]>
Commit | Line | Data |
---|---|---|
1 | # encoding: utf-8 | |
2 | ||
3 | from __future__ import unicode_literals | |
4 | ||
5 | import os | |
6 | import re | |
7 | ||
8 | from .common import InfoExtractor | |
9 | from .youtube import YoutubeIE | |
10 | from ..utils import ( | |
11 | compat_urllib_error, | |
12 | compat_urllib_parse, | |
13 | compat_urllib_request, | |
14 | compat_urlparse, | |
15 | compat_xml_parse_error, | |
16 | ||
17 | ExtractorError, | |
18 | float_or_none, | |
19 | HEADRequest, | |
20 | orderedSet, | |
21 | parse_xml, | |
22 | smuggle_url, | |
23 | unescapeHTML, | |
24 | unified_strdate, | |
25 | url_basename, | |
26 | ) | |
27 | from .brightcove import BrightcoveIE | |
28 | from .ooyala import OoyalaIE | |
29 | from .rutv import RUTVIE | |
30 | from .smotri import SmotriIE | |
31 | ||
32 | ||
33 | class GenericIE(InfoExtractor): | |
34 | IE_DESC = 'Generic downloader that works on some sites' | |
35 | _VALID_URL = r'.*' | |
36 | IE_NAME = 'generic' | |
37 | _TESTS = [ | |
38 | { | |
39 | 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', | |
40 | 'md5': '85b90ccc9d73b4acd9138d3af4c27f89', | |
41 | 'info_dict': { | |
42 | 'id': '13601338388002', | |
43 | 'ext': 'mp4', | |
44 | 'uploader': 'www.hodiho.fr', | |
45 | 'title': 'R\u00e9gis plante sa Jeep', | |
46 | } | |
47 | }, | |
48 | # bandcamp page with custom domain | |
49 | { | |
50 | 'add_ie': ['Bandcamp'], | |
51 | 'url': 'http://bronyrock.com/track/the-pony-mash', | |
52 | 'info_dict': { | |
53 | 'id': '3235767654', | |
54 | 'ext': 'mp3', | |
55 | 'title': 'The Pony Mash', | |
56 | 'uploader': 'M_Pallante', | |
57 | }, | |
58 | 'skip': 'There is a limit of 200 free downloads / month for the test song', | |
59 | }, | |
60 | # embedded brightcove video | |
61 | # it also tests brightcove videos that need to set the 'Referer' in the | |
62 | # http requests | |
63 | { | |
64 | 'add_ie': ['Brightcove'], | |
65 | 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', | |
66 | 'info_dict': { | |
67 | 'id': '2765128793001', | |
68 | 'ext': 'mp4', | |
69 | 'title': 'Le cours de bourse : l’analyse technique', | |
70 | 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9', | |
71 | 'uploader': 'BFM BUSINESS', | |
72 | }, | |
73 | 'params': { | |
74 | 'skip_download': True, | |
75 | }, | |
76 | }, | |
77 | { | |
78 | # https://github.com/rg3/youtube-dl/issues/2253 | |
79 | 'url': 'http://bcove.me/i6nfkrc3', | |
80 | 'md5': '0ba9446db037002366bab3b3eb30c88c', | |
81 | 'info_dict': { | |
82 | 'id': '3101154703001', | |
83 | 'ext': 'mp4', | |
84 | 'title': 'Still no power', | |
85 | 'uploader': 'thestar.com', | |
86 | 'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.', | |
87 | }, | |
88 | 'add_ie': ['Brightcove'], | |
89 | }, | |
90 | { | |
91 | 'url': 'http://www.championat.com/video/football/v/87/87499.html', | |
92 | 'md5': 'fb973ecf6e4a78a67453647444222983', | |
93 | 'info_dict': { | |
94 | 'id': '3414141473001', | |
95 | 'ext': 'mp4', | |
96 | 'title': 'Видео. Удаление Дзагоева (ЦСКА)', | |
97 | 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"', | |
98 | 'uploader': 'Championat', | |
99 | }, | |
100 | }, | |
101 | # Direct link to a video | |
102 | { | |
103 | 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4', | |
104 | 'md5': '67d406c2bcb6af27fa886f31aa934bbe', | |
105 | 'info_dict': { | |
106 | 'id': 'trailer', | |
107 | 'ext': 'mp4', | |
108 | 'title': 'trailer', | |
109 | 'upload_date': '20100513', | |
110 | } | |
111 | }, | |
112 | # ooyala video | |
113 | { | |
114 | 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219', | |
