]>
Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yahoo.py
2 from __future__
import unicode_literals
8 from .common
import InfoExtractor
, SearchInfoExtractor
24 from .brightcove
import (
28 from .nbc
import NBCSportsVPlayerIE
31 class YahooIE(InfoExtractor
):
32 IE_DESC
= 'Yahoo screen and movies'
33 _VALID_URL
= r
'(?P<host>https?://(?:(?P<country>[a-zA-Z]{2})\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?:(?P<display_id>.+)?-)?(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?'
36 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
38 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
40 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
41 'description': 'Julian and Travis watch Julian Smith',
46 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
47 'md5': '251af144a19ebc4a033e8ba91ac726bb',
49 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
51 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
52 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
55 'skip': 'HTTP Error 404',
58 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
59 'md5': '7993e572fac98e044588d0b5260f4352',
61 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
63 'title': "Yahoo Saves 'Community'",
64 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
69 'url': 'https://tw.news.yahoo.com/%E6%95%A2%E5%95%8F%E5%B8%82%E9%95%B7%20%E9%BB%83%E7%A7%80%E9%9C%9C%E6%89%B9%E8%B3%B4%E6%B8%85%E5%BE%B7%20%E9%9D%9E%E5%B8%B8%E9%AB%98%E5%82%B2-034024051.html',
70 'md5': '45c024bad51e63e9b6f6fad7a43a8c23',
72 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f',
74 'title': '敢問市長/黃秀霜批賴清德「非常高傲」',
75 'description': '直言台南沒捷運 交通居五都之末',
80 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
81 'md5': '71298482f7c64cbb7fa064e4553ff1c1',
83 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
85 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
86 'description': 'md5:f66c890e1490f4910a9953c941dee944',
91 'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html',
92 'md5': '57e06440778b1828a6079d2f744212c4',
94 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73',
96 'title': 'Program that makes hockey more affordable not offered in Manitoba',
97 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4',
100 'skip': 'Video gone',
102 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html',
107 'md5': '000887d0dc609bc3a47c974151a40fb8',
109 'id': 'e624c4bc-3389-34de-9dfc-025f74943409',
111 'title': '\'The Interview\' TV Spot: War',
112 'description': 'The Interview',
116 'md5': '81bc74faf10750fe36e4542f9a184c66',
118 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9',
120 'title': '\'The Interview\' TV Spot: Guys',
121 'description': 'The Interview',
126 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
127 'md5': '88e209b417f173d86186bef6e4d1f160',
129 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
131 'title': 'China Moses Is Crazy About the Blues',
132 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
136 'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html',
137 'md5': 'd9a083ccf1379127bf25699d67e4791b',
139 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c',
141 'title': 'Connect the Dots: Dark Side of Virgo',
142 'description': 'md5:1428185051cfd1949807ad4ff6d3686a',
145 'skip': 'Domain name in.lifestyle.yahoo.com gone',
147 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
148 'md5': '989396ae73d20c6f057746fb226aa215',
150 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
152 'title': '\'True Story\' Trailer',
153 'description': 'True Story',
157 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
158 'only_matching': True,
160 'note': 'NBC Sports embeds',
161 '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',
163 'id': '9CsDKds0kvHI',
165 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
166 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
167 'upload_date': '20150313',
168 'uploader': 'NBCU-SPORTS',
169 'timestamp': 1426270238,
172 'url': 'https://tw.news.yahoo.com/-100120367.html',
173 'only_matching': True,
175 # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
176 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
177 'md5': '4fbafb9c9b6f07aa8f870629f6671b35',
179 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
181 'title': 'Communitary - Community Episode 1: Ladders',
182 'description': 'md5:8fc39608213295748e1e289807838c97',
186 # it uses an alias to get the video_id
187 'url': 'https://www.yahoo.com/movies/the-stars-of-daddys-home-have-very-different-212843197.html',
