yt_dlp/extractor/ign.py

   1 import re
   2 import urllib.parse
   3
   4 from .common import InfoExtractor
   5 from ..networking.exceptions import HTTPError
   6 from ..utils import (
   7     ExtractorError,
   8     determine_ext,
   9     extract_attributes,
  10     int_or_none,
  11     merge_dicts,
  12     parse_iso8601,
  13     strip_or_none,
  14     traverse_obj,
  15     url_or_none,
  16     urljoin,
  17 )
  18
  19
  20 class IGNBaseIE(InfoExtractor):
  21     def _call_api(self, slug):
  22         return self._download_json(
  23             f'http://apis.ign.com/{self._PAGE_TYPE}/v3/{self._PAGE_TYPE}s/slug/{slug}', slug)
  24
  25     def _checked_call_api(self, slug):
  26         try:
  27             return self._call_api(slug)
  28         except ExtractorError as e:
  29             if isinstance(e.cause, HTTPError) and e.cause.status == 404:
  30                 e.cause.args = e.cause.args or [
  31                     e.cause.response.url, e.cause.status, e.cause.reason]
  32                 raise ExtractorError(
  33                     'Content not found: expired?', cause=e.cause,
  34                     expected=True)
  35             raise
  36
  37     def _extract_video_info(self, video, fatal=True):
  38         video_id = video['videoId']
  39
  40         formats = []
  41         refs = traverse_obj(video, 'refs', expected_type=dict) or {}
  42
  43         m3u8_url = url_or_none(refs.get('m3uUrl'))
  44         if m3u8_url:
  45             formats.extend(self._extract_m3u8_formats(
  46                 m3u8_url, video_id, 'mp4', 'm3u8_native',
  47                 m3u8_id='hls', fatal=False))
  48
  49         f4m_url = url_or_none(refs.get('f4mUrl'))
  50         if f4m_url:
  51             formats.extend(self._extract_f4m_formats(
  52                 f4m_url, video_id, f4m_id='hds', fatal=False))
  53
  54         for asset in (video.get('assets') or []):
  55             asset_url = url_or_none(asset.get('url'))
  56             if not asset_url:
  57                 continue
  58             formats.append({
  59                 'url': asset_url,
  60                 'tbr': int_or_none(asset.get('bitrate'), 1000),
  61                 'fps': int_or_none(asset.get('frame_rate')),
  62                 'height': int_or_none(asset.get('height')),
  63                 'width': int_or_none(asset.get('width')),
  64             })
  65
  66         mezzanine_url = traverse_obj(
  67             video, ('system', 'mezzanineUrl'), expected_type=url_or_none)
  68         if mezzanine_url:
  69             formats.append({
  70                 'ext': determine_ext(mezzanine_url, 'mp4'),
  71                 'format_id': 'mezzanine',
  72                 'quality': 1,
  73                 'url': mezzanine_url,
  74             })
  75
  76         thumbnails = traverse_obj(
  77             video, ('thumbnails', ..., {'url': 'url'}), expected_type=url_or_none)
  78         tags = traverse_obj(
  79             video, ('tags', ..., 'displayName'),
  80             expected_type=lambda x: x.strip() or None)
  81
  82         metadata = traverse_obj(video, 'metadata', expected_type=dict) or {}
  83         title = traverse_obj(
  84             metadata, 'longTitle', 'title', 'name',
  85             expected_type=lambda x: x.strip() or None)
  86
  87         return {
  88             'id': video_id,
  89             'title': title,
  90             'description': strip_or_none(metadata.get('description')),
  91             'timestamp': parse_iso8601(metadata.get('publishDate')),
  92             'duration': int_or_none(metadata.get('duration')),
  93             'thumbnails': thumbnails,
  94             'formats': formats,
  95             'tags': tags,
  96         }
  97
  98
  99 class IGNIE(IGNBaseIE):
 100     """
 101     Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com.
 102     Some videos of it.ign.com are also supported
 103     """
 104     _VIDEO_PATH_RE = r'/(?:\d{4}/\d{2}/\d{2}/)?(?P<id>.+?)'
 105     _PLAYLIST_PATH_RE = r'(?:/?\?(?P<filt>[^&#]+))?'
