yt_dlp/extractor/mlb.py

   1 import re
   2 import urllib.parse
   3 import uuid
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     determine_ext,
   8     int_or_none,
   9     join_nonempty,
  10     parse_duration,
  11     parse_iso8601,
  12     traverse_obj,
  13     try_get,
  14 )
  15
  16
  17 class MLBBaseIE(InfoExtractor):
  18     def _real_extract(self, url):
  19         display_id = self._match_id(url)
  20         video = self._download_video_data(display_id)
  21         video_id = video['id']
  22         title = video['title']
  23         feed = self._get_feed(video)
  24
  25         formats = []
  26         for playback in (feed.get('playbacks') or []):
  27             playback_url = playback.get('url')
  28             if not playback_url:
  29                 continue
  30             name = playback.get('name')
  31             ext = determine_ext(playback_url)
  32             if ext == 'm3u8':
  33                 formats.extend(self._extract_m3u8_formats(
  34                     playback_url, video_id, 'mp4',
  35                     'm3u8_native', m3u8_id=name, fatal=False))
  36             else:
  37                 f = {
  38                     'format_id': name,
  39                     'url': playback_url,
  40                 }
  41                 mobj = re.search(r'_(\d+)K_(\d+)X(\d+)', name)
  42                 if mobj:
  43                     f.update({
  44                         'height': int(mobj.group(3)),
  45                         'tbr': int(mobj.group(1)),
  46                         'width': int(mobj.group(2)),
  47                     })
  48                 mobj = re.search(r'_(\d+)x(\d+)_(\d+)_(\d+)K\.mp4', playback_url)
  49                 if mobj:
  50                     f.update({
  51                         'fps': int(mobj.group(3)),
  52                         'height': int(mobj.group(2)),
  53                         'tbr': int(mobj.group(4)),
  54                         'width': int(mobj.group(1)),
  55                     })
  56                 formats.append(f)
  57
  58         thumbnails = []
  59         for cut in (try_get(feed, lambda x: x['image']['cuts'], list) or []):
  60             src = cut.get('src')
  61             if not src:
  62                 continue
  63             thumbnails.append({
  64                 'height': int_or_none(cut.get('height')),
  65                 'url': src,
  66                 'width': int_or_none(cut.get('width')),
  67             })
  68
  69         language = (video.get('language') or 'EN').lower()
  70
  71         return {
  72             'id': video_id,
  73             'title': title,
  74             'formats': formats,
  75             'description': video.get('description'),
  76             'duration': parse_duration(feed.get('duration')),
  77             'thumbnails': thumbnails,
  78             'timestamp': parse_iso8601(video.get(self._TIMESTAMP_KEY)),
  79             'subtitles': self._extract_mlb_subtitles(feed, language),
  80         }
  81
  82
  83 class MLBIE(MLBBaseIE):
  84     _VALID_URL = r'''(?x)
  85                     https?://
  86                         (?:[\da-z_-]+\.)*mlb\.com/
  87                         (?:
  88                             (?:
  89                                 (?:[^/]+/)*video/[^/]+/c-|
  90                                 (?:
  91                                     shared/video/embed/(?:embed|m-internal-embed)\.html|
  92                                     (?:[^/]+/)+(?:play|index)\.jsp|
  93                                 )\?.*?\bcontent_id=
  94                             )
  95                             (?P<id>\d+)
  96                         )
  97                     '''
  98     _EMBED_REGEX = [
  99         r'<iframe[^>]+?src=(["\'])(?P<url>https?://m(?:lb)?\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
 100         r'data-video-link=["\'](?P<url>http://m\.mlb\.com/video/[^"\']+)',
 101     ]
 102     _TESTS = [
 103         {
 104             'url': 'https://www.mlb.com/mariners/video/ackleys-spectacular-catch/c-34698933',