115 | 'md5': '5644c6ca5d5782c1d0d350dad9bd840c', | |
116 | 'info_dict': { | |
117 | 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ', | |
118 | 'ext': 'mp4', | |
119 | 'title': '2cc213299525360.mov', # that's what we get | |
120 | }, | |
121 | }, | |
122 | # google redirect | |
123 | { | |
124 | 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE', | |
125 | 'info_dict': { | |
126 | 'id': 'cmQHVoWB5FY', | |
127 | 'ext': 'mp4', | |
128 | 'upload_date': '20130224', | |
129 | 'uploader_id': 'TheVerge', | |
130 | 'description': 'Chris Ziegler takes a look at the Alcatel OneTouch Fire and the ZTE Open; two of the first Firefox OS handsets to be officially announced.', | |
131 | 'uploader': 'The Verge', | |
132 | 'title': 'First Firefox OS phones side-by-side', | |
133 | }, | |
134 | 'params': { | |
135 | 'skip_download': False, | |
136 | } | |
137 | }, | |
138 | # embed.ly video | |
139 | { | |
140 | 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/', | |
141 | 'info_dict': { | |
142 | 'id': '9ODmcdjQcHQ', | |
143 | 'ext': 'mp4', | |
144 | 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second', | |
145 | 'upload_date': '20140225', | |
146 | 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff', | |
147 | 'uploader': 'Tested', | |
148 | 'uploader_id': 'testedcom', | |
149 | }, | |
150 | # No need to test YoutubeIE here | |
151 | 'params': { | |
152 | 'skip_download': True, | |
153 | }, | |
154 | }, | |
155 | # funnyordie embed | |
156 | { | |
157 | 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns', | |
158 | 'md5': '7cf780be104d40fea7bae52eed4a470e', | |
159 | 'info_dict': { | |
160 | 'id': '18e820ec3f', | |
161 | 'ext': 'mp4', | |
162 | 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama', | |
163 | 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.', | |
164 | }, | |
165 | }, | |
166 | # RUTV embed | |
167 | { | |
168 | 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html', | |
169 | 'info_dict': { | |
170 | 'id': '776940', | |
171 | 'ext': 'mp4', | |
172 | 'title': 'Охотское море стало целиком российским', | |
173 | 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43', | |
174 | }, | |
175 | 'params': { | |
176 | # m3u8 download | |
177 | 'skip_download': True, | |
178 | }, | |
179 | }, | |
180 | # Embedded TED video | |
181 | { | |
182 | 'url': 'http://en.support.wordpress.com/videos/ted-talks/', | |
183 | 'md5': 'deeeabcc1085eb2ba205474e7235a3d5', | |
184 | 'info_dict': { | |
185 | 'id': '981', | |
186 | 'ext': 'mp4', | |
187 | 'title': 'My web playroom', | |
188 | 'uploader': 'Ze Frank', | |
189 | 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b', | |
190 | } | |
191 | }, | |
192 | # Embeded Ustream video | |
193 | { | |
194 | 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm', | |
195 | 'md5': '27b99cdb639c9b12a79bca876a073417', | |
196 | 'info_dict': { | |
197 | 'id': '45734260', | |
198 | 'ext': 'flv', | |
199 | 'uploader': 'AU SPA: The NSA and Privacy', | |
200 | 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman' | |
201 | } | |
202 | }, | |
203 | # nowvideo embed hidden behind percent encoding | |
204 | { | |
205 | 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/', | |
206 | 'md5': '2baf4ddd70f697d94b1c18cf796d5107', | |
207 | 'info_dict': { | |
208 | 'id': '06e53103ca9aa', | |
209 | 'ext': 'flv', | |
210 | 'title': 'Macross Episode 001 Watch Macross Episode 001 onl', | |
211 | 'description': 'No description', | |
212 | }, | |
213 | }, | |
214 | # arte embed | |
215 | { | |
216 | 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html', | |
217 | 'md5': '7653032cbb25bf6c80d80f217055fa43', | |
