2 from __future__ import unicode_literals
8 from .common import InfoExtractor, SearchInfoExtractor
22 from .brightcove import (
26 from .nbc import NBCSportsVPlayerIE
29 class YahooIE(InfoExtractor):
30 IE_DESC = 'Yahoo screen and movies'
31 _VALID_URL = r'(?P<host>https?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?:(?P<display_id>.+)?-)?(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?'
34 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
36 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
38 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
39 'description': 'Julian and Travis watch Julian Smith',
44 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
45 'md5': '251af144a19ebc4a033e8ba91ac726bb',
47 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
49 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
50 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
53 'skip': 'HTTP Error 404',
56 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
57 'md5': '7993e572fac98e044588d0b5260f4352',
59 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
61 'title': "Yahoo Saves 'Community'",
62 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
67 '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',
68 'md5': '45c024bad51e63e9b6f6fad7a43a8c23',
70 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f',
72 'title': '敢問市長/黃秀霜批賴清德「非常高傲」',
73 'description': '直言台南沒捷運 交通居五都之末',
78 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
79 'md5': '71298482f7c64cbb7fa064e4553ff1c1',
81 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
83 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
84 'description': 'md5:f66c890e1490f4910a9953c941dee944',
89 'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html',
90 'md5': '57e06440778b1828a6079d2f744212c4',
92 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73',
94 'title': 'Program that makes hockey more affordable not offered in Manitoba',
95 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4',
100 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html',
105 'md5': '000887d0dc609bc3a47c974151a40fb8',
107 'id': 'e624c4bc-3389-34de-9dfc-025f74943409',
109 'title': '\'The Interview\' TV Spot: War',
110 'description': 'The Interview',
114 'md5': '81bc74faf10750fe36e4542f9a184c66',
116 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9',
118 'title': '\'The Interview\' TV Spot: Guys',
119 'description': 'The Interview',
124 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
125 'md5': '88e209b417f173d86186bef6e4d1f160',
127 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
129 'title': 'China Moses Is Crazy About the Blues',
130 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
134 'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html',
135 'md5': 'd9a083ccf1379127bf25699d67e4791b',
137 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c',
139 'title': 'Connect the Dots: Dark Side of Virgo',
140 'description': 'md5:1428185051cfd1949807ad4ff6d3686a',
143 'skip': 'Domain name in.lifestyle.yahoo.com gone',
145 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
146 'md5': '989396ae73d20c6f057746fb226aa215',
148 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
150 'title': '\'True Story\' Trailer',
151 'description': 'True Story',
155 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
156 'only_matching': True,
158 'note': 'NBC Sports embeds',
159 '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',
161 'id': '9CsDKds0kvHI',
163 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
164 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
165 'upload_date': '20150313',
166 'uploader': 'NBCU-SPORTS',
167 'timestamp': 1426270238,
170 'url': 'https://tw.news.yahoo.com/-100120367.html',
171 'only_matching': True,
173 # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
174 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
175 'md5': '4fbafb9c9b6f07aa8f870629f6671b35',
177 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
179 'title': 'Communitary - Community Episode 1: Ladders',
180 'description': 'md5:8fc39608213295748e1e289807838c97',
184 # it uses an alias to get the video_id
185 'url': 'https://www.yahoo.com/movies/the-stars-of-daddys-home-have-very-different-212843197.html',
187 'id': '40eda9c8-8e5f-3552-8745-830f67d0c737',
189 'title': 'Will Ferrell & Mark Wahlberg Are Pro-Spanking',
190 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.',
194 # config['models']['applet_model']['data']['sapi'] has no query
195 'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016',
196 'md5': 'dac0c72d502bc5facda80c9e6d5c98db',
198 'id': 'a6015640-e9e5-3efb-bb60-05589a183919',
200 'description': 'Galactic',
201 'title': 'Dolla Diva (feat. Maggie Koerner)',
203 'skip': 'redirect to https://www.yahoo.com/music',
207 'url': 'https://www.yahoo.com/movies/video/true-story-trailer-173000497.html',
209 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
211 'title': "'True Story' Trailer",
212 'description': 'True Story',
215 'skip_download': True,
220 'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html',
222 'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff',
224 'title': '單車天使 - 中文版預',
225 'description': '中文版預',
228 'skip_download': True,
233 'url': 'https://au.tv.yahoo.com/plus7/sunrise/-/watch/37083565/clown-entertainers-say-it-is-hurting-their-business/',
235 'id': '5575377707001',
237 'title': "Clown entertainers say 'It' is hurting their business",
238 'description': 'Stephen King s horror film has much to answer for. Jelby and Mr Loopy the Clowns join us.',
239 'timestamp': 1505341164,
240 'upload_date': '20170913',
241 'uploader_id': '2376984109001',
244 'skip_download': True,
249 def _real_extract(self, url):
250 mobj = re.match(self._VALID_URL, url)
251 page_id = mobj.group('id')
252 display_id = mobj.group('display_id') or page_id
253 host = mobj.group('host')
254 webpage, urlh = self._download_webpage_handle(url, display_id)
255 if 'err=404' in urlh.geturl():
256 raise ExtractorError('Video gone', expected=True)
258 # Look for iframed media first
260 iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
261 for idx, iframe_url in enumerate(iframe_urls):
262 entries.append(self.url_result(host + iframe_url, 'Yahoo'))
264 return self.playlist_result(entries, page_id)
266 # Look for NBCSports iframes
267 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
269 return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key())
271 # Look for Brightcove Legacy Studio embeds
272 bc_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
274 return self.url_result(bc_url, BrightcoveLegacyIE.ie_key())
276 # Look for Brightcove New Studio embeds
277 bc_url = BrightcoveNewIE._extract_url(self, webpage)
279 return self.url_result(bc_url, BrightcoveNewIE.ie_key())
281 brightcove_id = self._search_regex(
282 r'<iframe[^>]+data-video-id=["\'](\d+)', webpage, 'brightcove id',
285 return self.url_result(
286 'http://players.brightcove.net/2376984109001/default_default/index.html?videoId=%s' % brightcove_id,
287 BrightcoveNewIE.ie_key())
289 # Query result is often embedded in webpage as JSON. Sometimes explicit requests
290 # to video API results in a failure with geo restriction reason therefore using
291 # embedded query result when present sounds reasonable.