 105             'md5': '632358dacfceec06bad823b83d21df2d',
 106             'info_dict': {
 107                 'id': '34698933',
 108                 'ext': 'mp4',
 109                 'title': "Ackley's spectacular catch",
 110                 'description': 'md5:7f5a981eb4f3cbc8daf2aeffa2215bf0',
 111                 'duration': 66,
 112                 'timestamp': 1405995000,
 113                 'upload_date': '20140722',
 114                 'thumbnail': r're:^https?://.*\.jpg$',
 115             },
 116         },
 117         {
 118             'url': 'https://www.mlb.com/video/stanton-prepares-for-derby/c-34496663',
 119             'md5': 'bf2619bf9cacc0a564fc35e6aeb9219f',
 120             'info_dict': {
 121                 'id': '34496663',
 122                 'ext': 'mp4',
 123                 'title': 'Stanton prepares for Derby',
 124                 'description': 'md5:d00ce1e5fd9c9069e9c13ab4faedfa57',
 125                 'duration': 46,
 126                 'timestamp': 1405120200,
 127                 'upload_date': '20140711',
 128                 'thumbnail': r're:^https?://.*\.jpg$',
 129             },
 130         },
 131         {
 132             'url': 'https://www.mlb.com/video/cespedes-repeats-as-derby-champ/c-34578115',
 133             'md5': '99bb9176531adc600b90880fb8be9328',
 134             'info_dict': {
 135                 'id': '34578115',
 136                 'ext': 'mp4',
 137                 'title': 'Cespedes repeats as Derby champ',
 138                 'description': 'md5:08df253ce265d4cf6fb09f581fafad07',
 139                 'duration': 488,
 140                 'timestamp': 1405414336,
 141                 'upload_date': '20140715',
 142                 'thumbnail': r're:^https?://.*\.jpg$',
 143             },
 144         },
 145         {
 146             'url': 'https://www.mlb.com/video/bautista-on-home-run-derby/c-34577915',
 147             'md5': 'da8b57a12b060e7663ee1eebd6f330ec',
 148             'info_dict': {
 149                 'id': '34577915',
 150                 'ext': 'mp4',
 151                 'title': 'Bautista on Home Run Derby',
 152                 'description': 'md5:b80b34031143d0986dddc64a8839f0fb',
 153                 'duration': 52,
 154                 'timestamp': 1405405122,
 155                 'upload_date': '20140715',
 156                 'thumbnail': r're:^https?://.*\.jpg$',
 157             },
 158         },
 159         {
 160             'url': 'https://www.mlb.com/video/hargrove-homers-off-caldwell/c-1352023483?tid=67793694',
 161             'only_matching': True,
 162         },
 163         {
 164             'url': 'http://m.mlb.com/shared/video/embed/embed.html?content_id=35692085&topic_id=6479266&width=400&height=224&property=mlb',
 165             'only_matching': True,
 166         },
 167         {
 168             'url': 'http://mlb.mlb.com/shared/video/embed/embed.html?content_id=36599553',
 169             'only_matching': True,
 170         },
 171         {
 172             'url': 'http://mlb.mlb.com/es/video/play.jsp?content_id=36599553',
 173             'only_matching': True,
 174         },
 175         {
 176             'url': 'https://www.mlb.com/cardinals/video/piscottys-great-sliding-catch/c-51175783',
 177             'only_matching': True,
 178         },
 179         {
 180             # From http://m.mlb.com/news/article/118550098/blue-jays-kevin-pillar-goes-spidey-up-the-wall-to-rob-tim-beckham-of-a-homer
 181             'url': 'http://mlb.mlb.com/shared/video/embed/m-internal-embed.html?content_id=75609783&property=mlb&autoplay=true&hashmode=false&siteSection=mlb/multimedia/article_118550098/article_embed&club=mlb',
 182             'only_matching': True,
 183         },
 184     ]
 185     _TIMESTAMP_KEY = 'date'
 186
 187     @staticmethod
 188     def _get_feed(video):
 189         return video
 190
 191     @staticmethod
 192     def _extract_mlb_subtitles(feed, language):
 193         subtitles = {}
 194         for keyword in (feed.get('keywordsAll') or []):
 195             keyword_type = keyword.get('type')