218 | 'info_dict': { | |
219 | 'id': '048195-004_PLUS7-F', | |
220 | 'ext': 'flv', | |
221 | 'title': 'X:enius', | |
222 | 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168', | |
223 | 'upload_date': '20140320', | |
224 | }, | |
225 | 'params': { | |
226 | 'skip_download': 'Requires rtmpdump' | |
227 | } | |
228 | }, | |
229 | # smotri embed | |
230 | { | |
231 | 'url': 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml', | |
232 | 'md5': 'ec40048448e9284c9a1de77bb188108b', | |
233 | 'info_dict': { | |
234 | 'id': 'v27008541fad', | |
235 | 'ext': 'mp4', | |
236 | 'title': 'Крым и Севастополь вошли в состав России', | |
237 | 'description': 'md5:fae01b61f68984c7bd2fa741e11c3175', | |
238 | 'duration': 900, | |
239 | 'upload_date': '20140318', | |
240 | 'uploader': 'rbctv_2012_4', | |
241 | 'uploader_id': 'rbctv_2012_4', | |
242 | }, | |
243 | }, | |
244 | # Condé Nast embed | |
245 | { | |
246 | 'url': 'http://www.wired.com/2014/04/honda-asimo/', | |
247 | 'md5': 'ba0dfe966fa007657bd1443ee672db0f', | |
248 | 'info_dict': { | |
249 | 'id': '53501be369702d3275860000', | |
250 | 'ext': 'mp4', | |
251 | 'title': 'Honda’s New Asimo Robot Is More Human Than Ever', | |
252 | } | |
253 | }, | |
254 | # Dailymotion embed | |
255 | { | |
256 | 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/', | |
257 | 'md5': '441aeeb82eb72c422c7f14ec533999cd', | |
258 | 'info_dict': { | |
259 | 'id': 'k2mm4bCdJ6CQ2i7c8o2', | |
260 | 'ext': 'mp4', | |
261 | 'title': 'Le Zap de Spi0n n°216 - Zapping du Web', | |
262 | 'uploader': 'Spi0n', | |
263 | }, | |
264 | 'add_ie': ['Dailymotion'], | |
265 | }, | |
266 | # YouTube embed | |
267 | { | |
268 | 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html', | |
269 | 'info_dict': { | |
270 | 'id': 'FXRb4ykk4S0', | |
271 | 'ext': 'mp4', | |
272 | 'title': 'The NBL Auction 2014', | |
273 | 'uploader': 'BADMINTON England', | |
274 | 'uploader_id': 'BADMINTONEvents', | |
275 | 'upload_date': '20140603', | |
276 | 'description': 'md5:9ef128a69f1e262a700ed83edb163a73', | |
277 | }, | |
278 | 'add_ie': ['Youtube'], | |
279 | 'params': { | |
280 | 'skip_download': True, | |
281 | } | |
282 | }, | |
283 | # MTVSercices embed | |
284 | { | |
285 | 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too', | |
286 | 'md5': '35727f82f58c76d996fc188f9755b0d5', | |
287 | 'info_dict': { | |
288 | 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9', | |
289 | 'ext': 'mp4', | |
290 | 'title': 'Review', | |
291 | 'description': 'Mario\'s life in the fast lane has never looked so good.', | |
292 | }, | |
293 | }, | |
294 | # YouTube embed via <data-embed-url=""> | |
295 | { | |
296 | 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM', | |
297 | 'info_dict': { | |
298 | 'id': 'jpSGZsgga_I', | |
299 | 'ext': 'mp4', | |
300 | 'title': 'Asphalt 8: Airborne - Launch Trailer', | |
301 | 'uploader': 'Gameloft', | |
302 | 'uploader_id': 'gameloft', | |
303 | 'upload_date': '20130821', | |
304 | 'description': 'md5:87bd95f13d8be3e7da87a5f2c443106a', | |
305 | }, | |
306 | 'params': { | |
307 | 'skip_download': True, | |
308 | } | |
309 | }, | |
310 | # Camtasia studio | |
311 | { | |
312 | 'url': 'http://www.ll.mit.edu/workshops/education/videocourses/antennas/lecture1/video/', | |
313 | 'playlist': [{ | |
314 | 'md5': '0c5e352edabf715d762b0ad4e6d9ee67', | |
315 | 'info_dict': { | |
316 | 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final', | |
317 | 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - video1', | |
318 | 'ext': 'flv', | |
319 | 'duration': 2235.90, | |
320 | } | |
321 | }, { | |
322 | 'md5': '10e4bb3aaca9fd630e273ff92d9f3c63', | |
