6c6bd76e879edbef50e8839549efc7326edd485b
[youtube-dl] / youtube_dl / extractor / yahoo.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import hashlib
5 import itertools
6 import re
7
8 from .common import InfoExtractor, SearchInfoExtractor
9 from ..compat import (
10     compat_str,
11     compat_urllib_parse,
12 )
13 from ..utils import (
14     clean_html,
15     int_or_none,
16     mimetype2ext,
17     parse_iso8601,
18     smuggle_url,
19     try_get,
20     url_or_none,
21 )
22
23 from .brightcove import BrightcoveNewIE
24
25
26 class YahooIE(InfoExtractor):
27     IE_DESC = 'Yahoo screen and movies'
28     _VALID_URL = r'(?P<url>https?://(?:(?P<country>[a-zA-Z]{2}(?:-[a-zA-Z]{2})?|malaysia)\.)?(?:[\da-zA-Z_-]+\.)?yahoo\.com/(?:[^/]+/)*(?P<id>[^?&#]*-[0-9]+)\.html)'
29     _TESTS = [{
30         'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
31         'info_dict': {
32             'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
33             'ext': 'mp4',
34             'title': 'Julian Smith & Travis Legg Watch Julian Smith',
35             'description': 'Julian and Travis watch Julian Smith',
36             'duration': 6863,
37             'timestamp': 1369812016,
38             'upload_date': '20130529',
39         },
40     }, {
41         'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
42         'md5': '7993e572fac98e044588d0b5260f4352',
43         'info_dict': {
44             'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
45             'ext': 'mp4',
46             'title': "Yahoo Saves 'Community'",
47             'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
48             'duration': 170,
49             'timestamp': 1406838636,
50             'upload_date': '20140731',
51         },
52     }, {
53         'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
54         'md5': '71298482f7c64cbb7fa064e4553ff1c1',
55         'info_dict': {
56             'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
57             'ext': 'webm',
58             'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
59             'description': 'md5:f66c890e1490f4910a9953c941dee944',
60             'duration': 97,
61             'timestamp': 1414489862,
62             'upload_date': '20141028',
63         }
64     }, {
65         'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
66         'md5': '88e209b417f173d86186bef6e4d1f160',
67         'info_dict': {
68             'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
69             'ext': 'mp4',
70             'title': 'China Moses Is Crazy About the Blues',
71             'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
72             'duration': 128,
73             'timestamp': 1385722202,
74             'upload_date': '20131129',
75         }
76     }, {
77         'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
78         'md5': '2a9752f74cb898af5d1083ea9f661b58',
79         'info_dict': {
80             'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
81             'ext': 'mp4',
82             'title': '\'True Story\' Trailer',
83             'description': 'True Story',
84             'duration': 150,
85             'timestamp': 1418919206,
86             'upload_date': '20141218',
87         },
88     }, {
89         'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
90         'only_matching': True,
91     }, {
92         'note': 'NBC Sports embeds',
93         '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',
94         'info_dict': {
95             'id': '9CsDKds0kvHI',
96             'ext': 'flv',
97             'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
98             'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
99             'upload_date': '20150313',
100             'uploader': 'NBCU-SPORTS',
101             'timestamp': 1426270238,
102         },
103     }, {
104         'url': 'https://tw.news.yahoo.com/-100120367.html',
105         'only_matching': True,
106     }, {
107         # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
108         'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
109         'md5': '4fbafb9c9b6f07aa8f870629f6671b35',
110         'info_dict': {
111             'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
112             'ext': 'mp4',
113             'title': 'Communitary - Community Episode 1: Ladders',
114             'description': 'md5:8fc39608213295748e1e289807838c97',
115             'duration': 1646,
