]>
Commit | Line | Data |
---|---|---|
762958d5 | 1 | # coding: utf-8 |
2317e6b2 AS |
2 | from __future__ import unicode_literals |
3 | ||
934858ad | 4 | import itertools |
15251481 | 5 | import json |
d6039175 PH |
6 | import re |
7 | ||
934858ad | 8 | from .common import InfoExtractor, SearchInfoExtractor |
1cc79574 | 9 | from ..compat import ( |
934858ad | 10 | compat_urllib_parse, |
15707c7e | 11 | compat_urllib_parse_urlencode, |
9c15e9de | 12 | compat_urlparse, |
1cc79574 PH |
13 | ) |
14 | from ..utils import ( | |
9c15e9de | 15 | clean_html, |
3aa9176f | 16 | unescapeHTML, |
1cc79574 | 17 | ExtractorError, |
7217e148 | 18 | int_or_none, |
ecee5724 | 19 | mimetype2ext, |
d6039175 PH |
20 | ) |
21 | ||
6a4e659c | 22 | from .brightcove import BrightcoveNewIE |
a2a4d5fa YCH |
23 | from .nbc import NBCSportsVPlayerIE |
24 | ||
9c15e9de | 25 | |
d6039175 | 26 | class YahooIE(InfoExtractor): |
a4eb9578 | 27 | IE_DESC = 'Yahoo screen and movies' |
2a7c3883 | 28 | _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)?)' |
9c15e9de JMF |
29 | _TESTS = [ |
30 | { | |
2317e6b2 | 31 | 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html', |
2317e6b2 | 32 | 'info_dict': { |
fb6a5b96 | 33 | 'id': '2d25e626-2378-391f-ada0-ddaf1417e588', |
88f1c6de | 34 | 'ext': 'mp4', |
2317e6b2 AS |
35 | 'title': 'Julian Smith & Travis Legg Watch Julian Smith', |
36 | 'description': 'Julian and Travis watch Julian Smith', | |
762958d5 | 37 | 'duration': 6863, |
9c15e9de JMF |
38 | }, |
39 | }, | |
40 | { | |
2317e6b2 | 41 | 'url': 'http://screen.yahoo.com/wired/codefellas-s1-ep12-cougar-lies-103000935.html', |
4f549580 | 42 | 'md5': 'c3466d2b6d5dd6b9f41ba9ed04c24b23', |
2317e6b2 | 43 | 'info_dict': { |
fb6a5b96 | 44 | 'id': 'd1dedf8c-d58c-38c3-8963-e899929ae0a9', |
88f1c6de | 45 | 'ext': 'mp4', |
2317e6b2 | 46 | 'title': 'Codefellas - The Cougar Lies with Spanish Moss', |
762958d5 S |
47 | 'description': 'md5:66b627ab0a282b26352136ca96ce73c1', |
48 | 'duration': 151, | |
9c15e9de | 49 | }, |
6f5ac90c | 50 | }, |
49807b4a S |
51 | { |
52 | 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed', | |
4f549580 | 53 | 'md5': '75ffabdb87c16d4ffe8c036dc4d1c136', |
49807b4a S |
54 | 'info_dict': { |
55 | 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb', | |
56 | 'ext': 'mp4', | |
57 | 'title': "Yahoo Saves 'Community'", | |
58 | 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053', | |
762958d5 S |
59 | 'duration': 170, |
60 | } | |
61 | }, | |
62 | { | |
2a7dee8c YCH |
63 | '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', |
64 | 'md5': '9035d38f88b1782682a3e89f985be5bb', | |
762958d5 | 65 | 'info_dict': { |
85fab7e4 | 66 | 'id': 'cac903b3-fcf4-3c14-b632-643ab541712f', |
762958d5 | 67 | 'ext': 'mp4', |
85fab7e4 JMF |
68 | 'title': '敢問市長/黃秀霜批賴清德「非常高傲」', |
69 | 'description': '直言台南沒捷運 交通居五都之末', | |
70 | 'duration': 396, | |
2a7dee8c | 71 | }, |
762958d5 S |
72 | }, |
73 | { | |
3c06d371 | 74 | 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html', |
762958d5 S |
75 | 'md5': '0b51660361f0e27c9789e7037ef76f4b', |
76 | 'info_dict': { | |
77 | 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58', | |
78 | 'ext': 'mp4', | |
79 | 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder', | |
80 | 'description': 'md5:f66c890e1490f4910a9953c941dee944', | |
81 | 'duration': 97, | |
49807b4a S |
82 | } |
83 | }, | |
3c06d371 S |
84 | { |
