]>
jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/ign.py
4 from .common
import InfoExtractor
5 from ..compat
import compat_parse_qs
21 class IGNBaseIE(InfoExtractor
):
22 def _call_api(self
, slug
):
23 return self
._download
_json
(
24 'http://apis.ign.com/{0}/v3/{0}s/slug/{1}'.format(self
._PAGE
_TYPE
, slug
), slug
)
26 def _checked_call_api(self
, slug
):
28 return self
._call
_api
(slug
)
29 except ExtractorError
as e
:
30 if isinstance(e
.cause
, urllib
.error
.HTTPError
) and e
.cause
.code
== 404:
31 e
.cause
.args
= e
.cause
.args
or [
32 e
.cause
.geturl(), e
.cause
.getcode(), e
.cause
.reason
]
34 'Content not found: expired?', cause
=e
.cause
,
38 def _extract_video_info(self
, video
, fatal
=True):
39 video_id
= video
['videoId']
42 refs
= traverse_obj(video
, 'refs', expected_type
=dict) or {}
44 m3u8_url
= url_or_none(refs
.get('m3uUrl'))
46 formats
.extend(self
._extract
_m
3u8_formats
(
47 m3u8_url
, video_id
, 'mp4', 'm3u8_native',
48 m3u8_id
='hls', fatal
=False))
50 f4m_url
= url_or_none(refs
.get('f4mUrl'))
52 formats
.extend(self
._extract
_f
4m
_formats
(
53 f4m_url
, video_id
, f4m_id
='hds', fatal
=False))
55 for asset
in (video
.get('assets') or []):
56 asset_url
= url_or_none(asset
.get('url'))
61 'tbr': int_or_none(asset
.get('bitrate'), 1000),
62 'fps': int_or_none(asset
.get('frame_rate')),
63 'height': int_or_none(asset
.get('height')),
64 'width': int_or_none(asset
.get('width')),
67 mezzanine_url
= traverse_obj(
68 video
, ('system', 'mezzanineUrl'), expected_type
=url_or_none
)
71 'ext': determine_ext(mezzanine_url
, 'mp4'),
72 'format_id': 'mezzanine',
77 thumbnails
= traverse_obj(
78 video
, ('thumbnails', ..., {'url': 'url'}
), expected_type
=url_or_none
)
80 video
, ('tags', ..., 'displayName'),
81 expected_type
=lambda x
: x
.strip() or None)
83 metadata
= traverse_obj(video
, 'metadata', expected_type
=dict) or {}
85 metadata
, 'longTitle', 'title', 'name',
86 expected_type
=lambda x
: x
.strip() or None)
91 'description': strip_or_none(metadata
.get('description')),
92 'timestamp': parse_iso8601(metadata
.get('publishDate')),
93 'duration': int_or_none(metadata
.get('duration')),
94 'thumbnails': thumbnails
,
100 class IGNIE(IGNBaseIE
):
102 Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com.
103 Some videos of it.ign.com are also supported
105 _VIDEO_PATH_RE
= r
'/(?:\d{4}/\d{2}/\d{2}/)?(?P<id>.+?)'
106 _PLAYLIST_PATH_RE
= r
'(?:/?\?(?P<filt>[^&#]+))?'
108 r
'https?://(?:.+?\.ign|www\.pcmag)\.com/videos(?:%s)'
109 % '|'.join((_VIDEO_PATH_RE
+ r
'(?:[/?&#]|$)', _PLAYLIST_PATH_RE
)))
114 'url': 'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
115 'md5': 'd2e1586d9987d40fad7867bf96a018ea',
117 'id': '8f862beef863986b2785559b9e1aa599',
119 'title': 'The Last of Us Review',
120 'description': 'md5:c8946d4260a4d43a00d5ae8ed998870c',
121 'timestamp': 1370440800,
122 'upload_date': '20130605',
124 'display_id': 'the-last-of-us-review',
125 'thumbnail': 'https://assets1.ignimgs.com/vid/thumbnails/user/2014/03/26/lastofusreviewmimig2.jpg',
129 'nocheckcertificate': True,
132 'url': 'http://www.pcmag.com/videos/2015/01/06/010615-whats-new-now-is-gogo-snooping-on-your-data',
133 'md5': 'f1581a6fe8c5121be5b807684aeac3f6',
135 'id': 'ee10d774b508c9b8ec07e763b9125b91',
137 'title': 'What\'s New Now: Is GoGo Snooping on Your Data?',
138 'description': 'md5:817a20299de610bd56f13175386da6fa',
139 'timestamp': 1420571160,
140 'upload_date': '20150106',
143 'skip': '404 Not Found',
145 'url': 'https://www.ign.com/videos/is-a-resident-evil-4-remake-on-the-way-ign-daily-fix',
146 'only_matching': True,
150 def _extract_embed_urls(cls
, url
, webpage
):
152 r
'''(?s)<section\b[^>]+\bclass\s*=\s*['"](?:[\w-]+\s+)*?content-feed-grid(?!\B|-)[^>]+>(.+?)</section[^>]*>''',
155 (urljoin(url
, m
.group('path')) for m
in re
.finditer(
156 r
'''<a\b[^>]+\bhref\s*=\s*('|")(?P<path>/videos%s)\1'''
157 % cls
