengals at Kansas City Chiefs (2022-POST-3)'],
 360             'thumbnail': r're:^https?://.*\.jpg',
 361         },
 362         'params': {'skip_download': 'm3u8'},
 363     }]
 364
 365     def _real_extract(self, url):
 366         slug = self._match_id(url)
 367         self._get_auth_token(url, slug)
 368         video_id = self._download_json(
 369             f'https://api.nfl.com/content/v1/videos/episodes/{slug}', slug, headers={
 370                 'Authorization': f'Bearer {self._TOKEN}',
 371             })['mcpPlaybackId']
 372
 373         return self.url_result(f'{self._ANVATO_PREFIX}{video_id}', AnvatoIE, video_id)