323 | 'info_dict': { | |
324 | 'id': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final_PIP', | |
325 | 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final - pip', | |
326 | 'ext': 'flv', | |
327 | 'duration': 2235.93, | |
328 | } | |
329 | }], | |
330 | 'info_dict': { | |
331 | 'title': 'Fenn-AA_PA_Radar_Course_Lecture_1c_Final', | |
332 | } | |
333 | } | |
334 | ] | |
335 | ||
336 | def report_download_webpage(self, video_id): | |
337 | """Report webpage download.""" | |
338 | if not self._downloader.params.get('test', False): | |
339 | self._downloader.report_warning('Falling back on generic information extractor.') | |
340 | super(GenericIE, self).report_download_webpage(video_id) | |
341 | ||
342 | def report_following_redirect(self, new_url): | |
343 | """Report information extraction.""" | |
344 | self._downloader.to_screen('[redirect] Following redirect to %s' % new_url) | |
345 | ||
346 | def _send_head(self, url): | |
347 | """Check if it is a redirect, like url shorteners, in case return the new url.""" | |
348 | ||
349 | class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): | |
350 | """ | |
351 | Subclass the HTTPRedirectHandler to make it use our | |
352 | HEADRequest also on the redirected URL | |
353 | """ | |
354 | def redirect_request(self, req, fp, code, msg, headers, newurl): | |
355 | if code in (301, 302, 303, 307): | |
356 | newurl = newurl.replace(' ', '%20') | |
357 | newheaders = dict((k,v) for k,v in req.headers.items() | |
358 | if k.lower() not in ("content-length", "content-type")) | |
359 | try: | |
360 | # This function was deprecated in python 3.3 and removed in 3.4 | |
361 | origin_req_host = req.get_origin_req_host() | |
362 | except AttributeError: | |
363 | origin_req_host = req.origin_req_host | |
364 | return HEADRequest(newurl, | |
365 | headers=newheaders, | |
366 | origin_req_host=origin_req_host, | |
367 | unverifiable=True) | |
368 | else: | |
369 | raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) | |
370 | ||
371 | class HTTPMethodFallback(compat_urllib_request.BaseHandler): | |
372 | """ | |
373 | Fallback to GET if HEAD is not allowed (405 HTTP error) | |
374 | """ | |
375 | def http_error_405(self, req, fp, code, msg, headers): | |
376 | fp.read() | |
377 | fp.close() | |
378 | ||
379 | newheaders = dict((k,v) for k,v in req.headers.items() | |
380 | if k.lower() not in ("content-length", "content-type")) | |
381 | return self.parent.open(compat_urllib_request.Request(req.get_full_url(), | |
382 | headers=newheaders, | |
383 | origin_req_host=req.get_origin_req_host(), | |
384 | unverifiable=True)) | |
385 | ||
386 | # Build our opener | |
387 | opener = compat_urllib_request.OpenerDirector() | |
388 | for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, | |
389 | HTTPMethodFallback, HEADRedirectHandler, | |
390 | compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: | |
391 | opener.add_handler(handler()) | |
392 | ||
393 | response = opener.open(HEADRequest(url)) | |
394 | if response is None: | |
395 | raise ExtractorError('Invalid URL protocol') | |
396 | return response | |
397 | ||
398 | def _extract_rss(self, url, video_id, doc): | |
399 | playlist_title = doc.find('./channel/title').text | |
400 | playlist_desc_el = doc.find('./channel/description') | |
401 | playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text | |
402 | ||
403 | entries = [{ | |
404 | '_type': 'url', | |
405 | 'url': e.find('link').text, | |
406 | 'title': e.find('title').text, | |
407 | } for e in doc.findall('./channel/item')] | |
408 | ||
409 | return { | |