 106     _VALID_URL = (
 107         r'https?://(?:.+?\.ign|www\.pcmag)\.com/videos(?:{})'.format('|'.join((_VIDEO_PATH_RE + r'(?:[/?&#]|$)', _PLAYLIST_PATH_RE))))
 108     IE_NAME = 'ign.com'
 109     _PAGE_TYPE = 'video'
 110
 111     _TESTS = [{
 112         'url': 'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
 113         'md5': 'd2e1586d9987d40fad7867bf96a018ea',
 114         'info_dict': {
 115             'id': '8f862beef863986b2785559b9e1aa599',
 116             'ext': 'mp4',
 117             'title': 'The Last of Us Review',
 118             'description': 'md5:c8946d4260a4d43a00d5ae8ed998870c',
 119             'timestamp': 1370440800,
 120             'upload_date': '20130605',
 121             'tags': 'count:9',
 122             'display_id': 'the-last-of-us-review',
 123             'thumbnail': 'https://assets1.ignimgs.com/vid/thumbnails/user/2014/03/26/lastofusreviewmimig2.jpg',
 124             'duration': 440,
 125         },
 126         'params': {
 127             'nocheckcertificate': True,
 128         },
 129     }, {
 130         'url': 'http://www.pcmag.com/videos/2015/01/06/010615-whats-new-now-is-gogo-snooping-on-your-data',
 131         'md5': 'f1581a6fe8c5121be5b807684aeac3f6',
 132         'info_dict': {
 133             'id': 'ee10d774b508c9b8ec07e763b9125b91',
 134             'ext': 'mp4',
 135             'title': 'What\'s New Now: Is GoGo Snooping on Your Data?',
 136             'description': 'md5:817a20299de610bd56f13175386da6fa',
 137             'timestamp': 1420571160,
 138             'upload_date': '20150106',
 139             'tags': 'count:4',
 140         },
 141         'skip': '404 Not Found',
 142     }, {
 143         'url': 'https://www.ign.com/videos/is-a-resident-evil-4-remake-on-the-way-ign-daily-fix',
 144         'only_matching': True,
 145     }]
 146
 147     @classmethod
 148     def _extract_embed_urls(cls, url, webpage):
 149         grids = re.findall(
 150             r'''(?s)<section\b[^>]+\bclass\s*=\s*['"](?:[\w-]+\s+)*?content-feed-grid(?!\B|-)[^>]+>(.+?)</section[^>]*>''',
 151             webpage)
 152         return filter(
 153             None, (urljoin(url, m.group('path')) for m in re.finditer(
 154                 rf'''<a\b[^>]+\bhref\s*=\s*('|")(?P<path>/videos{cls._VIDEO_PATH_RE})\1''',
 155                 grids[0] if grids else '')))
 156
 157     def _real_extract(self, url):
 158         display_id, filt = self._match_valid_url(url).group('id', 'filt')
 159         if display_id:
 160             return self._extract_video(url, display_id)
 161         return self._extract_playlist(url, filt or 'all')
 162
 163     def _extract_playlist(self, url, display_id):
 164         webpage = self._download_webpage(url, display_id)
 165
 166         return self.playlist_result(
 167             (self.url_result(u, self.ie_key())
 168              for u in self._extract_embed_urls(url, webpage)),
 169             playlist_id=display_id)
 170
 171     def _extract_video(self, url, display_id):
 172         video = self._checked_call_api(display_id)
 173
 174         info = self._extract_video_info(video)
 175
 176         return merge_dicts({
 177             'display_id': display_id,
 178         }, info)
 179
 180
 181 class IGNVideoIE(IGNBaseIE):
 182     _VALID_URL = r'https?://.+?\.ign\.com/(?:[a-z]{2}/)?[^/]+/(?P<id>\d+)/(?:video|trailer)/'
 183     _TESTS = [{
 184         'url': 'http://me.ign.com/en/videos/112203/video/how-hitman-aims-to-be-different-than-every-other-s',
 185         'md5': 'dd9aca7ed2657c4e118d8b261e5e9de1',
 186         'info_dict': {
 187             'id': 'e9be7ea899a9bbfc0674accc22a36cc8',
 188             'ext': 'mp4',
 189             'title': 'How Hitman Aims to Be Different Than Every Other Stealth Game - NYCC 2015',
 190             'description': 'Taking out assassination targets in Hitman has never been more stylish.',
 191             'timestamp': 1444665600,
 192             'upload_date': '20151012',
 193             'display_id': '112203',
 194             'thumbnail': 'https://sm.ign.com/ign_me/video/h/how-hitman/how-hitman-aims-to-be-different-than-every-other-s_8z14.jpg',
 195             'duration': 298,
 196             'tags': 'count:13',
 197         },
 198         'expected_warnings': ['HTTP Error 400: Bad Request'],
 199     }, {
 200         'url': 'http://me.ign.com/ar/angry-birds-2/106533/video/lrd-ldyy-lwl-lfylm-angry-birds',
 201         'only_matching': True,
 202     }, {
 203         # Youtube embed
 204         'url': 'https://me.ign.com/ar/ratchet-clank-rift-apart/144327/trailer/embed',
 205         'only_matching': True,
 206     }, {