116             'timestamp': 1440436550,
117             'upload_date': '20150824',
118             'series': 'Communitary',
119             'season_number': 6,
120             'episode_number': 1,
121         },
122     }, {
123         # ytwnews://cavideo/
124         'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html',
125         'info_dict': {
126             'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff',
127             'ext': 'mp4',
128             'title': '單車天使 - 中文版預',
129             'description': '中文版預',
130             'timestamp': 1476696196,
131             'upload_date': '20161017',
132         },
133         'params': {
134             'skip_download': True,
135         },
136     }, {
137         # Contains both a Yahoo hosted video and multiple Youtube embeds
138         'url': 'https://www.yahoo.com/entertainment/gwen-stefani-reveals-the-pop-hit-she-passed-on-assigns-it-to-her-voice-contestant-instead-033045672.html',
139         'info_dict': {
140             'id': '46c5d95a-528f-3d03-b732-732fcadd51de',
141             'title': 'Gwen Stefani reveals the pop hit she passed on, assigns it to her \'Voice\' contestant instead',
142             'description': 'Gwen decided not to record this hit herself, but she decided it was the perfect fit for Kyndall Inskeep.',
143         },
144         'playlist': [{
145             'info_dict': {
146                 'id': '966d4262-4fd1-3aaa-b45b-049ca6e38ba6',
147                 'ext': 'mp4',
148                 'title': 'Gwen Stefani reveals she turned down one of Sia\'s best songs',
149                 'description': 'On "The Voice" Tuesday, Gwen Stefani told Taylor Swift which Sia hit was almost hers.',
150                 'timestamp': 1572406500,
151                 'upload_date': '20191030',
152             },
153         }, {
154             'info_dict': {
155                 'id': '352CFDOQrKg',
156                 'ext': 'mp4',
157                 'title': 'Kyndal Inskeep "Performs the Hell Out of" Sia\'s "Elastic Heart" - The Voice Knockouts 2019',
158                 'description': 'md5:35b61e94c2ae214bc965ff4245f80d11',
159                 'uploader': 'The Voice',
160                 'uploader_id': 'NBCTheVoice',
161                 'upload_date': '20191029',
162             },
163         }],
164         'params': {
165             'playlistend': 2,
166         },
167         'expected_warnings': ['HTTP Error 404'],
168     }, {
169         'url': 'https://malaysia.news.yahoo.com/video/bystanders-help-ontario-policeman-bust-190932818.html',
170         'only_matching': True,
171     }, {
172         'url': 'https://es-us.noticias.yahoo.com/es-la-puerta-irrompible-que-110539379.html',
173         'only_matching': True,
174     }]
175
176     def _real_extract(self, url):
177         url, country, display_id = re.match(self._VALID_URL, url).groups()
178         if not country:
179             country = 'us'
180         else:
181             country = country.split('-')[0]
182         api_base = 'https://%s.yahoo.com/_td/api/resource/' % country
183
184         for i, uuid in enumerate(['url=' + url, 'ymedia-alias=' + display_id]):
185             content = self._download_json(
186                 api_base + 'content;getDetailView=true;uuids=["%s"]' % uuid,
187                 display_id, 'Downloading content JSON metadata', fatal=i == 1)
188             if content:
189                 item = content['items'][0]
190                 break
191
192         if item.get('type') != 'video':
193             entries = []
194
195             cover = item.get('cover') or {}
196             if cover.get('type') == 'yvideo':
197                 cover_url = cover.get('url')
198                 if cover_url:
199                     entries.append(self.url_result(
200                         cover_url, 'Yahoo', cover.get('uuid')))
201
202             for e in item.get('body', []):
203                 if e.get('type') == 'videoIframe':
204                     iframe_url = e.get('url')
205                     if not iframe_url:
206                         continue
207                     entries.append(self.url_result(iframe_url))
208
209             return self.playlist_result(
210                 entries, item.get('uuid'),
211                 item.get('title'), item.get('summary'))
212
213         video_id = item['uuid']
214         video = self._download_json(
215             api_base + 'VideoService.videos;view=full;video_ids=["%s"]' % video_id,
216             video_id, 'Downloading video JSON metadata')[0]