410 | '_type': 'playlist', | |
411 | 'id': url, | |
412 | 'title': playlist_title, | |
413 | 'description': playlist_desc, | |
414 | 'entries': entries, | |
415 | } | |
416 | ||
417 | def _extract_camtasia(self, url, video_id, webpage): | |
418 | """ Returns None if no camtasia video can be found. """ | |
419 | ||
420 | camtasia_cfg = self._search_regex( | |
421 | r'fo\.addVariable\(\s*"csConfigFile",\s*"([^"]+)"\s*\);', | |
422 | webpage, 'camtasia configuration file', default=None) | |
423 | if camtasia_cfg is None: | |
424 | return None | |
425 | ||
426 | title = self._html_search_meta('DC.title', webpage, fatal=True) | |
427 | ||
428 | camtasia_url = compat_urlparse.urljoin(url, camtasia_cfg) | |
429 | camtasia_cfg = self._download_xml( | |
430 | camtasia_url, video_id, | |
431 | note='Downloading camtasia configuration', | |
432 | errnote='Failed to download camtasia configuration') | |
433 | fileset_node = camtasia_cfg.find('./playlist/array/fileset') | |
434 | ||
435 | entries = [] | |
436 | for n in fileset_node.getchildren(): | |
437 | url_n = n.find('./uri') | |
438 | if url_n is None: | |
439 | continue | |
440 | ||
441 | entries.append({ | |
442 | 'id': os.path.splitext(url_n.text.rpartition('/')[2])[0], | |
443 | 'title': '%s - %s' % (title, n.tag), | |
444 | 'url': compat_urlparse.urljoin(url, url_n.text), | |
445 | 'duration': float_or_none(n.find('./duration').text), | |
446 | }) | |
447 | ||
448 | return { | |
449 | '_type': 'playlist', | |
450 | 'entries': entries, | |
451 | 'title': title, | |
452 | } | |
453 | ||
454 | def _real_extract(self, url): | |
455 | if url.startswith('//'): | |
456 | return { | |
457 | '_type': 'url', | |
458 | 'url': self.http_scheme() + url, | |
459 | } | |
460 | ||
461 | parsed_url = compat_urlparse.urlparse(url) | |
462 | if not parsed_url.scheme: | |
463 | default_search = self._downloader.params.get('default_search') | |
464 | if default_search is None: | |
465 | default_search = 'fixup_error' | |
466 | ||
467 | if default_search in ('auto', 'auto_warning', 'fixup_error'): | |
468 | if '/' in url: | |
469 | self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') | |
470 | return self.url_result('http://' + url) | |
471 | elif default_search != 'fixup_error': | |
472 | if default_search == 'auto_warning': | |
473 | if re.match(r'^(?:url|URL)$', url): | |
474 | raise ExtractorError( | |
475 | 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url, | |
476 | expected=True) | |
477 | else: | |
478 | self._downloader.report_warning( | |
479 | 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url) | |
480 | return self.url_result('ytsearch:' + url) | |
481 | ||
482 | if default_search in ('error', 'fixup_error'): | |
483 | raise ExtractorError( | |
484 | ('%r is not a valid URL. ' | |
485 | 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube' | |
486 | ) % (url, url), expected=True) | |
487 | else: | |
488 | assert ':' in default_search | |
489 | return self.url_result(default_search + url) | |
490 | video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0] | |
491 | ||
492 | self.to_screen('%s: Requesting header' % video_id) | |
493 | ||
494 | try: | |
495 | response = self._send_head(url) | |
496 | ||
497 | # Check for redirect | |
498 | new_url = response.geturl() | |
499 | if url != new_url: | |
500 | self.report_following_redirect(new_url) | |
501 | return self.url_result(new_url) | |
502 | ||
503 | # Check for direct link to a video | |
504 | content_type = response.headers.get('Content-Type', '') | |
505 | m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type) | |
506 | if m: | |