85 | 'url': 'https://ca.sports.yahoo.com/video/program-makes-hockey-more-affordable-013127711.html', | |
86 | 'md5': '57e06440778b1828a6079d2f744212c4', | |
87 | 'info_dict': { | |
88 | 'id': 'c9fa2a36-0d4d-3937-b8f6-cc0fb1881e73', | |
89 | 'ext': 'mp4', | |
90 | 'title': 'Program that makes hockey more affordable not offered in Manitoba', | |
91 | 'description': 'md5:c54a609f4c078d92b74ffb9bf1f496f4', | |
92 | 'duration': 121, | |
2a7dee8c YCH |
93 | }, |
94 | 'skip': 'Video gone', | |
3c06d371 | 95 | }, { |
85fab7e4 | 96 | 'url': 'https://ca.finance.yahoo.com/news/hackers-sony-more-trouble-well-154609075.html', |
3c06d371 | 97 | 'info_dict': { |
d9ed3621 YCH |
98 | 'id': '154609075', |
99 | }, | |
100 | 'playlist': [{ | |
101 | 'md5': 'f8e336c6b66f503282e5f719641d6565', | |
102 | 'info_dict': { | |
103 | 'id': 'e624c4bc-3389-34de-9dfc-025f74943409', | |
104 | 'ext': 'mp4', | |
105 | 'title': '\'The Interview\' TV Spot: War', | |
106 | 'description': 'The Interview', | |
107 | 'duration': 30, | |
108 | }, | |
109 | }, { | |
110 | 'md5': '958bcb90b4d6df71c56312137ee1cd5a', | |
111 | 'info_dict': { | |
112 | 'id': '1fc8ada0-718e-3abe-a450-bf31f246d1a9', | |
113 | 'ext': 'mp4', | |
114 | 'title': '\'The Interview\' TV Spot: Guys', | |
115 | 'description': 'The Interview', | |
116 | 'duration': 30, | |
117 | }, | |
118 | }], | |
3c06d371 S |
119 | }, { |
120 | 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html', | |
daeb0f04 | 121 | 'md5': '88e209b417f173d86186bef6e4d1f160', |
3c06d371 S |
122 | 'info_dict': { |
123 | 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521', | |
124 | 'ext': 'mp4', | |
125 | 'title': 'China Moses Is Crazy About the Blues', | |
126 | 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0', | |
127 | 'duration': 128, | |
128 | } | |
129 | }, { | |
130 | 'url': 'https://in.lifestyle.yahoo.com/video/connect-dots-dark-side-virgo-090247395.html', | |
131 | 'md5': 'd9a083ccf1379127bf25699d67e4791b', | |
132 | 'info_dict': { | |
133 | 'id': '52aeeaa3-b3d1-30d8-9ef8-5d0cf05efb7c', | |
134 | 'ext': 'mp4', | |
135 | 'title': 'Connect the Dots: Dark Side of Virgo', | |
136 | 'description': 'md5:1428185051cfd1949807ad4ff6d3686a', | |
137 | 'duration': 201, | |
2a7dee8c YCH |
138 | }, |
139 | 'skip': 'Domain name in.lifestyle.yahoo.com gone', | |
3aa9176f JMF |
140 | }, { |
141 | 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html', | |
4f549580 | 142 | 'md5': 'b17ac378b1134fa44370fb27db09a744', |
3aa9176f JMF |
143 | 'info_dict': { |
144 | 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1', | |
145 | 'ext': 'mp4', | |
146 | 'title': '\'True Story\' Trailer', | |
147 | 'description': 'True Story', | |
148 | 'duration': 150, | |
149 | }, | |
3c06d371 S |
150 | }, { |
151 | 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html', | |
152 | 'only_matching': True, | |
a28ccbab YCH |
153 | }, { |
154 | 'note': 'NBC Sports embeds', | |
155 | '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', | |
156 | 'info_dict': { | |
157 | 'id': '9CsDKds0kvHI', | |
158 | 'ext': 'flv', | |
159 | 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d', | |
160 | 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson', | |
4f549580 YCH |
161 | 'upload_date': '20150313', |
162 | 'uploader': 'NBCU-SPORTS', | |
163 | 'timestamp': 1426270238, | |
a28ccbab | 164 | } |
861e65eb YCH |
165 | }, { |