189 'id': '40eda9c8-8e5f-3552-8745-830f67d0c737',
191 'title': 'Will Ferrell & Mark Wahlberg Are Pro-Spanking',
192 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.',
196 # config['models']['applet_model']['data']['sapi'] has no query
197 'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016',
198 'md5': 'dac0c72d502bc5facda80c9e6d5c98db',
200 'id': 'a6015640-e9e5-3efb-bb60-05589a183919',
202 'description': 'Galactic',
203 'title': 'Dolla Diva (feat. Maggie Koerner)',
205 'skip': 'redirect to https://www.yahoo.com/music',
209 'url': 'https://www.yahoo.com/movies/video/true-story-trailer-173000497.html',
211 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
213 'title': "'True Story' Trailer",
214 'description': 'True Story',
217 'skip_download': True,
222 'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html',
224 'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff',
226 'title': '單車天使 - 中文版預',
227 'description': '中文版預',
230 'skip_download': True,
235 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37083565/clown-entertainers-say-it-is-hurting-their-business/',
237 'id': '5575377707001',
239 'title': "Clown entertainers say 'It' is hurting their business",
240 'description': 'Stephen King s horror film has much to answer for. Jelby and Mr Loopy the Clowns join us.',
241 'timestamp': 1505341164,
242 'upload_date': '20170913',
243 'uploader_id': '2376984109001',
246 'skip_download': True,
250 # custom brightcove, geo-restricted to Australia, bypassable
251 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37263964/sunrise-episode-wed-27-sep/',
252 'only_matching': True,
256 def _real_extract(self
, url
):
257 mobj
= re
.match(self
._VALID
_URL
, url
)
258 page_id
= mobj
.group('id')
259 display_id
= mobj
.group('display_id') or page_id
260 host
= mobj
.group('host')
261 webpage
, urlh
= self
._download
_webpage
_handle
(url
, display_id
)
262 if 'err=404' in urlh
.geturl():
263 raise ExtractorError('Video gone', expected
=True)
265 # Look for iframed media first
267 iframe_urls
= re
.findall(r
'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage
)
268 for idx
, iframe_url
in enumerate(iframe_urls
):
269 entries
.append(self
.url_result(host
+ iframe_url
, 'Yahoo'))
271 return self
.playlist_result(entries
, page_id
)
273 # Look for NBCSports iframes
274 nbc_sports_url
= NBCSportsVPlayerIE
._extract
_url
(webpage
)
276 return self
.url_result(nbc_sports_url
, NBCSportsVPlayerIE
.ie_key())
278 # Look for Brightcove Legacy Studio embeds
279 bc_url
= BrightcoveLegacyIE
._extract
_brightcove
_url
(webpage
)
281 return self
.url_result(bc_url
, BrightcoveLegacyIE
.ie_key())
283 def brightcove_url_result(bc_url
):
284 return self
.url_result(
285 smuggle_url(bc_url
, {'geo_countries': [mobj
.group('country')]}),
286 BrightcoveNewIE
.ie_key())
288 # Look for Brightcove New Studio embeds
289 bc_url
= BrightcoveNewIE
._extract
_url
(self
, webpage
)
291 return brightcove_url_result(bc_url
)
293 brightcove_iframe
= self
._search
_regex
(
294 r
'(<iframe[^>]+data-video-id=["\']\d
+[^
>]+>)', webpage,
295 'brightcove iframe
', default=None)
296 if brightcove_iframe:
297 attr = extract_attributes(brightcove_iframe)
298 src = attr.get('src
')
300 parsed_src = compat_urlparse.urlparse(src)
301 qs = compat_urlparse.parse_qs(parsed_src.query)
302 account_id = qs.get('accountId
', ['2376984109001'])[0]
303 brightcove_id = attr.get('data
-video
-id') or qs.get('videoId
', [None])[0]
304 if account_id and brightcove_id:
305 return brightcove_url_result(
306 'http
://players
.brightcove
.net
/%s/default_default
/index
.html?videoId
=%s'
307 % (account_id, brightcove_id))
309 # Query result is often embedded in webpage as JSON. Sometimes explicit requests
310 # to video API results in a failure with geo restriction reason therefore using
311 # embedded query result when present sounds reasonable.
312 config_json = self._search_regex(
313 r'window\
.Af\
.bootstrap\
[[^\
]]+\
]\s
*=\s
*({.*?
"applet_type"\s
*:\s
*"td-applet-videoplayer".*?
});(?
:</script
>|$
)',
314 webpage, 'videoplayer applet
', default=None)
316 config = self._parse_json(config_json, display_id, fatal=False)
318 sapi = config.get('models
', {}).get('applet_model
', {}).get('data
', {}).get('sapi
')
319 if sapi and 'query
' in sapi:
320 info = self._extract_info(display_id, sapi, webpage)
321 self._sort_formats(info['formats
'])
324 items_json = self._search_regex(
325 r'mediaItems
: ({.*?