._VIDEO
_PATH
_RE
, grids
[0] if grids
else '')))
159 def _real_extract(self
, url
):
160 display_id
, filt
= self
._match
_valid
_url
(url
).group('id', 'filt')
162 return self
._extract
_video
(url
, display_id
)
163 return self
._extract
_playlist
(url
, filt
or 'all')
165 def _extract_playlist(self
, url
, display_id
):
166 webpage
= self
._download
_webpage
(url
, display_id
)
168 return self
.playlist_result(
169 (self
.url_result(u
, self
.ie_key())
170 for u
in self
._extract
_embed
_urls
(url
, webpage
)),
171 playlist_id
=display_id
)
173 def _extract_video(self
, url
, display_id
):
174 video
= self
._checked
_call
_api
(display_id
)
176 info
= self
._extract
_video
_info
(video
)
179 'display_id': display_id
,
183 class IGNVideoIE(IGNBaseIE
):
184 _VALID_URL
= r
'https?://.+?\.ign\.com/(?:[a-z]{2}/)?[^/]+/(?P<id>\d+)/(?:video|trailer)/'
186 'url': 'http://me.ign.com/en/videos/112203/video/how-hitman-aims-to-be-different-than-every-other-s',
187 'md5': 'dd9aca7ed2657c4e118d8b261e5e9de1',
189 'id': 'e9be7ea899a9bbfc0674accc22a36cc8',
191 'title': 'How Hitman Aims to Be Different Than Every Other Stealth Game - NYCC 2015',
192 'description': 'Taking out assassination targets in Hitman has never been more stylish.',
193 'timestamp': 1444665600,
194 'upload_date': '20151012',
195 'display_id': '112203',
196 'thumbnail': 'https://sm.ign.com/ign_me/video/h/how-hitman/how-hitman-aims-to-be-different-than-every-other-s_8z14.jpg',
199 'display_id': '112203',
200 'thumbnail': 'https://sm.ign.com/ign_me/video/h/how-hitman/how-hitman-aims-to-be-different-than-every-other-s_8z14.jpg',
204 'expected_warnings': ['HTTP Error 400: Bad Request'],
206 'url': 'http://me.ign.com/ar/angry-birds-2/106533/video/lrd-ldyy-lwl-lfylm-angry-birds',
207 'only_matching': True,
210 'url': 'https://me.ign.com/ar/ratchet-clank-rift-apart/144327/trailer/embed',
211 'only_matching': True,
214 'url': 'http://adria.ign.com/sherlock-season-4/9687/trailer/embed',
215 'only_matching': True,
218 'url': 'https://kr.ign.com/bic-2018/3307/trailer/embed',
219 'only_matching': True,
222 def _real_extract(self
, url
):
223 video_id
= self
._match
_id
(url
)
224 parsed_url
= urllib
.parse
.urlparse(url
)
225 embed_url
= urllib
.parse
.urlunparse(
226 parsed_url
._replace
(path
=parsed_url
.path
.rsplit('/', 1)[0] + '/embed'))
228 webpage
, urlh
= self
._download
_webpage
_handle
(embed_url
, video_id
)
229 new_url
= urlh
.geturl()
230 ign_url
= compat_parse_qs(
231 urllib
.parse
.urlparse(new_url
).query
).get('url', [None])[-1]
233 return self
.url_result(ign_url
, IGNIE
.ie_key())
234 video
= self
._search
_regex
(r
'(<div\b[^>]+\bdata-video-id\s*=\s*[^>]+>)', webpage
, 'video element', fatal
=False)
237 raise ExtractorError('Redirect loop: ' + url
)
238 return self
.url_result(new_url
)
239 video
= extract_attributes(video
)
240 video_data
= video
.get('data-settings') or '{}'
241 video_data
= self
._parse
_json
(video_data
, video_id
)['video']
242 info
= self
._extract
_video
_info
(video_data
)
245 'display_id': video_id
,
249 class IGNArticleIE(IGNBaseIE
):
250 _VALID_URL
= r
'https?://.+?\.ign\.com/(?:articles(?:/\d{4}/\d{2}/\d{2})?|(?:[a-z]{2}/)?(?:[\w-]+/)*?feature/\d+)/(?P<id>[^/?&#]+)'
251 _PAGE_TYPE
= 'article'
253 'url': 'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind',
256 'title': '100 Little Things in GTA 5 That Will Blow Your Mind',
261 'id': '5ebbd138523268b93c9141af17bec937',
263 'title': 'Grand Theft Auto V Video Review',
264 'description': 'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.',
265 'timestamp': 1379339880,
266 'upload_date': '20130916',
268 'thumbnail': 'https://assets1.ignimgs.com/thumbs/userUploaded/2021/8/16/gta-v-heistsjpg-e94705-1629138553533.jpeg',
269 'display_id': 'grand-theft-auto-v-video-review',
275 'id': '638672ee848ae4ff108df2a296418ee2',
277 'title': 'GTA 5 In Slow Motion',
278 'description': 'The twisted beauty of GTA 5 in stunning slow motion.',