166 | 'url': 'https://tw.news.yahoo.com/-100120367.html', | |
167 | 'only_matching': True, | |
364ca058 S |
168 | }, { |
169 | # Query result is embedded in webpage, but explicit request to video API fails with geo restriction | |
170 | 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html', | |
4f549580 | 171 | 'md5': '1ddbf7c850777548438e5c4f147c7b8c', |
364ca058 S |
172 | 'info_dict': { |
173 | 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504', | |
174 | 'ext': 'mp4', | |
175 | 'title': 'Communitary - Community Episode 1: Ladders', | |
176 | 'description': 'md5:8fc39608213295748e1e289807838c97', | |
177 | 'duration': 1646, | |
178 | }, | |
ca227c86 JMF |
179 | }, { |
180 | # it uses an alias to get the video_id | |
181 | 'url': 'https://www.yahoo.com/movies/the-stars-of-daddys-home-have-very-different-212843197.html', | |
182 | 'info_dict': { | |
183 | 'id': '40eda9c8-8e5f-3552-8745-830f67d0c737', | |
184 | 'ext': 'mp4', | |
185 | 'title': 'Will Ferrell & Mark Wahlberg Are Pro-Spanking', | |
186 | 'description': 'While they play feuding fathers in \'Daddy\'s Home,\' star Will Ferrell & Mark Wahlberg share their true feelings on parenthood.', | |
187 | }, | |
188 | }, | |
2a7c3883 YCH |
189 | { |
190 | # config['models']['applet_model']['data']['sapi'] has no query | |
191 | 'url': 'https://www.yahoo.com/music/livenation/event/galactic-2016', | |
192 | 'md5': 'dac0c72d502bc5facda80c9e6d5c98db', | |
193 | 'info_dict': { | |
194 | 'id': 'a6015640-e9e5-3efb-bb60-05589a183919', | |
195 | 'ext': 'mp4', | |
196 | 'description': 'Galactic', | |
197 | 'title': 'Dolla Diva (feat. Maggie Koerner)', | |
198 | }, | |
199 | }, | |
9c15e9de | 200 | ] |
d6039175 PH |
201 | |
202 | def _real_extract(self, url): | |
203 | mobj = re.match(self._VALID_URL, url) | |
861e65eb | 204 | display_id = mobj.group('display_id') or self._match_id(url) |
3aa9176f | 205 | page_id = mobj.group('id') |
49807b4a | 206 | url = mobj.group('url') |
3c06d371 | 207 | host = mobj.group('host') |
2a7dee8c YCH |
208 | webpage, urlh = self._download_webpage_handle(url, display_id) |
209 | if 'err=404' in urlh.geturl(): | |
210 | raise ExtractorError('Video gone', expected=True) | |
3c06d371 S |
211 | |
212 | # Look for iframed media first | |
d9ed3621 YCH |
213 | entries = [] |
214 | iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage) | |
215 | for idx, iframe_url in enumerate(iframe_urls): | |
3c06d371 | 216 | iframepage = self._download_webpage( |
d9ed3621 YCH |
217 | host + iframe_url, display_id, |
218 | note='Downloading iframe webpage for video #%d' % idx) | |
3c06d371 S |
219 | items_json = self._search_regex( |
220 | r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None) | |
221 | if items_json: | |
222 | items = json.loads(items_json) | |
223 | video_id = items[0]['id'] | |
d9ed3621 YCH |
224 | entries.append(self._get_info(video_id, display_id, webpage)) |
225 | if entries: | |
226 | return self.playlist_result(entries, page_id) | |
227 | ||
a28ccbab | 228 | # Look for NBCSports iframes |
a2a4d5fa YCH |
229 | nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage) |
230 | if nbc_sports_url: | |
6a4e659c S |
231 | return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key()) |
232 | ||
233 | # Look for Brightcove New Studio embeds | |
234 | bc_url = BrightcoveNewIE._extract_url(webpage) | |
235 | if bc_url: | |
236 | return self.url_result(bc_url, BrightcoveNewIE.ie_key()) | |