507 | upload_date = response.headers.get('Last-Modified') | |
508 | if upload_date: | |
509 | upload_date = unified_strdate(upload_date) | |
510 | return { | |
511 | 'id': video_id, | |
512 | 'title': os.path.splitext(url_basename(url))[0], | |
513 | 'formats': [{ | |
514 | 'format_id': m.group('format_id'), | |
515 | 'url': url, | |
516 | 'vcodec': 'none' if m.group('type') == 'audio' else None | |
517 | }], | |
518 | 'upload_date': upload_date, | |
519 | } | |
520 | ||
521 | except compat_urllib_error.HTTPError: | |
522 | # This may be a stupid server that doesn't like HEAD, our UA, or so | |
523 | pass | |
524 | ||
525 | try: | |
526 | webpage = self._download_webpage(url, video_id) | |
527 | except ValueError: | |
528 | # since this is the last-resort InfoExtractor, if | |
529 | # this error is thrown, it'll be thrown here | |
530 | raise ExtractorError('Failed to download URL: %s' % url) | |
531 | ||
532 | self.report_extraction(video_id) | |
533 | ||
534 | # Is it an RSS feed? | |
535 | try: | |
536 | doc = parse_xml(webpage) | |
537 | if doc.tag == 'rss': | |
538 | return self._extract_rss(url, video_id, doc) | |
539 | except compat_xml_parse_error: | |
540 | pass | |
541 | ||
542 | # Is it a Camtasia project? | |
543 | camtasia_res = self._extract_camtasia(url, video_id, webpage) | |
544 | if camtasia_res is not None: | |
545 | return camtasia_res | |
546 | ||
547 | # Sometimes embedded video player is hidden behind percent encoding | |
548 | # (e.g. https://github.com/rg3/youtube-dl/issues/2448) | |
549 | # Unescaping the whole page allows to handle those cases in a generic way | |
550 | webpage = compat_urllib_parse.unquote(webpage) | |
551 | ||
552 | # it's tempting to parse this further, but you would | |
553 | # have to take into account all the variations like | |
554 | # Video Title - Site Name | |
555 | # Site Name | Video Title | |
556 | # Video Title - Tagline | Site Name | |
557 | # and so on and so forth; it's just not practical | |
558 | video_title = self._html_search_regex( | |
559 | r'(?s)<title>(.*?)</title>', webpage, 'video title', | |
560 | default='video') | |
561 | ||
562 | # video uploader is domain name | |
563 | video_uploader = self._search_regex( | |
564 | r'^(?:https?://)?([^/]*)/.*', url, 'video uploader') | |
565 | ||
566 | # Helper method | |
567 | def _playlist_from_matches(matches, getter, ie=None): | |
568 | urlrs = orderedSet(self.url_result(getter(m), ie) for m in matches) | |
569 | return self.playlist_result( | |
570 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
571 | ||
572 | # Look for BrightCove: | |
573 | bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) | |
574 | if bc_urls: | |
575 | self.to_screen('Brightcove video detected.') | |
576 | entries = [{ | |
577 | '_type': 'url', | |
578 | 'url': smuggle_url(bc_url, {'Referer': url}), | |
579 | 'ie_key': 'Brightcove' | |
580 | } for bc_url in bc_urls] | |
581 | ||
582 | return { | |
583 | '_type': 'playlist', | |
584 | 'title': video_title, | |
585 | 'id': video_id, | |
586 | 'entries': entries, | |
587 | } | |
588 | ||
589 | # Look for embedded (iframe) Vimeo player | |
590 | mobj = re.search( | |
591 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage) | |
592 | if mobj: | |
593 | player_url = unescapeHTML(mobj.group('url')) | |
594 | surl = smuggle_url(player_url, {'Referer': url}) | |
595 | return self.url_result(surl, 'Vimeo') | |
596 | ||
597 | # Look for embedded (swf embed) Vimeo player | |
598 | mobj = re.search( | |
599 | r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) | |
600 | if mobj: | |
601 | return self.url_result(mobj.group(1), 'Vimeo') | |