279 'timestamp': 1386878820,
280 'upload_date': '20131212',
283 'display_id': 'gta-5-in-slow-motion',
284 'thumbnail': 'https://assets1.ignimgs.com/vid/thumbnails/user/2013/11/03/GTA-SLO-MO-1.jpg',
289 'skip_download': True,
291 'expected_warnings': ['Backend fetch failed'],
293 'url': 'http://www.ign.com/articles/2014/08/15/rewind-theater-wild-trailer-gamescom-2014?watch',
295 'id': '53ee806780a81ec46e0790f8',
296 'title': 'Rewind Theater - Wild Trailer Gamescom 2014',
299 'expected_warnings': ['Backend fetch failed'],
302 'url': 'http://www.ign.com/articles/2017/06/08/new-ducktales-short-donalds-birthday-doesnt-go-as-planned',
303 'only_matching': True,
306 'url': 'https://www.ign.com/articles/2021-mvp-named-in-puppy-bowl-xvii',
307 'only_matching': True,
310 'url': 'https://www.ign.com/articles/2014/08/07/sons-of-anarchy-final-season-trailer',
311 'only_matching': True,
314 'url': 'https://www.ign.com/articles/2017/09/20/marvels-the-punisher-watch-the-new-trailer-for-the-netflix-series',
315 'only_matching': True,
318 'url': 'https://www.ign.com/articles/2016/01/16/supergirl-goes-flying-with-martian-manhunter-in-new-clip',
319 'only_matching': True,
322 def _checked_call_api(self
, slug
):
324 return self
._call
_api
(slug
)
325 except ExtractorError
as e
:
326 if isinstance(e
.cause
, urllib
.error
.HTTPError
):
327 e
.cause
.args
= e
.cause
.args
or [
328 e
.cause
.geturl(), e
.cause
.getcode(), e
.cause
.reason
]
329 if e
.cause
.code
== 404:
330 raise ExtractorError(
331 'Content not found: expired?', cause
=e
.cause
,
333 elif e
.cause
.code
== 503:
334 self
.report_warning(error_to_compat_str(e
.cause
))
338 def _real_extract(self
, url
):
339 display_id
= self
._match
_id
(url
)
340 article
= self
._checked
_call
_api
(display_id
)
345 media_url
= traverse_obj(
346 article
, ('mediaRelations', 0, 'media', 'metadata', 'url'),
347 expected_type
=url_or_none
)
349 yield self
.url_result(media_url
, IGNIE
.ie_key())
350 for content
in (article
.get('content') or []):
351 for video_url
in re
.findall(r
'(?:\[(?:ignvideo\s+url|youtube\s+clip_id)|<iframe[^>]+src)="([^"]+)"', content
):
352 if url_or_none(video_url
):
353 yield self
.url_result(video_url
)
355 return self
.playlist_result(
356 entries(), article
.get('articleId'),
358 article
, ('metadata', 'headline'),
359 expected_type
=lambda x
: x
.strip() or None))
361 webpage
= self
._download
_webpage
(url
, display_id
)
363 playlist_id
= self
._html
_search
_meta
('dable:item_id', webpage
, default
=None)
367 for m
in re
.finditer(
368 r
'''(?s)<object\b[^>]+\bclass\s*=\s*("|')ign-videoplayer\1[^>]*>(?P<params>.+?)</object''',
370 flashvars
= self
._search
_regex
(
371 r
'''(<param\b[^>]+\bname\s*=\s*("|')flashvars\2[^>]*>)''',
372 m
.group('params'), 'flashvars', default
='')
373 flashvars
= compat_parse_qs(extract_attributes(flashvars
).get('value') or '')
374 v_url
= url_or_none((flashvars
.get('url') or [None])[-1])
376 yield self
.url_result(v_url
)
378 playlist_id
= self
._search
_regex
(
379 r
'''\bdata-post-id\s*=\s*("|')(?P<id>[\da-f]+)\1''',
380 webpage
, 'id', group
='id', default
=None)
382 nextjs_data
= self
._search
_nextjs
_data
(webpage
, display_id
)
385 for player
in traverse_obj(
387 ('props', 'apolloState', 'ROOT_QUERY', lambda k
, _
: k
.startswith('videoPlayerProps('), '__ref')):
388 # skip promo links (which may not always be served, eg GH CI servers)
389 if traverse_obj(nextjs_data
,
390 ('props', 'apolloState', player
.replace('PlayerProps', 'ModernContent')),
393 video
= traverse_obj(nextjs_data
, ('props', 'apolloState', player
), expected_type
=dict) or {}
394 info
= self
._extract
_video
_info
(video
, fatal
=False)
397 'display_id': display_id
,
400 return self
.playlist_result(
401 entries(), playlist_id
or display_id
,
402 re
.sub(r
'\s+-\s+IGN\s*$', '', self
._og
_search
_title
(webpage
, default
='')) or None)