d6039175 | 237 | |
364ca058 S |
238 | # Query result is often embedded in webpage as JSON. Sometimes explicit requests |
239 | # to video API results in a failure with geo restriction reason therefore using | |
240 | # embedded query result when present sounds reasonable. | |
241 | config_json = self._search_regex( | |
242 | r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:</script>|$)', | |
243 | webpage, 'videoplayer applet', default=None) | |
244 | if config_json: | |
245 | config = self._parse_json(config_json, display_id, fatal=False) | |
246 | if config: | |
247 | sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi') | |
2a7c3883 | 248 | if sapi and 'query' in sapi: |
364ca058 S |
249 | return self._extract_info(display_id, sapi, webpage) |
250 | ||
a4eb9578 PH |
251 | items_json = self._search_regex( |
252 | r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE, | |
253 | default=None) | |
254 | if items_json is None: | |
ca227c86 JMF |
255 | alias = self._search_regex( |
256 | r'"aliases":{"video":"(.*?)"', webpage, 'alias', default=None) | |
257 | if alias is not None: | |
258 | alias_info = self._download_json( | |
259 | 'https://www.yahoo.com/_td/api/resource/VideoService.videos;video_aliases=["%s"]' % alias, | |
260 | display_id, 'Downloading alias info') | |
261 | video_id = alias_info[0]['id'] | |
262 | else: | |
263 | CONTENT_ID_REGEXES = [ | |
264 | r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"', | |
265 | r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"', | |
266 | r'"first_videoid"\s*:\s*"([^"]+)"', | |
267 | r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id), | |
e87d98b0 S |
268 | r'<article[^>]data-uuid=["\']([^"\']+)', |
269 | r'yahoo://article/view\?.*\buuid=([^&"\']+)', | |
ca227c86 JMF |
270 | ] |
271 | video_id = self._search_regex( | |
272 | CONTENT_ID_REGEXES, webpage, 'content ID') | |
a4eb9578 PH |
273 | else: |
274 | items = json.loads(items_json) | |
275 | info = items['mediaItems']['query']['results']['mediaObj'][0] | |
276 | # The 'meta' field is not always in the video webpage, we request it | |
277 | # from another page | |
3c06d371 S |
278 | video_id = info['id'] |
279 | return self._get_info(video_id, display_id, webpage) | |
befd88b7 | 280 | |
364ca058 S |
281 | def _extract_info(self, display_id, query, webpage): |
282 | info = query['query']['results']['mediaObj'][0] | |
3c06d371 | 283 | meta = info.get('meta') |
364ca058 | 284 | video_id = info.get('id') |
3c06d371 S |
285 | |
286 | if not meta: | |
287 | msg = info['status'].get('msg') | |
288 | if msg: | |
289 | raise ExtractorError( | |
290 | '%s returned error: %s' % (self.IE_NAME, msg), expected=True) | |
291 | raise ExtractorError('Unable to extract media object meta') | |
9c15e9de JMF |
292 | |
293 | formats = [] | |
294 | for s in info['streams']: | |
295 | format_info = { | |
7217e148 PH |
296 | 'width': int_or_none(s.get('width')), |
297 | 'height': int_or_none(s.get('height')), | |
298 | 'tbr': int_or_none(s.get('bitrate')), | |
9c15e9de JMF |
299 | } |
300 | ||
301 | host = s['host'] | |
302 | path = s['path'] | |
303 | if host.startswith('rtmp'): | |
304 | format_info.update({ | |
305 | 'url': host, | |
306 | 'play_path': path, | |
307 | 'ext': 'flv', | |
308 | }) | |
309 | else: | |
364ca058 S |
310 | if s.get('format') == 'm3u8_playlist': |
311 | format_info['protocol'] = 'm3u8_native' | |