217         title = video['title']
218
219         if country == 'malaysia':
220             country = 'my'
221
222         is_live = video.get('live_state') == 'live'
223         fmts = ('m3u8',) if is_live else ('webm', 'mp4')
224
225         urls = []
226         formats = []
227         subtitles = {}
228         for fmt in fmts:
229             media_obj = self._download_json(
230                 'https://video-api.yql.yahoo.com/v1/video/sapi/streams/' + video_id,
231                 video_id, 'Downloading %s JSON metadata' % fmt,
232                 headers=self.geo_verification_headers(), query={
233                     'format': fmt,
234                     'region': country.upper(),
235                 })['query']['results']['mediaObj'][0]
236             msg = media_obj.get('status', {}).get('msg')
237
238             for s in media_obj.get('streams', []):
239                 host = s.get('host')
240                 path = s.get('path')
241                 if not host or not path:
242                     continue
243                 s_url = host + path
244                 if s.get('format') == 'm3u8':
245                     formats.extend(self._extract_m3u8_formats(
246                         s_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
247                     continue
248                 tbr = int_or_none(s.get('bitrate'))
249                 formats.append({
250                     'url': s_url,
251                     'format_id': fmt + ('-%d' % tbr if tbr else ''),
252                     'width': int_or_none(s.get('width')),
253                     'height': int_or_none(s.get('height')),
254                     'tbr': tbr,
255                     'fps': int_or_none(s.get('framerate')),
256                 })
257
258             for cc in media_obj.get('closedcaptions', []):
259                 cc_url = cc.get('url')
260                 if not cc_url or cc_url in urls:
261                     continue
262                 urls.append(cc_url)
263                 subtitles.setdefault(cc.get('lang') or 'en-US', []).append({
264                     'url': cc_url,
265                     'ext': mimetype2ext(cc.get('content_type')),
266                 })
267
268         streaming_url = video.get('streaming_url')
269         if streaming_url and not is_live:
270             formats.extend(self._extract_m3u8_formats(
271                 streaming_url, video_id, 'mp4',
272                 'm3u8_native', m3u8_id='hls', fatal=False))
273
274         if not formats and msg == 'geo restricted':
275             self.raise_geo_restricted()
276
277         self._sort_formats(formats)
278
279         thumbnails = []
280         for thumb in video.get('thumbnails', []):
281             thumb_url = thumb.get('url')
282             if not thumb_url:
283                 continue
284             thumbnails.append({
285                 'id': thumb.get('tag'),
286                 'url': thumb.get('url'),
287                 'width': int_or_none(thumb.get('width')),
288                 'height': int_or_none(thumb.get('height')),
289             })
290
291         series_info = video.get('series_info') or {}
292
293         return {
294             'id': video_id,
295             'title': self._live_title(title) if is_live else title,
296             'formats': formats,
297             'display_id': display_id,
298             'thumbnails': thumbnails,
299             'description': clean_html(video.get('description')),
300             'timestamp': parse_iso8601(video.get('publish_time')),
301             'subtitles': subtitles,
302             'duration': int_or_none(video.get('duration')),
303             'view_count': int_or_none(video.get('view_count')),
304             'is_live': is_live,
305             'series': video.get('show_name'),
306             'season_number': int_or_none(series_info.get('season_number')),
307             'episode_number': int_or_none(series_info.get('episode_number')),
308         }
309
310
311 class YahooSearchIE(SearchInfoExtractor):
312     IE_DESC = 'Yahoo screen search'
313     _MAX_RESULTS = 1000
314     IE_NAME = 'screen.yahoo:search'
315     _SEARCH_KEY = 'yvsearch'
316
317     def _get_n_results(self, query, n):
318         """Get a specified number of results for a query"""
319         entries = []
320         for pagenum in itertools.count(0):
321             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)
322             info = self._download_json(result_url, query,