602 | ||
603 | # Look for embedded YouTube player | |
604 | matches = re.findall(r'''(?x) | |
605 | (?: | |
606 | <iframe[^>]+?src=| | |
607 | data-video-url=| | |
608 | <embed[^>]+?src=| | |
609 | embedSWF\(?:\s* | |
610 | ) | |
611 | (["\']) | |
612 | (?P<url>(?:https?:)?//(?:www\.)?youtube\.com/ | |
613 | (?:embed|v)/.+?) | |
614 | \1''', webpage) | |
615 | if matches: | |
616 | return _playlist_from_matches( | |
617 | matches, lambda m: unescapeHTML(m[1]), ie='Youtube') | |
618 | ||
619 | # Look for embedded Dailymotion player | |
620 | matches = re.findall( | |
621 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) | |
622 | if matches: | |
623 | return _playlist_from_matches( | |
624 | matches, lambda m: unescapeHTML(m[1])) | |
625 | ||
626 | # Look for embedded Wistia player | |
627 | match = re.search( | |
628 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) | |
629 | if match: | |
630 | return { | |
631 | '_type': 'url_transparent', | |
632 | 'url': unescapeHTML(match.group('url')), | |
633 | 'ie_key': 'Wistia', | |
634 | 'uploader': video_uploader, | |
635 | 'title': video_title, | |
636 | 'id': video_id, | |
637 | } | |
638 | ||
639 | # Look for embedded blip.tv player | |
640 | mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage) | |
641 | if mobj: | |
642 | return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV') | |
643 | mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage) | |
644 | if mobj: | |
645 | return self.url_result(mobj.group(1), 'BlipTV') | |
646 | ||
647 | # Look for embedded condenast player | |
648 | matches = re.findall( | |
649 | r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")', | |
650 | webpage) | |
651 | if matches: | |
652 | return { | |
653 | '_type': 'playlist', | |
654 | 'entries': [{ | |
655 | '_type': 'url', | |
656 | 'ie_key': 'CondeNast', | |
657 | 'url': ma, | |
658 | } for ma in matches], | |
659 | 'title': video_title, | |
660 | 'id': video_id, | |
661 | } | |
662 | ||
663 | # Look for Bandcamp pages with custom domain | |
664 | mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) | |
665 | if mobj is not None: | |
666 | burl = unescapeHTML(mobj.group(1)) | |
667 | # Don't set the extractor because it can be a track url or an album | |
668 | return self.url_result(burl) | |
669 | ||
670 | # Look for embedded Vevo player | |
671 | mobj = re.search( | |
672 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage) | |
673 | if mobj is not None: | |
674 | return self.url_result(mobj.group('url')) | |
675 | ||
676 | # Look for Ooyala videos | |
677 | mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or | |
678 | re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage)) | |
679 | if mobj is not None: | |
680 | return OoyalaIE._build_url_result(mobj.group('ec')) | |
681 | ||
682 | # Look for Aparat videos | |
683 | mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage) | |
684 | if mobj is not None: | |
685 | return self.url_result(mobj.group(1), 'Aparat') | |
686 | ||
687 | # Look for MPORA videos | |
688 | mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage) | |
689 | if mobj is not None: | |
690 | return self.url_result(mobj.group(1), 'Mpora') | |
691 | ||
692 | # Look for embedded NovaMov-based player | |
693 | mobj = re.search( | |
694 | r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\']) | |
695 | (?P<url>http://(?:(?:embed|www)\.)? | |
696 | (?:novamov\.com| | |
697 | nowvideo\.(?:ch|sx|eu|at|ag|co)| | |
698 | videoweed\.(?:es|com)| | |
699 | movshare\.(?:net|sx|ag)| | |
700 | divxstage\.(?:eu|net|ch|co|at|ag)) | |