312 | format_info['ext'] = 'mp4' | |
9c15e9de JMF |
313 | format_url = compat_urlparse.urljoin(host, path) |
314 | format_info['url'] = format_url | |
9c15e9de | 315 | formats.append(format_info) |
7217e148 PH |
316 | |
317 | self._sort_formats(formats) | |
9c15e9de | 318 | |
ecee5724 YCH |
319 | closed_captions = self._html_search_regex( |
320 | r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions', | |
321 | default='[]') | |
322 | ||
323 | cc_json = self._parse_json(closed_captions, video_id, fatal=False) | |
324 | subtitles = {} | |
325 | if cc_json: | |
326 | for closed_caption in cc_json: | |
327 | lang = closed_caption['lang'] | |
328 | if lang not in subtitles: | |
329 | subtitles[lang] = [] | |
330 | subtitles[lang].append({ | |
331 | 'url': closed_caption['url'], | |
332 | 'ext': mimetype2ext(closed_caption['content_type']), | |
333 | }) | |
334 | ||
35907e23 | 335 | return { |
9c15e9de | 336 | 'id': video_id, |
3c06d371 | 337 | 'display_id': display_id, |
3aa9176f | 338 | 'title': unescapeHTML(meta['title']), |
9c15e9de JMF |
339 | 'formats': formats, |
340 | 'description': clean_html(meta['description']), | |
dede691a | 341 | 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage), |
762958d5 | 342 | 'duration': int_or_none(meta.get('duration')), |
ecee5724 | 343 | 'subtitles': subtitles, |
9c15e9de | 344 | } |
d6039175 | 345 | |
364ca058 S |
346 | def _get_info(self, video_id, display_id, webpage): |
347 | region = self._search_regex( | |
348 | r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"', | |
349 | webpage, 'region', fatal=False, default='US') | |
15707c7e | 350 | data = compat_urllib_parse_urlencode({ |
364ca058 | 351 | 'protocol': 'http', |
55290788 | 352 | 'region': region.upper(), |
364ca058 S |
353 | }) |
354 | query_url = ( | |
355 | 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/' | |
356 | '{id}?{data}'.format(id=video_id, data=data)) | |
357 | query_result = self._download_json( | |
358 | query_url, display_id, 'Downloading video info') | |
359 | return self._extract_info(display_id, query_result, webpage) | |
360 | ||
934858ad PH |
361 | |
362 | class YahooSearchIE(SearchInfoExtractor): | |
2317e6b2 | 363 | IE_DESC = 'Yahoo screen search' |
934858ad | 364 | _MAX_RESULTS = 1000 |
2317e6b2 | 365 | IE_NAME = 'screen.yahoo:search' |
934858ad PH |
366 | _SEARCH_KEY = 'yvsearch' |
367 | ||
368 | def _get_n_results(self, query, n): | |
369 | """Get a specified number of results for a query""" | |
88f1c6de JMF |
370 | entries = [] |
371 | for pagenum in itertools.count(0): | |
2317e6b2 | 372 | 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) |
88f1c6de | 373 | info = self._download_json(result_url, query, |
9e1a5b84 | 374 | note='Downloading results page ' + str(pagenum + 1)) |
2317e6b2 AS |
375 | m = info['m'] |
376 | results = info['results'] | |
934858ad PH |
377 | |
378 | for (i, r) in enumerate(results): | |
88f1c6de | 379 | if (pagenum * 30) + i >= n: |
934858ad PH |
380 | break |
381 | mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r) | |
382 | e = self.url_result('http://' + mobj.group('url'), 'Yahoo') | |
88f1c6de JMF |
383 | entries.append(e) |
384 | if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)): | |
934858ad PH |
385 | break |
386 | ||
88f1c6de JMF |
387 | return { |
388 | '_type': 'playlist', | |
389 | 'id': query, | |
390 | 'entries': entries, | |
391 | } |