]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/yahoo.py
Annotate some highlights of the release.
[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 ExtractorError,
16 int_or_none,
17 mimetype2ext,
18 parse_iso8601,
19 smuggle_url,
20 try_get,
21 url_or_none,
22 )
23
24 from .brightcove import BrightcoveNewIE
25
26
27 class YahooIE(InfoExtractor):
28 IE_DESC = 'Yahoo screen and movies'
29 _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)'
30 _TESTS = [{
31 'url': 'http://screen.yahoo.com/julian-smith-travis-legg-watch-214727115.html',
32 'info_dict': {
33 'id': '2d25e626-2378-391f-ada0-ddaf1417e588',
34 'ext': 'mp4',
35 'title': 'Julian Smith & Travis Legg Watch Julian Smith',
36 'description': 'Julian and Travis watch Julian Smith',
37 'duration': 6863,
38 'timestamp': 1369812016,
39 'upload_date': '20130529',
40 },
41 }, {
42 'url': 'https://screen.yahoo.com/community/community-sizzle-reel-203225340.html?format=embed',
43 'md5': '7993e572fac98e044588d0b5260f4352',
44 'info_dict': {
45 'id': '4fe78544-8d48-39d8-97cd-13f205d9fcdb',
46 'ext': 'mp4',
47 'title': "Yahoo Saves 'Community'",
48 'description': 'md5:4d4145af2fd3de00cbb6c1d664105053',
49 'duration': 170,
50 'timestamp': 1406838636,
51 'upload_date': '20140731',
52 },
53 }, {
54 'url': 'https://uk.screen.yahoo.com/editor-picks/cute-raccoon-freed-drain-using-091756545.html',
55 'md5': '71298482f7c64cbb7fa064e4553ff1c1',
56 'info_dict': {
57 'id': 'b3affa53-2e14-3590-852b-0e0db6cd1a58',
58 'ext': 'webm',
59 'title': 'Cute Raccoon Freed From Drain\u00a0Using Angle Grinder',
60 'description': 'md5:f66c890e1490f4910a9953c941dee944',
61 'duration': 97,
62 'timestamp': 1414489862,
63 'upload_date': '20141028',
64 }
65 }, {
66 'url': 'http://news.yahoo.com/video/china-moses-crazy-blues-104538833.html',
67 'md5': '88e209b417f173d86186bef6e4d1f160',
68 'info_dict': {
69 'id': 'f885cf7f-43d4-3450-9fac-46ac30ece521',
70 'ext': 'mp4',
71 'title': 'China Moses Is Crazy About the Blues',
72 'description': 'md5:9900ab8cd5808175c7b3fe55b979bed0',
73 'duration': 128,
74 'timestamp': 1385722202,
75 'upload_date': '20131129',
76 }
77 }, {
78 'url': 'https://www.yahoo.com/movies/v/true-story-trailer-173000497.html',
79 'md5': '2a9752f74cb898af5d1083ea9f661b58',
80 'info_dict': {
81 'id': '071c4013-ce30-3a93-a5b2-e0413cd4a9d1',
82 'ext': 'mp4',
83 'title': '\'True Story\' Trailer',
84 'description': 'True Story',
85 'duration': 150,
86 'timestamp': 1418919206,
87 'upload_date': '20141218',
88 },
89 }, {
90 'url': 'https://gma.yahoo.com/pizza-delivery-man-surprised-huge-tip-college-kids-195200785.html',
91 'only_matching': True,
92 }, {
93 'note': 'NBC Sports embeds',
94 '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',
95 'info_dict': {
96 'id': '9CsDKds0kvHI',
97 'ext': 'flv',
98 'description': 'md5:df390f70a9ba7c95ff1daace988f0d8d',
99 'title': 'Tyler Kalinoski hits buzzer-beater to lift Davidson',
100 'upload_date': '20150313',
101 'uploader': 'NBCU-SPORTS',
102 'timestamp': 1426270238,
103 },
104 }, {
105 'url': 'https://tw.news.yahoo.com/-100120367.html',
106 'only_matching': True,
107 }, {
108 # Query result is embedded in webpage, but explicit request to video API fails with geo restriction
109 'url': 'https://screen.yahoo.com/community/communitary-community-episode-1-ladders-154501237.html',
110 'md5': '4fbafb9c9b6f07aa8f870629f6671b35',
111 'info_dict': {
112 'id': '1f32853c-a271-3eef-8cb6-f6d6872cb504',
113 'ext': 'mp4',
114 'title': 'Communitary - Community Episode 1: Ladders',
115 'description': 'md5:8fc39608213295748e1e289807838c97',