 196             if keyword_type and keyword_type.startswith('closed_captions_location_'):
 197                 cc_location = keyword.get('value')
 198                 if cc_location:
 199                     subtitles.setdefault(language, []).append({
 200                         'url': cc_location,
 201                     })
 202         return subtitles
 203
 204     def _download_video_data(self, display_id):
 205         return self._download_json(
 206             f'http://content.mlb.com/mlb/item/id/v1/{display_id}/details/web-v1.json',
 207             display_id)
 208
 209
 210 class MLBVideoIE(MLBBaseIE):
 211     _VALID_URL = r'https?://(?:www\.)?mlb\.com/(?:[^/]+/)*video/(?P<id>[^/?&#]+)'
 212     _TEST = {
 213         'url': 'https://www.mlb.com/mariners/video/ackley-s-spectacular-catch-c34698933',
 214         'md5': '632358dacfceec06bad823b83d21df2d',
 215         'info_dict': {
 216             'id': 'c04a8863-f569-42e6-9f87-992393657614',
 217             'ext': 'mp4',
 218             'title': "Ackley's spectacular catch",
 219             'description': 'md5:7f5a981eb4f3cbc8daf2aeffa2215bf0',
 220             'duration': 66,
 221             'timestamp': 1405995000,
 222             'upload_date': '20140722',
 223             'thumbnail': r're:^https?://.+',
 224         },
 225     }
 226     _TIMESTAMP_KEY = 'timestamp'
 227
 228     @classmethod
 229     def suitable(cls, url):
 230         return False if MLBIE.suitable(url) else super().suitable(url)
 231
 232     @staticmethod
 233     def _get_feed(video):
 234         return video['feeds'][0]
 235
 236     @staticmethod
 237     def _extract_mlb_subtitles(feed, language):
 238         subtitles = {}
 239         for cc_location in (feed.get('closedCaptions') or []):
 240             subtitles.setdefault(language, []).append({
 241                 'url': cc_location,
 242             })
 243
 244     def _download_video_data(self, display_id):
 245         # https://www.mlb.com/data-service/en/videos/[SLUG]
 246         return self._download_json(
 247             'https://fastball-gateway.mlb.com/graphql',
 248             display_id, query={
 249                 'query': '''{
 250   mediaPlayback(ids: "%s") {
 251     description
 252     feeds(types: CMS) {
 253       closedCaptions
 254       duration
 255       image {
 256         cuts {
 257           width
 258           height
 259           src
 260         }
 261       }
 262       playbacks {
 263         name
 264         url
 265       }
 266     }
 267     id
 268     timestamp
 269     title
 270   }
 271 }''' % display_id,  # noqa: UP031
 272             })['data']['mediaPlayback'][0]
 273
 274
 275 class MLBTVIE(InfoExtractor):
 276     _VALID_URL = r'https?://(?:www\.)?mlb\.com/tv/g(?P<id>\d{6})'
 277     _NETRC_MACHINE = 'mlb'
 278
 279     _TESTS = [{
 280         'url': 'https://www.mlb.com/tv/g661581/vee2eff5f-a7df-4c20-bdb4-7b926fa12638',
 281         'info_dict': {
 282             'id': '661581',
 283             'ext': 'mp4',
 284             'title': '2022-07-02 - St. Louis Cardinals @ Philadelphia Phillies',
 285         },
 286         'params': {
 287             'skip_download': True,
 288         },
 289     }]
 290     _access_token = None
 291
 292     def _real_initialize(self):
 293         if not self._access_token:
 294             self.raise_login_required(
 295                 'All videos are only available to registered users', method='password')
 296
 297     def _perform_login(self, username, password):
 298         data = f'grant_type=password&username={urllib.parse.quote(username)}&password={urllib.parse.quote(password)}&scope=openid offline_access&client_id=0oa3e1nutA1HLzAKG356'
 299         access_token = self._download_json(
 300             'https://ids.mlb.com/oauth2/aus1m088yK07noBfh356/v1/token', None,
 301             headers={
 302                 'User-Agent': 'okhttp/3.12.1',
 303                 'Content-Type': 'application/x-www-form-urlencoded',
 304             }, data=data.encode())['access_token']
 305
 306         entitlement = self._download_webpage(
