]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yahoo.py
debian/control: Move phantomjs from Recommends to Suggests.
[youtubedl] / 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]+(?:-[a-z]+)?)\.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 'url': 'https://www.yahoo.com/entertainment/v/longtime-cbs-news-60-minutes-032036500-cbs.html',
176 'only_matching': True,
177 }]
178
179 def _real_extract(self, url):
180 url, country, display_id = re.match(self._VALID_URL, url).groups()
181 if not country:
182 country = 'us'
183 else:
184 country = country.split('-')[0]
185 api_base = 'https://%s.yahoo.com/_td/api/resource/' % country
186
187 for i, uuid in enumerate(['url=' + url, 'ymedia-alias=' + display_id]):
188 content = self._download_json(
189 api_base + 'content;getDetailView=true;uuids=["%s"]' % uuid,
190 display_id, 'Downloading content JSON metadata', fatal=i == 1)
191 if content:
192 item = content['items'][0]
193 break
194
195 if item.get('type') != 'video':
196 entries = []
197
198 cover = item.get('cover') or {}
199 if cover.get('type') == 'yvideo':
200 cover_url = cover.get('url')
201 if cover_url:
202 entries.append(self.url_result(
203 cover_url, 'Yahoo', cover.get('uuid')))
204
205 for e in item.get('body', []):
206 if e.get('type') == 'videoIframe':
207 iframe_url = e.get('url')
208 if not iframe_url:
209 continue
210 entries.append(self.url_result(iframe_url))
211
212 return self.playlist_result(
213 entries, item.get('uuid'),
214 item.get('title'), item.get('summary'))
215
216 video_id = item['uuid']
217 video = self._download_json(
218 api_base + 'VideoService.videos;view=full;video_ids=["%s"]' % video_id,
219 video_id, 'Downloading video JSON metadata')[0]
220 title = video['title']
221
222 if country == 'malaysia':
223 country = 'my'
224
225 is_live = video.get('live_state') == 'live'
226 fmts = ('m3u8',) if is_live else ('webm', 'mp4')
227
228 urls = []
229 formats = []
230 subtitles = {}
231 for fmt in fmts:
232 media_obj = self._download_json(
233 'https://video-api.yql.yahoo.com/v1/video/sapi/streams/' + video_id,
234 video_id, 'Downloading %s JSON metadata' % fmt,
235 headers=self.geo_verification_headers(), query={
236 'format': fmt,
237 'region': country.upper(),
238 })['query']['results']['mediaObj'][0]
239 msg = media_obj.get('status', {}).get('msg')
240
241 for s in media_obj.get('streams', []):
242 host = s.get('host')
243 path = s.get('path')
244 if not host or not path:
245 continue
246 s_url = host + path
247 if s.get('format') == 'm3u8':
248 formats.extend(self._extract_m3u8_formats(
249 s_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
250 continue
251 tbr = int_or_none(s.get('bitrate'))
252 formats.append({
253 'url': s_url,
254 'format_id': fmt + ('-%d' % tbr if tbr else ''),
255 'width': int_or_none(s.get('width')),
256 'height': int_or_none(s.get('height')),
257 'tbr': tbr,
258 'fps': int_or_none(s.get('framerate')),
259 })
260
261 for cc in media_obj.get('closedcaptions', []):
262 cc_url = cc.get('url')
263 if not cc_url or cc_url in urls:
264 continue
265 urls.append(cc_url)
266 subtitles.setdefault(cc.get('lang') or 'en-US', []).append({
267 'url': cc_url,
268 'ext': mimetype2ext(cc.get('content_type')),
269 })
270
271 streaming_url = video.get('streaming_url')
272 if streaming_url and not is_live:
273 formats.extend(self._extract_m3u8_formats(
274 streaming_url, video_id, 'mp4',
275 'm3u8_native', m3u8_id='hls', fatal=False))
276
277 if not formats and msg == 'geo restricted':
278 self.raise_geo_restricted()
279
280 self._sort_formats(formats)
281
282 thumbnails = []
283 for thumb in video.get('thumbnails', []):
284 thumb_url = thumb.get('url')
285 if not thumb_url:
286 continue
287 thumbnails.append({
288 'id': thumb.get('tag'),
289 'url': thumb.get('url'),
290 'width': int_or_none(thumb.get('width')),
291 'height': int_or_none(thumb.get('height')),
292 })
293
294 series_info = video.get('series_info') or {}
295
296 return {
297 'id': video_id,
298 'title': self._live_title(title) if is_live else title,
299 'formats': formats,
300 'display_id': display_id,