 207         # Twitter embed
 208         'url': 'http://adria.ign.com/sherlock-season-4/9687/trailer/embed',
 209         'only_matching': True,
 210     }, {
 211         # Vimeo embed
 212         'url': 'https://kr.ign.com/bic-2018/3307/trailer/embed',
 213         'only_matching': True,
 214     }]
 215
 216     def _real_extract(self, url):
 217         video_id = self._match_id(url)
 218         parsed_url = urllib.parse.urlparse(url)
 219         embed_url = urllib.parse.urlunparse(
 220             parsed_url._replace(path=parsed_url.path.rsplit('/', 1)[0] + '/embed'))
 221
 222         webpage, urlh = self._download_webpage_handle(embed_url, video_id)
 223         new_url = urlh.url
 224         ign_url = urllib.parse.parse_qs(
 225             urllib.parse.urlparse(new_url).query).get('url', [None])[-1]
 226         if ign_url:
 227             return self.url_result(ign_url, IGNIE.ie_key())
 228         video = self._search_regex(r'(<div\b[^>]+\bdata-video-id\s*=\s*[^>]+>)', webpage, 'video element', fatal=False)
 229         if not video:
 230             if new_url == url:
 231                 raise ExtractorError('Redirect loop: ' + url)
 232             return self.url_result(new_url)
 233         video = extract_attributes(video)
 234         video_data = video.get('data-settings') or '{}'
 235         video_data = self._parse_json(video_data, video_id)['video']
 236         info = self._extract_video_info(video_data)
 237
 238         return merge_dicts({
 239             'display_id': video_id,
 240         }, info)
 241
 242
 243 class IGNArticleIE(IGNBaseIE):
 244     _VALID_URL = r'https?://.+?\.ign\.com/(?:articles(?:/\d{4}/\d{2}/\d{2})?|(?:[a-z]{2}/)?(?:[\w-]+/)*?feature/\d+)/(?P<id>[^/?&#]+)'
 245     _PAGE_TYPE = 'article'
 246     _TESTS = [{
 247         'url': 'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind',
 248         'info_dict': {
 249             'id': '72113',
 250             'title': '100 Little Things in GTA 5 That Will Blow Your Mind',
 251         },
 252         'playlist': [
 253             {
 254                 'info_dict': {
 255                     'id': '5ebbd138523268b93c9141af17bec937',
 256                     'ext': 'mp4',
 257                     'title': 'Grand Theft Auto V Video Review',
 258                     'description': 'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.',
 259                     'timestamp': 1379339880,
 260                     'upload_date': '20130916',
 261                     'tags': 'count:12',
 262                     'thumbnail': 'https://assets1.ignimgs.com/thumbs/userUploaded/2021/8/16/gta-v-heistsjpg-e94705-1629138553533.jpeg',
 263                     'display_id': 'grand-theft-auto-v-video-review',
 264                     'duration': 501,
 265                 },
 266             },
 267             {
 268                 'info_dict': {
 269                     'id': '638672ee848ae4ff108df2a296418ee2',
 270                     'ext': 'mp4',
 271                     'title': 'GTA 5 In Slow Motion',
 272                     'description': 'The twisted beauty of GTA 5 in stunning slow motion.',
 273                     'timestamp': 1386878820,
 274                     'upload_date': '20131212',
 275                     'duration': 202,
 276                     'tags': 'count:25',
 277                     'display_id': 'gta-5-in-slow-motion',
 278                     'thumbnail': 'https://assets1.ignimgs.com/vid/thumbnails/user/2013/11/03/GTA-SLO-MO-1.jpg',
 279                 },
 280             },
 281         ],
 282         'params': {
 283             'skip_download': True,
 284         },
 285         'expected_warnings': ['Backend fetch failed'],
 286     }, {
 287         'url': 'http://www.ign.com/articles/2014/08/15/rewind-theater-wild-trailer-gamescom-2014?watch',
 288         'info_dict': {
 289             'id': '53ee806780a81ec46e0790f8',
 290             'title': 'Rewind Theater - Wild Trailer Gamescom 2014',
 291         },
 292         'playlist_count': 1,
 293         'expected_warnings': ['Backend fetch failed'],
 294     }, {
 295         # videoId pattern
 296         'url': 'http://www.ign.com/articles/2017/06/08/new-ducktales-short-donalds-birthday-doesnt-go-as-planned',
 297         'only_matching': True,
 298     }, {
 299         # Youtube embed
 300         'url': 'https://www.ign.com/articles/2021-mvp-named-in-puppy-bowl-xvii',
 301         'only_matching': True,
 302     }, {
 303         # IMDB embed
 304         'url': 'https://www.ign.com/articles/2014/08/07/sons-of-anarchy-final-season-trailer',
 305         'only_matching': True,
 306     }, {