116 'duration': 1646,
117 'timestamp': 1440436550,
118 'upload_date': '20150824',
119 'series': 'Communitary',
120 'season_number': 6,
121 'episode_number': 1,
122 },
123 }, {
124 # ytwnews://cavideo/
125 'url': 'https://tw.video.yahoo.com/movie-tw/單車天使-中文版預-092316541.html',
126 'info_dict': {
127 'id': 'ba133ff2-0793-3510-b636-59dfe9ff6cff',
128 'ext': 'mp4',
129 'title': '單車天使 - 中文版預',
130 'description': '中文版預',
131 'timestamp': 1476696196,
132 'upload_date': '20161017',
133 },
134 'params': {
135 'skip_download': True,
136 },
137 }, {
138 # Contains both a Yahoo hosted video and multiple Youtube embeds
139 'url': 'https://www.yahoo.com/entertainment/gwen-stefani-reveals-the-pop-hit-she-passed-on-assigns-it-to-her-voice-contestant-instead-033045672.html',
140 'info_dict': {
141 'id': '46c5d95a-528f-3d03-b732-732fcadd51de',
142 'title': 'Gwen Stefani reveals the pop hit she passed on, assigns it to her \'Voice\' contestant instead',
143 'description': 'Gwen decided not to record this hit herself, but she decided it was the perfect fit for Kyndall Inskeep.',
144 },
145 'playlist': [{
146 'info_dict': {
147 'id': '966d4262-4fd1-3aaa-b45b-049ca6e38ba6',
148 'ext': 'mp4',
149 'title': 'Gwen Stefani reveals she turned down one of Sia\'s best songs',
150 'description': 'On "The Voice" Tuesday, Gwen Stefani told Taylor Swift which Sia hit was almost hers.',
151 'timestamp': 1572406500,
152 'upload_date': '20191030',
153 },
154 }, {
155 'info_dict': {
156 'id': '352CFDOQrKg',
157 'ext': 'mp4',
158 'title': 'Kyndal Inskeep "Performs the Hell Out of" Sia\'s "Elastic Heart" - The Voice Knockouts 2019',
159 'description': 'md5:35b61e94c2ae214bc965ff4245f80d11',
160 'uploader': 'The Voice',
161 'uploader_id': 'NBCTheVoice',
162 'upload_date': '20191029',
163 },
164 }],
165 'params': {
166 'playlistend': 2,
167 },
168 'expected_warnings': ['HTTP Error 404'],
169 }, {
170 'url': 'https://malaysia.news.yahoo.com/video/bystanders-help-ontario-policeman-bust-190932818.html',
171 'only_matching': True,
172 }, {
173 'url': 'https://es-us.noticias.yahoo.com/es-la-puerta-irrompible-que-110539379.html',
174 'only_matching': True,
175 }, {
176 'url': 'https://www.yahoo.com/entertainment/v/longtime-cbs-news-60-minutes-032036500-cbs.html',
177 'only_matching': True,
178 }]
179
180 def _real_extract(self, url):
181 url, country, display_id = re.match(self._VALID_URL, url).groups()
182 if not country:
183 country = 'us'
184 else:
185 country = country.split('-')[0]
186 api_base = 'https://%s.yahoo.com/_td/api/resource/' % country
187
188 for i, uuid in enumerate(['url=' + url, 'ymedia-alias=' + display_id]):
189 content = self._download_json(
190 api_base + 'content;getDetailView=true;uuids=["%s"]' % uuid,
191 display_id, 'Downloading content JSON metadata', fatal=i == 1)
192 if content:
193 item = content['items'][0]
194 break
195
196 if item.get('type') != 'video':
197 entries = []
198
199 cover = item.get('cover') or {}
200 if cover.get('type') == 'yvideo':
201 cover_url = cover.get('url')
202 if cover_url:
203 entries.append(self.url_result(
204 cover_url, 'Yahoo', cover.get('uuid')))
205
206 for e in item.get('body', []):
207 if e.get('type') == 'videoIframe':
208 iframe_url = e.get('url')
209 if not iframe_url:
210 continue
211 entries.append(self.url_result(iframe_url))
212
213 return self.playlist_result(
214 entries, item.get('uuid'),
215 item.get('title'), item.get('summary'))
216
217 video_id = item['uuid']
218 video = self._download_json(
219 api_base + 'VideoService.videos;view=full;video_ids=["%s"]' % video_id,