301 'thumbnails': thumbnails,
302 'description': clean_html(video.get('description')),
303 'timestamp': parse_iso8601(video.get('publish_time')),
304 'subtitles': subtitles,
305 'duration': int_or_none(video.get('duration')),
306 'view_count': int_or_none(video.get('view_count')),
307 'is_live': is_live,
308 'series': video.get('show_name'),
309 'season_number': int_or_none(series_info.get('season_number')),
310 'episode_number': int_or_none(series_info.get('episode_number')),
311 }
312
313
314 class YahooSearchIE(SearchInfoExtractor):
315 IE_DESC = 'Yahoo screen search'
316 _MAX_RESULTS = 1000
317 IE_NAME = 'screen.yahoo:search'
318 _SEARCH_KEY = 'yvsearch'
319
320 def _get_n_results(self, query, n):
321 """Get a specified number of results for a query"""
322 entries = []
323 for pagenum in itertools.count(0):
324 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)
325 info = self._download_json(result_url, query,
326 note='Downloading results page ' + str(pagenum + 1))
327 m = info['m']
328 results = info['results']
329
330 for (i, r) in enumerate(results):
331 if (pagenum * 30) + i >= n:
332 break
333 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
334 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
335 entries.append(e)
336 if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
337 break
338
339 return {
340 '_type': 'playlist',
341 'id': query,
342 'entries': entries,
343 }
344
345
346 class YahooGyaOPlayerIE(InfoExtractor):
347 IE_NAME = 'yahoo:gyao:player'
348 _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})'
349 _TESTS = [{
350 'url': 'https://gyao.yahoo.co.jp/player/00998/v00818/v0000000000000008564/',
351 'info_dict': {
352 'id': '5993125228001',
353 'ext': 'mp4',
354 'title': 'フューリー 【字幕版】',
355 'description': 'md5:21e691c798a15330eda4db17a8fe45a5',
356 'uploader_id': '4235717419001',
357 'upload_date': '20190124',
358 'timestamp': 1548294365,
359 },
360 'params': {
361 # m3u8 download
362 'skip_download': True,
363 },
364 }, {
365 'url': 'https://streaming.yahoo.co.jp/c/y/01034/v00133/v0000000000000000706/',
366 'only_matching': True,
367 }, {
368 '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',
369 'only_matching': True,
370 }]
371
372 def _real_extract(self, url):
373 video_id = self._match_id(url).replace('/', ':')
374 video = self._download_json(
375 'https://gyao.yahoo.co.jp/dam/v1/videos/' + video_id,
376 video_id, query={
377 'fields': 'longDescription,title,videoId',
378 }, headers={
379 'X-User-Agent': 'Unknown Pc GYAO!/2.0.0 Web',
380 })
381 return {
382 '_type': 'url_transparent',
383 'id': video_id,
384 'title': video['title'],
385 'url': smuggle_url(
386 'http://players.brightcove.net/4235717419001/SyG5P0gjb_default/index.html?videoId=' + video['videoId'],
387 {'geo_countries': ['JP']}),
388 'description': video.get('longDescription'),
389 'ie_key': BrightcoveNewIE.ie_key(),
390 }
391
392
393 class YahooGyaOIE(InfoExtractor):
394 IE_NAME = 'yahoo:gyao'
395 _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})'
396 _TESTS = [{
397 'url': 'https://gyao.yahoo.co.jp/p/00449/v03102/',
398 'info_dict': {
399 'id': '00449:v03102',
400 },
401 'playlist_count': 2,
402 }, {
403 'url': 'https://streaming.yahoo.co.jp/p/y/01034/v00133/',
404 'only_matching': True,
405 }, {
406 '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',
407 'only_matching': True,
408 }]
409
410 def _real_extract(self, url):
411 program_id = self._match_id(url).replace('/', ':')
412 videos = self._download_json(
413 'https://gyao.yahoo.co.jp/api/programs/%s/videos' % program_id, program_id)['videos']
414 entries = []
415 for video in videos:
416 video_id = video.get('id')
417 if not video_id:
418 continue
419 entries.append(self.url_result(
420 'https://gyao.yahoo.co.jp/player/%s/' % video_id.replace(':', '/'),
421 YahooGyaOPlayerIE.ie_key(), video_id))
422 return self.playlist_result(entries, program_id)
423
424
425 class YahooJapanNewsIE(InfoExtractor):
426 IE_NAME = 'yahoo:japannews'
427 IE_DESC = 'Yahoo! Japan News'
428 _VALID_URL = r'https?://(?P<host>(?:news|headlines)\.yahoo\.co\.jp)[^\d]*(?P<id>\d[\d-]*\d)?'