 307             f'https://media-entitlement.mlb.com/api/v3/jwt?os=Android&appname=AtBat&did={uuid.uuid4()}', None,
 308             headers={
 309                 'User-Agent': 'okhttp/3.12.1',
 310                 'Authorization': f'Bearer {access_token}',
 311             })
 312
 313         data = f'grant_type=urn:ietf:params:oauth:grant-type:token-exchange&subject_token={entitlement}&subject_token_type=urn:ietf:params:oauth:token-type:jwt&platform=android-tv'
 314         self._access_token = self._download_json(
 315             'https://us.edge.bamgrid.com/token', None,
 316             headers={
 317                 'Accept': 'application/json',
 318                 'Authorization': 'Bearer bWxidHYmYW5kcm9pZCYxLjAuMA.6LZMbH2r--rbXcgEabaDdIslpo4RyZrlVfWZhsAgXIk',
 319                 'Content-Type': 'application/x-www-form-urlencoded',
 320             }, data=data.encode())['access_token']
 321
 322     def _real_extract(self, url):
 323         video_id = self._match_id(url)
 324         airings = self._download_json(
 325             f'https://search-api-mlbtv.mlb.com/svc/search/v2/graphql/persisted/query/core/Airings?variables=%7B%22partnerProgramIds%22%3A%5B%22{video_id}%22%5D%2C%22applyEsniMediaRightsLabels%22%3Atrue%7D',
 326             video_id)['data']['Airings']
 327
 328         formats, subtitles = [], {}
 329         for airing in airings:
 330             m3u8_url = self._download_json(
 331                 airing['playbackUrls'][0]['href'].format(scenario='browser~csai'), video_id,
 332                 headers={
 333                     'Authorization': self._access_token,
 334                     'Accept': 'application/vnd.media-service+json; version=2',
 335                 })['stream']['complete']
 336             f, s = self._extract_m3u8_formats_and_subtitles(
 337                 m3u8_url, video_id, 'mp4', m3u8_id=join_nonempty(airing.get('feedType'), airing.get('feedLanguage')))
 338             formats.extend(f)
 339             self._merge_subtitles(s, target=subtitles)
 340
 341         return {
 342             'id': video_id,
 343             'title': traverse_obj(airings, (..., 'titles', 0, 'episodeName'), get_all=False),
 344             'is_live': traverse_obj(airings, (..., 'mediaConfig', 'productType'), get_all=False) == 'LIVE',
 345             'formats': formats,
 346             'subtitles': subtitles,
 347             'http_headers': {'Authorization': f'Bearer {self._access_token}'},
 348         }
 349
 350
 351 class MLBArticleIE(InfoExtractor):
 352     _VALID_URL = r'https?://www\.mlb\.com/news/(?P<id>[\w-]+)'
 353     _TESTS = [{
 354         'url': 'https://www.mlb.com/news/manny-machado-robs-guillermo-heredia-reacts',
 355         'info_dict': {
 356             'id': '36db7394-343c-4ea3-b8ca-ead2e61bca9a',
 357             'title': 'Machado\'s grab draws hilarious irate reaction',
 358             'modified_timestamp': 1675888370,
 359             'description': 'md5:a19d4eb0487b2cb304e9a176f6b67676',
 360             'modified_date': '20230208',
 361         },
 362         'playlist_mincount': 2,
 363     }]
 364
 365     def _real_extract(self, url):
 366         display_id = self._match_id(url)
 367         webpage = self._download_webpage(url, display_id)
 368         apollo_cache_json = self._search_json(r'window\.initState\s*=', webpage, 'window.initState', display_id)['apolloCache']
 369
 370         content_real_info = traverse_obj(
 371             apollo_cache_json, ('ROOT_QUERY', lambda k, _: k.startswith('getArticle')), get_all=False)
 372
 373         return self.playlist_from_matches(
 374             traverse_obj(content_real_info, ('parts', lambda _, v: v['__typename'] == 'Video' or v['type'] == 'video')),
 375             getter=lambda x: f'https://www.mlb.com/video/{x["slug"]}',
 376             ie=MLBVideoIE, playlist_id=content_real_info.get('translationId'),
 377             title=self._html_search_meta('og:title', webpage),
 378             description=content_real_info.get('summary'),
 379             modified_timestamp=parse_iso8601(content_real_info.get('lastUpdatedDate')))