220 video_id, 'Downloading video JSON metadata')[0]
221 title = video['title']
222
223 if country == 'malaysia':
224 country = 'my'
225
226 is_live = video.get('live_state') == 'live'
227 fmts = ('m3u8',) if is_live else ('webm', 'mp4')
228
229 urls = []
230 formats = []
231 subtitles = {}
232 for fmt in fmts:
233 media_obj = self._download_json(
234 'https://video-api.yql.yahoo.com/v1/video/sapi/streams/' + video_id,
235 video_id, 'Downloading %s JSON metadata' % fmt,
236 headers=self.geo_verification_headers(), query={
237 'format': fmt,
238 'region': country.upper(),
239 })['query']['results']['mediaObj'][0]
240 msg = media_obj.get('status', {}).get('msg')
241
242 for s in media_obj.get('streams', []):
243 host = s.get('host')
244 path = s.get('path')
245 if not host or not path:
246 continue
247 s_url = host + path
248 if s.get('format') == 'm3u8':
249 formats.extend(self._extract_m3u8_formats(
250 s_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
251 continue
252 tbr = int_or_none(s.get('bitrate'))
253 formats.append({
254 'url': s_url,
255 'format_id': fmt + ('-%d' % tbr if tbr else ''),
256 'width': int_or_none(s.get('width')),
257 'height': int_or_none(s.get('height')),
258 'tbr': tbr,
259 'fps': int_or_none(s.get('framerate')),
260 })
261
262 for cc in media_obj.get('closedcaptions', []):
263 cc_url = cc.get('url')
264 if not cc_url or cc_url in urls:
265 continue
266 urls.append(cc_url)
267 subtitles.setdefault(cc.get('lang') or 'en-US', []).append({
268 'url': cc_url,
269 'ext': mimetype2ext(cc.get('content_type')),
270 })
271
272 streaming_url = video.get('streaming_url')
273 if streaming_url and not is_live:
274 formats.extend(self._extract_m3u8_formats(
275 streaming_url, video_id, 'mp4',
276 'm3u8_native', m3u8_id='hls', fatal=False))
277
278 if not formats and msg == 'geo restricted':
279 self.raise_geo_restricted()
280
281 self._sort_formats(formats)
282
283 thumbnails = []
284 for thumb in video.get('thumbnails', []):
285 thumb_url = thumb.get('url')
286 if not thumb_url:
287 continue
288 thumbnails.append({
289 'id': thumb.get('tag'),
290 'url': thumb.get('url'),
291 'width': int_or_none(thumb.get('width')),
292 'height': int_or_none(thumb.get('height')),
293 })
294
295 series_info = video.get('series_info') or {}
296
297 return {
298 'id': video_id,
299 'title': self._live_title(title) if is_live else title,
300 'formats': formats,
301 'display_id': display_id,
302 'thumbnails': thumbnails,
303 'description': clean_html(video.get('description')),
304 'timestamp': parse_iso8601(video.get('publish_time')),
305 'subtitles': subtitles,
306 'duration': int_or_none(video.get('duration')),
307 'view_count': int_or_none(video.get('view_count')),
308 'is_live': is_live,
309 'series': video.get('show_name'),
310 'season_number': int_or_none(series_info.get('season_number')),
311 'episode_number': int_or_none(series_info.get('episode_number')),
312 }
313
314
315 class YahooSearchIE(SearchInfoExtractor):
316 IE_DESC = 'Yahoo screen search'
317 _MAX_RESULTS = 1000
318 IE_NAME = 'screen.yahoo:search'
319 _SEARCH_KEY = 'yvsearch'
320
321 def _get_n_results(self, query, n):
322 """Get a specified number of results for a query"""
323 entries = []
324 for pagenum in itertools.count(0):
325 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)
326 info = self._download_json(result_url, query,
327 note='Downloading results page ' + str(pagenum + 1))
328 m = info['m']
329 results = info['results']
330
331 for (i, r) in enumerate(results):
332 if (pagenum * 30) + i >= n:
333 break