429 _GEO_COUNTRIES = ['JP']
430 _TESTS = [{
431 'url': 'https://headlines.yahoo.co.jp/videonews/ann?a=20190716-00000071-ann-int',
432 'info_dict': {
433 'id': '1736242',
434 'ext': 'mp4',
435 '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',
436 '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))',
437 'thumbnail': r're:^https?://.*\.[a-zA-Z\d]{3,4}$',
438 },
439 'params': {
440 'skip_download': True,
441 },
442 }, {
443 # geo restricted
444 'url': 'https://headlines.yahoo.co.jp/hl?a=20190721-00000001-oxv-l04',
445 'only_matching': True,
446 }, {
447 'url': 'https://headlines.yahoo.co.jp/videonews/',
448 'only_matching': True,
449 }, {
450 'url': 'https://news.yahoo.co.jp',
451 'only_matching': True,
452 }, {
453 'url': 'https://news.yahoo.co.jp/byline/hashimotojunji/20190628-00131977/',
454 'only_matching': True,
455 }, {
456 'url': 'https://news.yahoo.co.jp/feature/1356',
457 'only_matching': True
458 }]
459
460 def _extract_formats(self, json_data, content_id):
461 formats = []
462
463 video_data = try_get(
464 json_data,
465 lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
466 list)
467 for vid in video_data or []:
468 delivery = vid.get('delivery')
469 url = url_or_none(vid.get('Url'))
470 if not delivery or not url:
471 continue
472 elif delivery == 'hls':
473 formats.extend(
474 self._extract_m3u8_formats(
475 url, content_id, 'mp4', 'm3u8_native',
476 m3u8_id='hls', fatal=False))
477 else:
478 formats.append({
479 'url': url,
480 'format_id': 'http-%s' % compat_str(vid.get('bitrate', '')),
481 'height': int_or_none(vid.get('height')),
482 'width': int_or_none(vid.get('width')),
483 'tbr': int_or_none(vid.get('bitrate')),
484 })
485 self._remove_duplicate_formats(formats)
486 self._sort_formats(formats)
487
488 return formats
489
490 def _real_extract(self, url):
491 mobj = re.match(self._VALID_URL, url)
492 host = mobj.group('host')
493 display_id = mobj.group('id') or host
494
495 webpage = self._download_webpage(url, display_id)
496
497 title = self._html_search_meta(
498 ['og:title', 'twitter:title'], webpage, 'title', default=None
499 ) or self._html_search_regex('<title>([^<]+)</title>', webpage, 'title')
500
501 if display_id == host:
502 # Headline page (w/ multiple BC playlists) ('news.yahoo.co.jp', 'headlines.yahoo.co.jp/videonews/', ...)
503 stream_plists = re.findall(r'plist=(\d+)', webpage) or re.findall(r'plist["\']:\s*["\']([^"\']+)', webpage)
504 entries = [
505 self.url_result(
506 smuggle_url(
507 'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=%s' % plist_id,
508 {'geo_countries': ['JP']}),
509 ie='BrightcoveNew', video_id=plist_id)
510 for plist_id in stream_plists]
511 return self.playlist_result(entries, playlist_title=title)
512
513 # Article page
514 description = self._html_search_meta(
515 ['og:description', 'description', 'twitter:description'],
516 webpage, 'description', default=None)
517 thumbnail = self._og_search_thumbnail(
518 webpage, default=None) or self._html_search_meta(
519 'twitter:image', webpage, 'thumbnail', default=None)
520 space_id = self._search_regex([
521 r'<script[^>]+class=["\']yvpub-player["\'][^>]+spaceid=([^&"\']+)',
522 r'YAHOO\.JP\.srch\.\w+link\.onLoad[^;]+spaceID["\' ]*:["\' ]+([^"\']+)',
523 r'<!--\s+SpaceID=(\d+)'
524 ], webpage, 'spaceid')
525
526 content_id = self._search_regex(
527 r'<script[^>]+class=["\']yvpub-player["\'][^>]+contentid=(?P<contentid>[^&"\']+)',
528 webpage, 'contentid', group='contentid')
529
530 json_data = self._download_json(
531 'https://feapi-yvpub.yahooapis.jp/v1/content/%s' % content_id,
532 content_id,
533 query={
534 'appid': 'dj0zaiZpPVZMTVFJR0FwZWpiMyZzPWNvbnN1bWVyc2VjcmV0Jng9YjU-',
535 'output': 'json',
536 'space_id': space_id,
537 'domain': host,
538 'ak': hashlib.md5('_'.join((space_id, host)).encode()).hexdigest(),
539 'device_type': '1100',
540 })
541 formats = self._extract_formats(json_data, content_id)
542
543 return {
544 'id': content_id,
545 'title': title,
546 'description': description,
547 'thumbnail': thumbnail,
548 'formats': formats,
549 }