2 from __future__ import unicode_literals
8 from .common import InfoExtractor, SearchInfoExtractor
11 compat_urllib_parse_urlencode,
22 from .nbc import NBCSportsVPlayerIE
25 class YahooIE(InfoExtractor):
26 IE_DESC = 'Yahoo screen and movies'
27 _VALID_URL = r'(?P<url>(?P<host>https?://(?:[a-zA-Z]{2}\.)?[\da-zA-Z_-]+\.yahoo\.com)/(?:[^/]+/)*(?P<display_id>.+)?-(?P<id>[0-9]+)(?:-[a-z]+)?(?:\.html)?)'
30 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
32 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
34 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
35 'description': 'Julian and Travis watch Julian Smith',
40 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html',
41 'md5': 'c3466d2b6d5dd6b9f41ba9ed04c24b23',
43 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9',
45 'title': 'Codefellas - The Cougar Lies with Spanish Moss',
46 'description': 'md5:66b627ab0a282b26352136ca96ce73c1',
51 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
52 'md5': '75ffabdb87c16d4ffe8c036dc4d1c136',
54 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
56 'title': "Yahoo Saves 'Community'",
57 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
62 'url': 'https://tw.screen.yahoo.com/election-2014-askmayor/敢問市長-黃秀霜批賴清德-非常高傲-033009720.html',
63 'md5': '3a09cf59349cfaddae1797acc3c087fc',
65 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f',
67 'title': '敢問市長/黃秀霜批賴清德「非常高傲」',
68 'description': '直言台南沒捷運 交通居五都之末',
73 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
74 'md5': '0b51660361f0e27c9789e7037ef76f4b',
76 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
78 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
79 'description': 'md5:f66c890e1490f4910a9953c941dee944',
84 'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html',
85 'md5': '57e06440778b1828a6079d2f744212c4',
87 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73',
89 'title': 'Program that makes hockey more affordable not offered in Manitoba',
90 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4',
94 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html',
95 'md5': '226a895aae7e21b0129e2a2006fe9690',
97 'id': 'e624c4bc-3389-34de-9dfc-025f74943409',
99 'title': '\'The Interview\' TV Spot: War',
100 'description': 'The Interview',
104 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
105 'md5': '88e209b417f173d86186bef6e4d1f160',
107 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
109 'title': 'China Moses Is Crazy About the Blues',
110 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
114 'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html',
115 'md5': 'd9a083ccf1379127bf25699d67e4791b',
117 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c',
119 'title': 'Connect the Dots: Dark Side of Virgo',
120 'description': 'md5:1428185051cfd1949807ad4ff6d3686a',
124 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
125 'md5': 'b17ac378b1134fa44370fb27db09a744',
127 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
129 'title': '\'True Story\' Trailer',
130 'description': 'True Story',
134 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
135 'only_matching': True,
137 'note': 'NBC Sports embeds',
138 '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',
140 'id': '9CsDKds0kvHI',
142 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
143 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
144 'upload_date': '20150313',
145 'uploader': 'NBCU-SPORTS',
146 'timestamp': 1426270238,
149 'url': 'https://tw.news.yahoo.com/-100120367.html',
150 'only_matching': True,
152 # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
153 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
154 'md5': '1ddbf7c850777548438e5c4f147c7b8c',
156 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
158 'title': 'Communitary - Community Episode 1: Ladders',
159 'description': 'md5:8fc39608213295748e1e289807838c97',
163 # it uses an alias to get the video_id
164 'url': 'https://www.yahoo.com/movies/the-stars-of-daddys-home-have-very-different-212843197.html',
166 'id': '40eda9c8-8e5f-3552-8745-830f67d0c737',
168 'title': 'Will Ferrell & Mark Wahlberg Are Pro-Spanking',
169 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.',
173 # config['models']['applet_model']['data']['sapi'] has no query
174 'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016',
175 'md5': 'dac0c72d502bc5facda80c9e6d5c98db',
177 'id': 'a6015640-e9e5-3efb-bb60-05589a183919',
179 'description': 'Galactic',
180 'title': 'Dolla Diva (feat. Maggie Koerner)',
185 def _real_extract(self, url):
186 mobj = re.match(self._VALID_URL, url)
187 display_id = mobj.group('display_id') or self._match_id(url)
188 page_id = mobj.group('id')
189 url = mobj.group('url')
190 host = mobj.group('host')
191 webpage = self._download_webpage(url, display_id)
193 # Look for iframed media first
194 iframe_m = re.search(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
196 iframepage = self._download_webpage(
197 host + iframe_m.group(1), display_id, 'Downloading iframe webpage')
198 items_json = self._search_regex(
199 r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None)
201 items = json.loads(items_json)
202 video_id = items[0]['id']
203 return self._get_info(video_id, display_id, webpage)
204 # Look for NBCSports iframes
205 nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
207 return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
209 # Query result is often embedded in webpage as JSON. Sometimes explicit requests
210 # to video API results in a failure with geo restriction reason therefore using
211 # embedded query result when present sounds reasonable.