 307         # Facebook embed
 308         'url': 'https://www.ign.com/articles/2017/09/20/marvels-the-punisher-watch-the-new-trailer-for-the-netflix-series',
 309         'only_matching': True,
 310     }, {
 311         # Brightcove embed
 312         'url': 'https://www.ign.com/articles/2016/01/16/supergirl-goes-flying-with-martian-manhunter-in-new-clip',
 313         'only_matching': True,
 314     }]
 315
 316     def _checked_call_api(self, slug):
 317         try:
 318             return self._call_api(slug)
 319         except ExtractorError as e:
 320             if isinstance(e.cause, HTTPError):
 321                 e.cause.args = e.cause.args or [
 322                     e.cause.response.url, e.cause.status, e.cause.reason]
 323                 if e.cause.status == 404:
 324                     raise ExtractorError(
 325                         'Content not found: expired?', cause=e.cause,
 326                         expected=True)
 327                 elif e.cause.status == 503:
 328                     self.report_warning(str(e.cause))
 329                     return
 330             raise
 331
 332     def _real_extract(self, url):
 333         display_id = self._match_id(url)
 334         article = self._checked_call_api(display_id)
 335
 336         if article:
 337             # obsolete ?
 338             def entries():
 339                 media_url = traverse_obj(
 340                     article, ('mediaRelations', 0, 'media', 'metadata', 'url'),
 341                     expected_type=url_or_none)
 342                 if media_url:
 343                     yield self.url_result(media_url, IGNIE.ie_key())
 344                 for content in (article.get('content') or []):
 345                     for video_url in re.findall(r'(?:\[(?:ignvideo\s+url|youtube\s+clip_id)|<iframe[^>]+src)="([^"]+)"', content):
 346                         if url_or_none(video_url):
 347                             yield self.url_result(video_url)
 348
 349             return self.playlist_result(
 350                 entries(), article.get('articleId'),
 351                 traverse_obj(
 352                     article, ('metadata', 'headline'),
 353                     expected_type=lambda x: x.strip() or None))
 354
 355         webpage = self._download_webpage(url, display_id)
 356
 357         playlist_id = self._html_search_meta('dable:item_id', webpage, default=None)
 358         if playlist_id:
 359
 360             def entries():
 361                 for m in re.finditer(
 362                         r'''(?s)<object\b[^>]+\bclass\s*=\s*("|')ign-videoplayer\1[^>]*>(?P<params>.+?)</object''',
 363                         webpage):
 364                     flashvars = self._search_regex(
 365                         r'''(<param\b[^>]+\bname\s*=\s*("|')flashvars\2[^>]*>)''',
 366                         m.group('params'), 'flashvars', default='')
 367                     flashvars = urllib.parse.parse_qs(extract_attributes(flashvars).get('value') or '')
 368                     v_url = url_or_none((flashvars.get('url') or [None])[-1])
 369                     if v_url:
 370                         yield self.url_result(v_url)
 371         else:
 372             playlist_id = self._search_regex(
 373                 r'''\bdata-post-id\s*=\s*("|')(?P<id>[\da-f]+)\1''',
 374                 webpage, 'id', group='id', default=None)
 375
 376             nextjs_data = self._search_nextjs_data(webpage, display_id)
 377
 378             def entries():
 379                 for player in traverse_obj(
 380                         nextjs_data,
 381                         ('props', 'apolloState', 'ROOT_QUERY', lambda k, _: k.startswith('videoPlayerProps('), '__ref')):
 382                     # skip promo links (which may not always be served, eg GH CI servers)
 383                     if traverse_obj(nextjs_data,
 384                                     ('props', 'apolloState', player.replace('PlayerProps', 'ModernContent')),
 385                                     expected_type=dict):
 386                         continue
 387                     video = traverse_obj(nextjs_data, ('props', 'apolloState', player), expected_type=dict) or {}
 388                     info = self._extract_video_info(video, fatal=False)
 389                     if info:
 390                         yield merge_dicts({
 391                             'display_id': display_id,
 392                         }, info)
 393
 394         return self.playlist_result(
 395             entries(), playlist_id or display_id,
 396             re.sub(r'\s+-\s+IGN\s*$', '', self._og_search_title(webpage, default='')) or None)