yt_dlp/extractor/yahoo.py

   1 import hashlib
   2 import itertools
   3 import urllib.parse
   4
   5 from .common import InfoExtractor, SearchInfoExtractor
   6 from .youtube import YoutubeIE
   7 from ..utils import (
   8     ExtractorError,
   9     clean_html,
  10     int_or_none,
  11     mimetype2ext,
  12     parse_iso8601,
  13     traverse_obj,
  14     try_get,
  15     url_or_none,
  16 )
  17
  18
  19 class YahooIE(InfoExtractor):
  20     IE_DESC = 'Yahoo screen and movies'
  21     _VALID_URL = r'(?P<url>https?://(?:(?P<country>[a-zA-Z]{2}(?:-[a-zA-Z]{2})?|malaysia)\.)?(?:[\da-zA-Z_-]+\.)?yahoo\.com/(?:[^/]+/)*(?P<id>[^?&#]*-[0-9]+(?:-[a-z]+)?)\.html)'
  22     _EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1']
  23
  24     _TESTS = [{
  25         'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
  26         'info_dict': {
  27             'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
  28             'ext': 'mp4',
  29             'title': 'Julian Smith & Travis Legg Watch Julian Smith',
  30             'description': 'Julian and Travis watch Julian Smith',
  31             'duration': 6863,
  32             'timestamp': 1369812016,
  33             'upload_date': '20130529',
  34         },
  35         'skip': 'No longer exists',
  36     }, {
  37         'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
  38         'md5': '7993e572fac98e044588d0b5260f4352',
  39         'info_dict': {
  40             'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
  41             'ext': 'mp4',
  42             'title': "Yahoo Saves 'Community'",
  43             'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
  44             'duration': 170,
  45             'timestamp': 1406838636,
  46             'upload_date': '20140731',
  47         },
  48         'skip': 'Unfortunately, this video is not available in your region',
  49     }, {
  50         'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
  51         'md5': '71298482f7c64cbb7fa064e4553ff1c1',
  52         'info_dict': {
  53             'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
  54             'ext': 'webm',
  55             'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
  56             'description': 'md5:f66c890e1490f4910a9953c941dee944',
  57             'duration': 97,
  58             'timestamp': 1414489862,
  59             'upload_date': '20141028',
  60         },
  61         'skip': 'No longer exists',
  62     }, {
  63         'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
  64         'md5': '88e209b417f173d86186bef6e4d1f160',
  65         'info_dict': {
  66             'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
  67             'ext': 'mp4',
  68             'title': 'China Moses Is Crazy About the Blues',
  69             'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
  70             'duration': 128,
  71             'timestamp': 1385722202,
  72             'upload_date': '20131129',
  73         }
  74     }, {
  75         'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
  76         'md5': '2a9752f74cb898af5d1083ea9f661b58',
  77         'info_dict': {
  78             'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
  79             'ext': 'mp4',
  80             'title': '\'True Story\' Trailer',
  81             'description': 'True Story',
  82             'duration': 150,
  83             'timestamp': 1418919206,
  84             'upload_date': '20141218',
  85         },
  86     }, {
  87         'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
  88         'only_matching': True,
  89     }, {
  90         'note': 'NBC Sports embeds',
  91         'url': 'http://sports.yahoo.com/blogs/ncaab-the-dagger/tyler-kalinoski-s-buzzer-beater-caps-davidson-s-comeback-win-185609842.html?guid=nbc_cbk_davidsonbuzzerbeater_150313',
  92         'info_dict': {
  93             'id': '9CsDKds0kvHI',
  94             'ext': 'flv',
  95             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
  96             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
  97             'upload_date': '20150313',
  98             'uploader': 'NBCU-SPORTS',
  99             'timestamp': 1426270238,
 100         },
 101     }, {
 102         'url': 'https://tw.news.yahoo.com/-100120367.html',
 103         'only_matching': True,
 104     }, {
 105         # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
 106         'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
 107         'md5': '4fbafb9c9b6f07aa8f870629f6671b35',
 108         'info_dict': {
 109             'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
 110             'ext': 'mp4',
 111             'title': 'Communitary - Community Episode 1: Ladders',
 112             'description': 'md5:8fc39608213295748e1e289807838c97',
 113             'duration': 1646,