323                                        note='Downloading results page ' + str(pagenum + 1))
324             m = info['m']
325             results = info['results']
326
327             for (i, r) in enumerate(results):
328                 if (pagenum * 30) + i >= n:
329                     break
330                 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
331                 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
332                 entries.append(e)
333             if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
334                 break
335
336         return {
337             '_type': 'playlist',
338             'id': query,
339             'entries': entries,
340         }
341
342
343 class YahooGyaOPlayerIE(InfoExtractor):
344     IE_NAME = 'yahoo:gyao:player'
345     _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:player|episode/[^/]+)|streaming\.yahoo\.co\.jp/c/y)/(?P<id>\d+/v\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
346     _TESTS = [{
347         'url': 'https://gyao.yahoo.co.jp/player/00998/v00818/v0000000000000008564/',
348         'info_dict': {
349             'id': '5993125228001',
350             'ext': 'mp4',
351             'title': 'フューリー 【字幕版】',
352             'description': 'md5:21e691c798a15330eda4db17a8fe45a5',
353             'uploader_id': '4235717419001',
354             'upload_date': '20190124',
355             'timestamp': 1548294365,
356         },
357         'params': {
358             # m3u8 download
359             'skip_download': True,
360         },
361     }, {
362         'url': 'https://streaming.yahoo.co.jp/c/y/01034/v00133/v0000000000000000706/',
363         'only_matching': True,
364     }, {
365         'url': 'https://gyao.yahoo.co.jp/episode/%E3%81%8D%E3%81%AE%E3%81%86%E4%BD%95%E9%A3%9F%E3%81%B9%E3%81%9F%EF%BC%9F%20%E7%AC%AC2%E8%A9%B1%202019%2F4%2F12%E6%94%BE%E9%80%81%E5%88%86/5cb02352-b725-409e-9f8d-88f947a9f682',
366         'only_matching': True,
367     }]
368
369     def _real_extract(self, url):
370         video_id = self._match_id(url).replace('/', ':')
371         video = self._download_json(
372             'https://gyao.yahoo.co.jp/dam/v1/videos/' + video_id,
373             video_id, query={
374                 'fields': 'longDescription,title,videoId',
375             }, headers={
376                 'X-User-Agent': 'Unknown Pc GYAO!/2.0.0 Web',
377             })
378         return {
379             '_type': 'url_transparent',
380             'id': video_id,
381             'title': video['title'],
382             'url': smuggle_url(
383                 'http://players.brightcove.net/4235717419001/default_default/index.html?videoId=' + video['videoId'],
384                 {'geo_countries': ['JP']}),
385             'description': video.get('longDescription'),
386             'ie_key': BrightcoveNewIE.ie_key(),
387         }
388
389
390 class YahooGyaOIE(InfoExtractor):
391     IE_NAME = 'yahoo:gyao'
392     _VALID_URL = r'https?://(?:gyao\.yahoo\.co\.jp/(?:p|title/[^/]+)|streaming\.yahoo\.co\.jp/p/y)/(?P<id>\d+/v\d+|[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
393     _TESTS = [{
394         'url': 'https://gyao.yahoo.co.jp/p/00449/v03102/',
395         'info_dict': {
396             'id': '00449:v03102',
397         },
398         'playlist_count': 2,
399     }, {
400         'url': 'https://streaming.yahoo.co.jp/p/y/01034/v00133/',
401         'only_matching': True,
402     }, {
403         'url': 'https://gyao.yahoo.co.jp/title/%E3%81%97%E3%82%83%E3%81%B9%E3%81%8F%E3%82%8A007/5b025a49-b2e5-4dc7-945c-09c6634afacf',
404         'only_matching': True,
405     }]
406
407     def _real_extract(self, url):
408         program_id = self._match_id(url).replace('/', ':')
409         videos = self._download_json(
410             'https://gyao.yahoo.co.jp/api/programs/%s/videos' % program_id, program_id)['videos']
411         entries = []
412         for video in videos:
413             video_id = video.get('id')
414             if not video_id:
415                 continue
416             entries.append(self.url_result(
417                 'https://gyao.yahoo.co.jp/player/%s/' % video_id.replace(':', '/'),
418                 YahooGyaOPlayerIE.ie_key(), video_id))
419         return self.playlist_result(entries, program_id)
420
421
422 class YahooJapanNewsIE(InfoExtractor):
423     IE_NAME = 'yahoo:japannews'
424     IE_DESC = 'Yahoo! Japan News'
425     _VALID_URL = r'https?://(?P<host>(?:news|headlines)\.yahoo\.co\.jp)[^\d]*(?P<id>\d[\d-]*\d)?'