334 mobj = re.search(r'(?P<url>screen\.yahoo\.com/.*?-\d*?\.html)"', r)
335 e = self.url_result('http://' + mobj.group('url'), 'Yahoo')
336 entries.append(e)
337 if (pagenum * 30 + i >= n) or (m['last'] >= (m['total'] - 1)):
338 break
339
340 return {
341 '_type': 'playlist',
342 'id': query,
343 'entries': entries,
344 }
345
346
347 class YahooGyaOPlayerIE(InfoExtractor):
348 IE_NAME = 'yahoo:gyao:player'
349 _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})'
350 _TESTS = [{
351 'url': 'https://gyao.yahoo.co.jp/player/00998/v00818/v0000000000000008564/',
352 'info_dict': {
353 'id': '5993125228001',
354 'ext': 'mp4',
355 'title': 'フューリー 【字幕版】',
356 'description': 'md5:21e691c798a15330eda4db17a8fe45a5',
357 'uploader_id': '4235717419001',
358 'upload_date': '20190124',
359 'timestamp': 1548294365,
360 },
361 'params': {
362 # m3u8 download
363 'skip_download': True,
364 },
365 }, {
366 'url': 'https://streaming.yahoo.co.jp/c/y/01034/v00133/v0000000000000000706/',
367 'only_matching': True,
368 }, {
369 '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',
370 'only_matching': True,
371 }]
372 _GEO_BYPASS = False
373
374 def _real_extract(self, url):
375 video_id = self._match_id(url).replace('/', ':')
376 headers = self.geo_verification_headers()
377 headers['Accept'] = 'application/json'
378 resp = self._download_json(
379 'https://gyao.yahoo.co.jp/apis/playback/graphql', video_id, query={
380 'appId': 'dj00aiZpPUNJeDh2cU1RazU3UCZzPWNvbnN1bWVyc2VjcmV0Jng9NTk-',
381 'query': '''{
382 content(parameter: {contentId: "%s", logicaAgent: PC_WEB}) {
383 video {
384 delivery {
385 id
386 }
387 title
388 }
389 }
390 }''' % video_id,
391 }, headers=headers)
392 content = resp['data']['content']
393 if not content:
394 msg = resp['errors'][0]['message']
395 if msg == 'not in japan':
396 self.raise_geo_restricted(countries=['JP'])
397 raise ExtractorError(msg)
398 video = content['video']
399 return {
400 '_type': 'url_transparent',
401 'id': video_id,
402 'title': video['title'],
403 'url': smuggle_url(
404 'http://players.brightcove.net/4235717419001/SyG5P0gjb_default/index.html?videoId=' + video['delivery']['id'],
405 {'geo_countries': ['JP']}),
406 'ie_key': BrightcoveNewIE.ie_key(),
407 }
408
409
410 class YahooGyaOIE(InfoExtractor):
411 IE_NAME = 'yahoo:gyao'
412 _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})'
413 _TESTS = [{
414 'url': 'https://gyao.yahoo.co.jp/p/00449/v03102/',
415 'info_dict': {
416 'id': '00449:v03102',
417 },
418 'playlist_count': 2,
419 }, {
420 'url': 'https://streaming.yahoo.co.jp/p/y/01034/v00133/',
421 'only_matching': True,
422 }, {
423 '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',
424 'only_matching': True,
425 }, {
426 'url': 'https://gyao.yahoo.co.jp/title/5b025a49-b2e5-4dc7-945c-09c6634afacf',
427 'only_matching': True,
428 }]
429
430 def _real_extract(self, url):
431 program_id = self._match_id(url).replace('/', ':')
432 videos = self._download_json(
433 'https://gyao.yahoo.co.jp/api/programs/%s/videos' % program_id, program_id)['videos']
434 entries = []
435 for video in videos:
436 video_id = video.get('id')
437 if not video_id:
438 continue
439 entries.append(self.url_result(
440 'https://gyao.yahoo.co.jp/player/%s/' % video_id.replace(':', '/'),
441 YahooGyaOPlayerIE.ie_key(), video_id))
442 return self.playlist_result(entries, program_id)
443
444
445 class YahooJapanNewsIE(InfoExtractor):
446 IE_NAME = 'yahoo:japannews'
447 IE_DESC = 'Yahoo! Japan News'
448 _VALID_URL = r'https?://(?P<host>(?:news|headlines)\.yahoo\.co\.jp)[^\d]*(?P<id>\d[\d-]*\d)?'