212 config_json = self._search_regex(
213 r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:</script>|$)',
214 webpage, 'videoplayer applet', default=None)
216 config = self._parse_json(config_json, display_id, fatal=False)
218 sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi')
219 if sapi and 'query' in sapi:
220 return self._extract_info(display_id, sapi, webpage)
222 items_json = self._search_regex(
223 r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
225 if items_json is None:
226 alias = self._search_regex(
227 r'"aliases":{"video":"(.*?)"', webpage, 'alias', default=None)
228 if alias is not None:
229 alias_info = self._download_json(
230 'https://www.yahoo.com/_td/api/resource/VideoService.videos;video_aliases=["%s"]' % alias,
231 display_id, 'Downloading alias info')
232 video_id = alias_info[0]['id']
234 CONTENT_ID_REGEXES = [
235 r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
236 r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
237 r'"first_videoid"\s*:\s*"([^"]+)"',
238 r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id),
239 r'<article[^>]data-uuid=["\']([^"\']+)',
240 r'yahoo://article/view\?.*\buuid=([^&"\']+)',
242 video_id = self._search_regex(
243 CONTENT_ID_REGEXES, webpage, 'content ID')
245 items = json.loads(items_json)
246 info = items['mediaItems']['query']['results']['mediaObj'][0]
247 # The 'meta' field is not always in the video webpage, we request it
249 video_id = info['id']
250 return self._get_info(video_id, display_id, webpage)
252 def _extract_info(self, display_id, query, webpage):
253 info = query['query']['results']['mediaObj'][0]
254 meta = info.get('meta')
255 video_id = info.get('id')
258 msg = info['status'].get('msg')
260 raise ExtractorError(
261 '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
262 raise ExtractorError('Unable to extract media object meta')
265 for s in info['streams']:
267 'width': int_or_none(s.get('width')),
268 'height': int_or_none(s.get('height')),
269 'tbr': int_or_none(s.get('bitrate')),
274 if host.startswith('rtmp'):
281 if s.get('format') == 'm3u8_playlist':
282 format_info['protocol'] = 'm3u8_native'
283 format_info['ext'] = 'mp4'
284 format_url = compat_urlparse.urljoin(host, path)
285 format_info['url'] = format_url
286 formats.append(format_info)
288 self._sort_formats(formats)
290 closed_captions = self._html_search_regex(
291 r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions',
294 cc_json = self._parse_json(closed_captions, video_id, fatal=False)
297 for closed_caption in cc_json:
298 lang = closed_caption['lang']
299 if lang not in subtitles:
301 subtitles[lang].append({
302 'url': closed_caption['url'],
303 'ext': mimetype2ext(closed_caption['content_type']),
308 'display_id': display_id,
309 'title': unescapeHTML(meta['title']),
311 'description': clean_html(meta['description']),
312 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
313 'duration': int_or_none(meta.get('duration')),
314 'subtitles': subtitles,
317 def _get_info(self, video_id, display_id, webpage):
318 region = self._search_regex(
319 r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
320 webpage, 'region', fatal=False, default='US')
321 data = compat_urllib_parse_urlencode({
326 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/'
327 '{id}?{data}'.format(id=video_id, data=data))
328 query_result = self._download_json(
329 query_url, display_id, 'Downloading video info')
330 return self._extract_info(display_id, query_result, webpage)
333 class YahooSearchIE(SearchInfoExtractor):
334 IE_DESC = 'Yahoo screen search'
336 IE_NAME = 'screen.yahoo:search'
337 _SEARCH_KEY = 'yvsearch'
339 def _get_n_results(self, query, n):
340 """Get a specified number of results for a query"""
342 for pagenum in itertools.count(0):
343 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)
344 info = self._download_json(result_url, query,
345 note='Downloading results page ' + str(pagenum + 1))
347 results = info['results']
349 for (i, r) in enumerate(results):
350 if (pagenum * 30) + i >= n:
352 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
353 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
355 if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):