426     _GEO_COUNTRIES = ['JP']
427     _TESTS = [{
428         'url': 'https://headlines.yahoo.co.jp/videonews/ann?a=20190716-00000071-ann-int',
429         'info_dict': {
430             'id': '1736242',
431             'ext': 'mp4',
432             'title': 'ムン大統領が対日批判を強化“現金化”効果は?(テレビ朝日系(ANN)) - Yahoo!ニュース',
433             'description': '韓国の元徴用工らを巡る裁判の原告が弁護士が差し押さえた三菱重工業の資産を売却して - Yahoo!ニュース(テレビ朝日系(ANN))',
434             'thumbnail': r're:^https?://.*\.[a-zA-Z\d]{3,4}$',
435         },
436         'params': {
437             'skip_download': True,
438         },
439     }, {
440         # geo restricted
441         'url': 'https://headlines.yahoo.co.jp/hl?a=20190721-00000001-oxv-l04',
442         'only_matching': True,
443     }, {
444         'url': 'https://headlines.yahoo.co.jp/videonews/',
445         'only_matching': True,
446     }, {
447         'url': 'https://news.yahoo.co.jp',
448         'only_matching': True,
449     }, {
450         'url': 'https://news.yahoo.co.jp/byline/hashimotojunji/20190628-00131977/',
451         'only_matching': True,
452     }, {
453         'url': 'https://news.yahoo.co.jp/feature/1356',
454         'only_matching': True
455     }]
456
457     def _extract_formats(self, json_data, content_id):
458         formats = []
459
460         video_data = try_get(
461             json_data,
462             lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
463             list)
464         for vid in video_data or []:
465             delivery = vid.get('delivery')
466             url = url_or_none(vid.get('Url'))
467             if not delivery or not url:
468                 continue
469             elif delivery == 'hls':
470                 formats.extend(
471                     self._extract_m3u8_formats(
472                         url, content_id, 'mp4', 'm3u8_native',
473                         m3u8_id='hls', fatal=False))
474             else:
475                 formats.append({
476                     'url': url,
477                     'format_id': 'http-%s' % compat_str(vid.get('bitrate', '')),
478                     'height': int_or_none(vid.get('height')),
479                     'width': int_or_none(vid.get('width')),
480                     'tbr': int_or_none(vid.get('bitrate')),
481                 })
482         self._remove_duplicate_formats(formats)
483         self._sort_formats(formats)
484
485         return formats
486
487     def _real_extract(self, url):
488         mobj = re.match(self._VALID_URL, url)
489         host = mobj.group('host')
490         display_id = mobj.group('id') or host
491
492         webpage = self._download_webpage(url, display_id)
493
494         title = self._html_search_meta(
495             ['og:title', 'twitter:title'], webpage, 'title', default=None
496         ) or self._html_search_regex('<title>([^<]+)</title>', webpage, 'title')
497
498         if display_id == host:
499             # Headline page (w/ multiple BC playlists) ('news.yahoo.co.jp', 'headlines.yahoo.co.jp/videonews/', ...)
500             stream_plists = re.findall(r'plist=(\d+)', webpage) or re.findall(r'plist["\']:\s*["\']([^"\']+)', webpage)
501             entries = [
502                 self.url_result(
503                     smuggle_url(
504                         'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=%s' % plist_id,
505                         {'geo_countries': ['JP']}),
506                     ie='BrightcoveNew', video_id=plist_id)
507                 for plist_id in stream_plists]
508             return self.playlist_result(entries, playlist_title=title)
509
510         # Article page
511         description = self._html_search_meta(
512             ['og:description', 'description', 'twitter:description'],
513             webpage, 'description', default=None)
514         thumbnail = self._og_search_thumbnail(
515             webpage, default=None) or self._html_search_meta(
516             'twitter:image', webpage, 'thumbnail', default=None)
517         space_id = self._search_regex([
518             r'<script[^>]+class=["\']yvpub-player["\'][^>]+spaceid=([^&"\']+)',
519             r'YAHOO\.JP\.srch\.\w+link\.onLoad[^;]+spaceID["\' ]*:["\' ]+([^"\']+)',
520             r'<!--\s+SpaceID=(\d+)'
521         ], webpage, 'spaceid')
522
523         content_id = self._search_regex(
524             r'<script[^>]+class=["\']yvpub-player["\'][^>]+contentid=(?P<contentid>[^&"\']+)',
525             webpage, 'contentid', group='contentid')
526
527         json_data = self._download_json(
528             'https://feapi-yvpub.yahooapis.jp/v1/content/%s' % content_id,
529             content_id,
530             query={
531                 'appid': 'dj0zaiZpPVZMTVFJR0FwZWpiMyZzPWNvbnN1bWVyc2VjcmV0Jng9YjU-',
532                 'output': 'json',
533                 'space_id': space_id,
534                 'domain': host,
535                 'ak': hashlib.md5('_'.join((space_id, host)).encode()).hexdigest(),
536                 'device_type': '1100',
537             })
538         formats = self._extract_formats(json_data, content_id)
539
540         return {
541             'id': content_id,
542             'title': title,
543             'description': description,
544             'thumbnail': thumbnail,
545             'formats': formats,
546         }