449 _GEO_COUNTRIES = ['JP']
450 _TESTS = [{
451 'url': 'https://headlines.yahoo.co.jp/videonews/ann?a=20190716-00000071-ann-int',
452 'info_dict': {
453 'id': '1736242',
454 'ext': 'mp4',
455 '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',
456 '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))',
457 'thumbnail': r're:^https?://.*\.[a-zA-Z\d]{3,4}$',
458 },
459 'params': {
460 'skip_download': True,
461 },
462 }, {
463 # geo restricted
464 'url': 'https://headlines.yahoo.co.jp/hl?a=20190721-00000001-oxv-l04',
465 'only_matching': True,
466 }, {
467 'url': 'https://headlines.yahoo.co.jp/videonews/',
468 'only_matching': True,
469 }, {
470 'url': 'https://news.yahoo.co.jp',
471 'only_matching': True,
472 }, {
473 'url': 'https://news.yahoo.co.jp/byline/hashimotojunji/20190628-00131977/',
474 'only_matching': True,
475 }, {
476 'url': 'https://news.yahoo.co.jp/feature/1356',
477 'only_matching': True
478 }]
479
480 def _extract_formats(self, json_data, content_id):
481 formats = []
482
483 video_data = try_get(
484 json_data,
485 lambda x: x['ResultSet']['Result'][0]['VideoUrlSet']['VideoUrl'],
486 list)
487 for vid in video_data or []:
488 delivery = vid.get('delivery')
489 url = url_or_none(vid.get('Url'))
490 if not delivery or not url:
491 continue
492 elif delivery == 'hls':
493 formats.extend(
494 self._extract_m3u8_formats(
495 url, content_id, 'mp4', 'm3u8_native',
496 m3u8_id='hls', fatal=False))
497 else:
498 formats.append({
499 'url': url,
500 'format_id': 'http-%s' % compat_str(vid.get('bitrate', '')),
501 'height': int_or_none(vid.get('height')),
502 'width': int_or_none(vid.get('width')),
503 'tbr': int_or_none(vid.get('bitrate')),
504 })
505 self._remove_duplicate_formats(formats)
506 self._sort_formats(formats)
507
508 return formats
509
510 def _real_extract(self, url):
511 mobj = re.match(self._VALID_URL, url)
512 host = mobj.group('host')
513 display_id = mobj.group('id') or host
514
515 webpage = self._download_webpage(url, display_id)
516
517 title = self._html_search_meta(
518 ['og:title', 'twitter:title'], webpage, 'title', default=None
519 ) or self._html_search_regex('<title>([^<]+)</title>', webpage, 'title')
520
521 if display_id == host:
522 # Headline page (w/ multiple BC playlists) ('news.yahoo.co.jp', 'headlines.yahoo.co.jp/videonews/', ...)
523 stream_plists = re.findall(r'plist=(\d+)', webpage) or re.findall(r'plist["\']:\s*["\']([^"\']+)', webpage)
524 entries = [
525 self.url_result(
526 smuggle_url(
527 'http://players.brightcove.net/5690807595001/HyZNerRl7_default/index.html?playlistId=%s' % plist_id,
528 {'geo_countries': ['JP']}),
529 ie='BrightcoveNew', video_id=plist_id)
530 for plist_id in stream_plists]
531 return self.playlist_result(entries, playlist_title=title)
532
533 # Article page
534 description = self._html_search_meta(
535 ['og:description', 'description', 'twitter:description'],
536 webpage, 'description', default=None)
537 thumbnail = self._og_search_thumbnail(
538 webpage, default=None) or self._html_search_meta(
539 'twitter:image', webpage, 'thumbnail', default=None)
540 space_id = self._search_regex([
541 r'<script[^>]+class=["\']yvpub-player["\'][^>]+spaceid=([^&"\']+)',
542 r'YAHOO\.JP\.srch\.\w+link\.onLoad[^;]+spaceID["\' ]*:["\' ]+([^"\']+)',
543 r'<!--\s+SpaceID=(\d+)'
544 ], webpage, 'spaceid')
545
546 content_id = self._search_regex(
547 r'<script[^>]+class=["\']yvpub-player["\'][^>]+contentid=(?P<contentid>[^&"\']+)',
548 webpage, 'contentid', group='contentid')
549
550 json_data = self._download_json(
551 'https://feapi-yvpub.yahooapis.jp/v1/content/%s' % content_id,
552 content_id,
553 query={
554 'appid': 'dj0zaiZpPVZMTVFJR0FwZWpiMyZzPWNvbnN1bWVyc2VjcmV0Jng9YjU-',
555 'output': 'json',
556 'space_id': space_id,
557 'domain': host,
558 'ak': hashlib.md5('_'.join((space_id, host)).encode()).hexdigest(),
559 'device_type': '1100',
560 })
561 formats = self._extract_formats(json_data, content_id)
562
563 return {
564 'id': content_id,
565 'title': title,
566 'description': description,
567 'thumbnail': thumbnail,
568 'formats': formats,
569 }