292 config_json = self._search_regex(
293 r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:</script>|$)',
294 webpage, 'videoplayer applet', default=None)
296 config = self._parse_json(config_json, display_id, fatal=False)
298 sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi')
299 if sapi and 'query' in sapi:
300 info = self._extract_info(display_id, sapi, webpage)
301 self._sort_formats(info['formats'])
304 items_json = self._search_regex(
305 r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
307 if items_json is None:
308 alias = self._search_regex(
309 r'"aliases":{"video":"(.*?)"', webpage, 'alias', default=None)
310 if alias is not None:
311 alias_info = self._download_json(
312 'https://www.yahoo.com/_td/api/resource/VideoService.videos;video_aliases=["%s"]' % alias,
313 display_id, 'Downloading alias info')
314 video_id = alias_info[0]['id']
316 CONTENT_ID_REGEXES = [
317 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
318 r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
319 r'"first_videoid"\s*:\s*"([^"]+)"',
320 r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id),
321 r'<article[^>]data-uuid=["\']([^"\']+)',
322 r'<meta[^<>]+yahoo://article/view\?.*\buuid=([^&"\']+)',
323 r'<meta[^<>]+["\']ytwnews://cavideo/(?:[^/]+/)+([\da-fA-F-]+)[&"\']',
325 video_id = self._search_regex(
326 CONTENT_ID_REGEXES, webpage, 'content ID')
328 items = json.loads(items_json)
329 info = items['mediaItems']['query']['results']['mediaObj'][0]
330 # The 'meta' field is not always in the video webpage, we request it
332 video_id = info['id']
333 return self._get_info(video_id, display_id, webpage)
335 def _extract_info(self, display_id, query, webpage):
336 info = query['query']['results']['mediaObj'][0]
337 meta = info.get('meta')
338 video_id = info.get('id')
341 msg = info['status'].get('msg')
343 raise ExtractorError(
344 '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
345 raise ExtractorError('Unable to extract media object meta')
348 for s in info['streams']:
349 tbr = int_or_none(s.get('bitrate'))
351 'width': int_or_none(s.get('width')),
352 'height': int_or_none(s.get('height')),
358 if host.startswith('rtmp'):
366 if s.get('format') == 'm3u8_playlist':
369 'protocol': 'm3u8_native',
373 fmt = format_info['ext'] = determine_ext(path)
374 format_url = compat_urlparse.urljoin(host, path)
375 format_info['url'] = format_url
376 format_info['format_id'] = fmt + ('-%d' % tbr if tbr else '')
377 formats.append(format_info)
379 closed_captions = self._html_search_regex(
380 r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions',
383 cc_json = self._parse_json(closed_captions, video_id, fatal=False)
386 for closed_caption in cc_json:
387 lang = closed_caption['lang']
388 if lang not in subtitles:
390 subtitles[lang].append({
391 'url': closed_caption['url'],
392 'ext': mimetype2ext(closed_caption['content_type']),
397 'display_id': display_id,
398 'title': unescapeHTML(meta['title']),
400 'description': clean_html(meta['description']),
401 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
402 'duration': int_or_none(meta.get('duration')),
403 'subtitles': subtitles,
406 def _get_info(self, video_id, display_id, webpage):
407 region = self._search_regex(
408 r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
409 webpage, 'region', fatal=False, default='US').upper()
412 for fmt in ('webm', 'mp4'):
413 query_result = self._download_json(
414 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' + video_id,
415 display_id, 'Downloading %s video info' % fmt, query={
420 info = self._extract_info(display_id, query_result, webpage)
421 formats.extend(info['formats'])
422 formats.extend(self._extract_m3u8_formats(
423 'http://video.media.yql.yahoo.com/v1/hls/%s?region=%s' % (video_id, region),
424 video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
425 self._sort_formats(formats)
426 info['formats'] = formats
430 class YahooSearchIE(SearchInfoExtractor):
431 IE_DESC = 'Yahoo screen search'
433 IE_NAME = 'screen.yahoo:search'
434 _SEARCH_KEY = 'yvsearch'
436 def _get_n_results(self, query, n):
437 """Get a specified number of results for a query"""
439 for pagenum in itertools.count(0):
440 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)
441 info = self._download_json(result_url, query,
442 note='Downloading results page ' + str(pagenum + 1))
444 results = info['results']
446 for (i, r) in enumerate(results):
447 if (pagenum * 30) + i >= n:
449 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
450 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
452 if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):