 114             'timestamp': 1440436550,
 115             'upload_date': '20150824',
 116             'series': 'Communitary',
 117             'season_number': 6,
 118             'episode_number': 1,
 119         },
 120         'skip': 'No longer exists',
 121     }, {
 122         # ytwnews://cavideo/
 123         'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html',
 124         'info_dict': {
 125             'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff',
 126             'ext': 'mp4',
 127             'title': '單車天使 - 中文版預',
 128             'description': '中文版預',
 129             'timestamp': 1476696196,
 130             'upload_date': '20161017',
 131         },
 132         'params': {
 133             'skip_download': True,
 134         },
 135     }, {
 136         # Contains both a Yahoo hosted video and multiple Youtube embeds
 137         'url': 'https://www.yahoo.com/entertainment/gwen-stefani-reveals-the-pop-hit-she-passed-on-assigns-it-to-her-voice-contestant-instead-033045672.html',
 138         'info_dict': {
 139             'id': '46c5d95a-528f-3d03-b732-732fcadd51de',
 140             'title': 'Gwen Stefani reveals the pop hit she passed on, assigns it to her \'Voice\' contestant instead',
 141             'description': 'Gwen decided not to record this hit herself, but she decided it was the perfect fit for Kyndall Inskeep.',
 142         },
 143         'playlist': [{
 144             'info_dict': {
 145                 'id': '966d4262-4fd1-3aaa-b45b-049ca6e38ba6',
 146                 'ext': 'mp4',
 147                 'title': 'Gwen Stefani reveals she turned down one of Sia\'s best songs',
 148                 'description': 'On "The Voice" Tuesday, Gwen Stefani told Taylor Swift which Sia hit was almost hers.',
 149                 'timestamp': 1572406500,
 150                 'upload_date': '20191030',
 151             },
 152         }, {
 153             'info_dict': {
 154                 'id': '352CFDOQrKg',
 155                 'ext': 'mp4',
 156                 'title': 'Kyndal Inskeep "Performs the Hell Out of" Sia\'s "Elastic Heart" - The Voice Knockouts 2019',
 157                 'description': 'md5:7fe8e3d5806f96002e55f190d1d94479',
 158                 'uploader': 'The Voice',
 159                 'uploader_id': 'NBCTheVoice',
 160                 'upload_date': '20191029',
 161             },
 162         }],
 163         'params': {
 164             'playlistend': 2,
 165         },
 166         'expected_warnings': ['HTTP Error 404', 'Ignoring subtitle tracks'],
 167     }, {
 168         'url': 'https://malaysia.news.yahoo.com/video/bystanders-help-ontario-policeman-bust-190932818.html',
 169         'only_matching': True,
 170     }, {
 171         'url': 'https://es-us.noticias.yahoo.com/es-la-puerta-irrompible-que-110539379.html',
 172         'only_matching': True,
 173     }, {
 174         'url': 'https://www.yahoo.com/entertainment/v/longtime-cbs-news-60-minutes-032036500-cbs.html',
 175         'only_matching': True,
 176     }]
 177
 178     def _extract_yahoo_video(self, video_id, country):
 179         video = self._download_json(
 180             'https://%s.yahoo.com/_td/api/resource/VideoService.videos;view=full;video_ids=["%s"]' % (country, video_id),
 181             video_id, 'Downloading video JSON metadata')[0]
 182         title = video['title']
 183
 184         if country == 'malaysia':
 185             country = 'my'
 186
 187         is_live = video.get('live_state') == 'live'
 188         fmts = ('m3u8',) if is_live else ('webm', 'mp4')
 189
 190         urls = []
 191         formats = []
 192         subtitles = {}
 193         for fmt in fmts:
 194             media_obj = self._download_json(
 195                 'https://video-api.yql.yahoo.com/v1/video/sapi/streams/' + video_id,
 196                 video_id, 'Downloading %s JSON metadata' % fmt,
 197                 headers=self.geo_verification_headers(), query={
 198                     'format': fmt,
 199                     'region': country.upper(),
 200                 })['query']['results']['mediaObj'][0]
 201             msg = media_obj.get('status', {}).get('msg')
 202
 203             for s in media_obj.get('streams', []):
 204                 host = s.get('host')
 205                 path = s.get('path')
 206                 if not host or not path:
 207                     continue
 208                 s_url = host + path
 209                 if s.get('format') == 'm3u8':
 210                     formats.extend(self._extract_m3u8_formats(
 211                         s_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 212                     continue
 213                 tbr = int_or_none(s.get('bitrate'))
 214                 formats.append({
 215                     'url': s_url,
 216                     'format_id': fmt + ('-%d' % tbr if tbr else ''),