})$
', webpage, 'items
', flags=re.MULTILINE,
327 if items_json is None:
328 alias = self._search_regex(
329 r'"aliases":{"video":"(.*?)"', webpage, 'alias
', default=None)
330 if alias is not None:
331 alias_info = self._download_json(
332 'https
://www
.yahoo
.com
/_td
/api
/resource
/VideoService
.videos
;video_aliases
=["%s"]' % alias,
333 display_id, 'Downloading alias info
')
334 video_id = alias_info[0]['id']
336 CONTENT_ID_REGEXES = [
337 r'YUI\
.namespace\
("Media"\
)\
.CONTENT_ID\s
*=\s
*"([^"]+)"',
338 r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^
"]+)"',
339 r'"first_videoid"\s
*:\s
*"([^"]+)"',
340 r'%s[^}]*"ccm_id
"\s*:\s*"([^
"]+)"' % re.escape(page_id),
341 r'<article
[^
>]data
-uuid
=["\']([^"\']+)',
342 r'<meta
[^
<>]+yahoo
://article
/view
\?.*\buuid
=([^
&"\']+)',
343 r'<meta[^<>]+["\']ytwnews
://cavideo
/(?
:[^
/]+/)+([\da
-fA
-F
-]+)[&"\']',
345 video_id = self._search_regex(
346 CONTENT_ID_REGEXES, webpage, 'content ID')
348 items = json.loads(items_json)
349 info = items['mediaItems']['query']['results']['mediaObj'][0]
350 # The 'meta' field is not always in the video webpage, we request it
352 video_id = info['id']
353 return self._get_info(video_id, display_id, webpage)
355 def _extract_info(self, display_id, query, webpage):
356 info = query['query']['results']['mediaObj'][0]
357 meta = info.get('meta')
358 video_id = info.get('id')
361 msg = info['status'].get('msg')
363 raise ExtractorError(
364 '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
365 raise ExtractorError('Unable to extract media object meta')
368 for s in info['streams']:
369 tbr = int_or_none(s.get('bitrate'))
371 'width': int_or_none(s.get('width')),
372 'height': int_or_none(s.get('height')),
378 if host.startswith('rtmp'):
386 if s.get('format') == 'm3u8_playlist':
389 'protocol': 'm3u8_native',
393 fmt = format_info['ext'] = determine_ext(path)
394 format_url = compat_urlparse.urljoin(host, path)
395 format_info['url'] = format_url
396 format_info['format_id'] = fmt + ('-%d' % tbr if tbr else '')
397 formats.append(format_info)
399 closed_captions = self._html_search_regex(
400 r'"closedcaptions
":(\[[^\]]+\])', webpage, 'closed captions',
403 cc_json = self._parse_json(closed_captions, video_id, fatal=False)
406 for closed_caption in cc_json:
407 lang = closed_caption['lang']
408 if lang not in subtitles:
410 subtitles[lang].append({
411 'url': closed_caption['url'],
412 'ext': mimetype2ext(closed_caption['content_type']),
417 'display_id': display_id,
418 'title': unescapeHTML(meta['title']),
420 'description': clean_html(meta['description']),
421 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
422 'duration': int_or_none(meta.get('duration')),
423 'subtitles': subtitles,
426 def _get_info(self, video_id, display_id, webpage):
427 region = self._search_regex(
428 r'\\?"region
\\?
"\s*:\s*\\?"([^
"]+?)\\?"',
429 webpage, 'region
', fatal=False, default='US
').upper()
432 for fmt in ('webm
', 'mp4
'):
433 query_result = self._download_json(
434 'https
://video
.media
.yql
.yahoo
.com
/v1
/video
/sapi
/streams
/' + video_id,
435 display_id, 'Downloading
%s video info
' % fmt, query={
440 info = self._extract_info(display_id, query_result, webpage)
441 formats.extend(info['formats
'])
442 formats.extend(self._extract_m3u8_formats(
443 'http
://video
.media
.yql
.yahoo
.com
/v1
/hls
/%s?region
=%s' % (video_id, region),
444 video_id, 'mp4
', 'm3u8_native
', m3u8_id='hls
', fatal=False))
445 self._sort_formats(formats)
446 info['formats
'] = formats
450 class YahooSearchIE(SearchInfoExtractor):
451 IE_DESC = 'Yahoo screen search
'
453 IE_NAME = 'screen
.yahoo
:search
'
454 _SEARCH_KEY = 'yvsearch
'
456 def _get_n_results(self, query, n):
457 """Get a specified number of results for a query"""
459 for pagenum in itertools.count(0):
460 result_url = 'http
://video
.search
.yahoo
.com
/search
/?p
=%s&fr
=screen
&o
=js
&gs
=0&b
=%d' % (compat_urllib_parse.quote_plus(query), pagenum * 30)
461 info = self._download_json(result_url, query,
462 note='Downloading results page
' + str(pagenum + 1))
464 results = info['results
']
466 for (i, r) in enumerate(results):
467 if (pagenum * 30) + i >= n:
469 mobj = re.search(r'(?P
<url
>screen\
.yahoo\
.com
/.*?
-\d
*?\
.html
)"', r)
470 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
472 if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):