 217                     'width': int_or_none(s.get('width')),
 218                     'height': int_or_none(s.get('height')),
 219                     'tbr': tbr,
 220                     'fps': int_or_none(s.get('framerate')),
 221                 })
 222
 223             for cc in media_obj.get('closedcaptions', []):
 224                 cc_url = cc.get('url')
 225                 if not cc_url or cc_url in urls:
 226                     continue
 227                 urls.append(cc_url)
 228                 subtitles.setdefault(cc.get('lang') or 'en-US', []).append({
 229                     'url': cc_url,
 230                     'ext': mimetype2ext(cc.get('content_type')),
 231                 })
 232
 233         streaming_url = video.get('streaming_url')
 234         if streaming_url and not is_live:
 235             formats.extend(self._extract_m3u8_formats(
 236                 streaming_url, video_id, 'mp4',
 237                 'm3u8_native', m3u8_id='hls', fatal=False))
 238
 239         if not formats and msg == 'geo restricted':
 240             self.raise_geo_restricted(metadata_available=True)
 241
 242         thumbnails = []
 243         for thumb in video.get('thumbnails', []):
 244             thumb_url = thumb.get('url')
 245             if not thumb_url:
 246                 continue
 247             thumbnails.append({
 248                 'id': thumb.get('tag'),
 249                 'url': thumb.get('url'),
 250                 'width': int_or_none(thumb.get('width')),
 251                 'height': int_or_none(thumb.get('height')),
 252             })
 253
 254         series_info = video.get('series_info') or {}
 255
 256         return {
 257             'id': video_id,
 258             'title': title,
 259             'formats': formats,
 260             'thumbnails': thumbnails,
 261             'description': clean_html(video.get('description')),
 262             'timestamp': parse_iso8601(video.get('publish_time')),
 263             'subtitles': subtitles,
 264             'duration': int_or_none(video.get('duration')),
 265             'view_count': int_or_none(video.get('view_count')),
 266             'is_live': is_live,
 267             'series': video.get('show_name'),
 268             'season_number': int_or_none(series_info.get('season_number')),
 269             'episode_number': int_or_none(series_info.get('episode_number')),
 270         }
 271
 272     def _real_extract(self, url):
 273         url, country, display_id = self._match_valid_url(url).groups()
 274         if not country:
 275             country = 'us'
 276         else:
 277             country = country.split('-')[0]
 278
 279         items = self._download_json(
 280             'https://%s.yahoo.com/caas/content/article' % country, display_id,
 281             'Downloading content JSON metadata', query={
 282                 'url': url
 283             })['items'][0]
 284
 285         item = items['data']['partnerData']
 286         if item.get('type') != 'video':
 287             entries = []
 288
 289             cover = item.get('cover') or {}
 290             if cover.get('type') == 'yvideo':
 291                 cover_url = cover.get('url')
 292                 if cover_url:
 293                     entries.append(self.url_result(
 294                         cover_url, 'Yahoo', cover.get('uuid')))
 295
 296             for e in (item.get('body') or []):
 297                 if e.get('type') == 'videoIframe':
 298                     iframe_url = e.get('url')
 299                     if iframe_url:
 300                         entries.append(self.url_result(iframe_url))
 301
 302             if item.get('type') == 'storywithleadvideo':
 303                 iframe_url = try_get(item, lambda x: x['meta']['player']['url'])
 304                 if iframe_url:
 305                     entries.append(self.url_result(iframe_url))
 306                 else:
 307                     self.report_warning("Yahoo didn't provide an iframe url for this storywithleadvideo")
 308
 309             if items.get('markup'):
 310                 entries.extend(
 311                     self.url_result(yt_url) for yt_url in YoutubeIE._extract_embed_urls(url, items['markup']))
 312
 313             return self.playlist_result(
 314                 entries, item.get('uuid'),
 315                 item.get('title'), item.get('summary'))
 316
 317         info = self._extract_yahoo_video(item['uuid'], country)
 318         info['display_id'] = display_id
 319         return info
 320
 321
 322 class YahooSearchIE(SearchInfoExtractor):
 323     IE_DESC = 'Yahoo screen search'
 324     _MAX_RESULTS = 1000
 325     IE_NAME = 'screen.yahoo:search'
 326     _SEARCH_KEY = 'yvsearch'
 327
 328     def _search_results(self, query):
 329         for pagenum in itertools.count(0):
 330             result_url = 'http://video.search.yahoo.com/search/?p=%s&fr=screen&o=js&gs=0&b=%d' % (urllib.parse.quote_plus(query), pagenum * 30)
 331             info = self._download_json(result_url, query,
 332                                        note='Downloading results page ' + str(pagenum + 1))
 333             yield from (self.url_result(result['rurl']) for result in info['results'])
 334             if info['m']['last'] >= info['m']['total'] - 1:
 335                 break
 336
 337
 338 class YahooJapanNewsIE(InfoExtractor):
 339     IE_NAME = 'yahoo:japannews'
 340     IE_DESC = 'Yahoo! Japan News'
 341     _VALID_URL = r'https?://news\.yahoo\.co\.jp/(?:articles|feature)/(?P<id>[a-zA-Z0-9]+)'
 342     _GEO_COUNTRIES = ['JP']
 343     _TESTS = [{
 344         'url': 'https://news.yahoo.co.jp/articles/a70fe3a064f1cfec937e2252c7fc6c1ba3201c0e',
 345         'info_dict': {
 346             'id': 'a70fe3a064f1cfec937e2252c7fc6c1ba3201c0e',
 347             'ext': 'mp4',
 348             'title': '【独自】安倍元総理「国葬」中止求め“脅迫メール”…「子ども誘拐」“送信者”を追跡',
 349             'description': 'md5:1c06974575f930f692d8696fbcfdc546',
 350             'thumbnail': r're:https://.+',
 351         },
 352         'params': {
 353             'skip_download': True,
 354         },
 355     }, {
 356         'url': 'https://news.yahoo.co.jp/feature/1356',
 357         'only_matching': True
 358     }]
 359
 360     def _extract_formats(self, json_data, content_id):
 361         formats = []
 362
 363         for vid in traverse_obj(json_data, ('ResultSet', 'Result', ..., 'VideoUrlSet', 'VideoUrl', ...)) or []:
 364             delivery = vid.get('delivery')
 365             url = url_or_none(vid.get('Url'))
 366             if not delivery or not url:
 367                 continue
 368             elif delivery == 'hls':
 369                 formats.extend(
 370                     self._extract_m3u8_formats(
 371                         url, content_id, 'mp4', 'm3u8_native',
 372                         m3u8_id='hls', fatal=False))
 373             else:
 374                 formats.append({
 375                     'url': url,
 376                     'format_id': f'http-{vid.get("bitrate")}',
 377                     'height': int_or_none(vid.get('height')),
 378                     'width': int_or_none(vid.get('width')),
 379                     'tbr': int_or_none(vid.get('bitrate')),
 380                 })
 381         self._remove_duplicate_formats(formats)
 382
 383         return formats
 384
 385     def _real_extract(self, url):
 386         video_id = self._match_id(url)
 387         webpage = self._download_webpage(url, video_id)
 388         preloaded_state = self._search_json(r'__PRELOADED_STATE__\s*=', webpage, 'preloaded state', video_id)
 389
 390         content_id = traverse_obj(
 391             preloaded_state, ('articleDetail', 'paragraphs', ..., 'objectItems', ..., 'video', 'vid'),
 392             get_all=False, expected_type=int)
 393         if content_id is None:
 394             raise ExtractorError('This article does not contain a video', expected=True)
 395
 396         HOST = 'news.yahoo.co.jp'
 397         space_id = traverse_obj(preloaded_state, ('pageData', 'spaceId'), expected_type=str)
 398         json_data = self._download_json(
 399             f'https://feapi-yvpub.yahooapis.jp/v1/content/{content_id}',
 400             video_id, query={
 401                 'appid': 'dj0zaiZpPVZMTVFJR0FwZWpiMyZzPWNvbnN1bWVyc2VjcmV0Jng9YjU-',
 402                 'output': 'json',
 403                 'domain': HOST,
 404                 'ak': hashlib.md5('_'.join((space_id, HOST)).encode()).hexdigest() if space_id else '',
 405                 'device_type': '1100',
 406             })
 407
 408         title = (
 409             traverse_obj(preloaded_state,
 410                          ('articleDetail', 'headline'), ('pageData', 'pageParam', 'title'),
 411                          expected_type=str)
 412             or self._html_search_meta(('og:title', 'twitter:title'), webpage, 'title', default=None)
 413             or self._html_extract_title(webpage))
 414         description = (
 415             traverse_obj(preloaded_state, ('pageData', 'description'), expected_type=str)
 416             or self._html_search_meta(
 417                 ('og:description', 'description', 'twitter:description'),
 418                 webpage, 'description', default=None))
 419         thumbnail = (
 420             traverse_obj(preloaded_state, ('pageData', 'ogpImage'), expected_type=str)
 421             or self._og_search_thumbnail(webpage, default=None)
 422             or self._html_search_meta('twitter:image', webpage, 'thumbnail', default=None))
 423
 424         return {
 425             'id': video_id,
 426             'title': title,
 427             'description': description,
 428             'thumbnail': thumbnail,
 429             'formats': self._extract_formats(json_data, video_id),
 430         }