701 | /embed\.php.+?)\1''', webpage) | |
702 | if mobj is not None: | |
703 | return self.url_result(mobj.group('url')) | |
704 | ||
705 | # Look for embedded Facebook player | |
706 | mobj = re.search( | |
707 | r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage) | |
708 | if mobj is not None: | |
709 | return self.url_result(mobj.group('url'), 'Facebook') | |
710 | ||
711 | # Look for embedded VK player | |
712 | mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage) | |
713 | if mobj is not None: | |
714 | return self.url_result(mobj.group('url'), 'VK') | |
715 | ||
716 | # Look for embedded ivi player | |
717 | mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage) | |
718 | if mobj is not None: | |
719 | return self.url_result(mobj.group('url'), 'Ivi') | |
720 | ||
721 | # Look for embedded Huffington Post player | |
722 | mobj = re.search( | |
723 | r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage) | |
724 | if mobj is not None: | |
725 | return self.url_result(mobj.group('url'), 'HuffPost') | |
726 | ||
727 | # Look for embed.ly | |
728 | mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage) | |
729 | if mobj is not None: | |
730 | return self.url_result(mobj.group('url')) | |
731 | mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage) | |
732 | if mobj is not None: | |
733 | return self.url_result(compat_urllib_parse.unquote(mobj.group('url'))) | |
734 | ||
735 | # Look for funnyordie embed | |
736 | matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage) | |
737 | if matches: | |
738 | return _playlist_from_matches( | |
739 | matches, getter=unescapeHTML, ie='FunnyOrDie') | |
740 | ||
741 | # Look for embedded RUTV player | |
742 | rutv_url = RUTVIE._extract_url(webpage) | |
743 | if rutv_url: | |
744 | return self.url_result(rutv_url, 'RUTV') | |
745 | ||
746 | # Look for embedded TED player | |
747 | mobj = re.search( | |
748 | r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage) | |
749 | if mobj is not None: | |
750 | return self.url_result(mobj.group('url'), 'TED') | |
751 | ||
752 | # Look for embedded Ustream videos | |
753 | mobj = re.search( | |
754 | r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage) | |
755 | if mobj is not None: | |
756 | return self.url_result(mobj.group('url'), 'Ustream') | |
757 | ||
758 | # Look for embedded arte.tv player | |
759 | mobj = re.search( | |
760 | r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"', | |
761 | webpage) | |
762 | if mobj is not None: | |
763 | return self.url_result(mobj.group('url'), 'ArteTVEmbed') | |
764 | ||
765 | # Look for embedded smotri.com player | |
766 | smotri_url = SmotriIE._extract_url(webpage) | |
767 | if smotri_url: | |
768 | return self.url_result(smotri_url, 'Smotri') | |
769 | ||
770 | # Look for embeded soundcloud player | |
771 | mobj = re.search( | |
772 | r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"', | |
773 | webpage) | |
774 | if mobj is not None: | |
775 | url = unescapeHTML(mobj.group('url')) | |
776 | return self.url_result(url) | |
777 | ||
778 | # Look for embedded vulture.com player | |
779 | mobj = re.search( | |
780 | r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"', | |
781 | webpage) | |
782 | if mobj is not None: | |
783 | url = unescapeHTML(mobj.group('url')) | |
784 | return self.url_result(url, ie='Vulture') | |
785 | ||
786 | # Look for embedded mtvservices player | |
787 | mobj = re.search( | |
788 | r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"', | |
789 | webpage) | |
790 | if mobj is not None: | |
791 | url = unescapeHTML(mobj.group('url')) | |
792 | return self.url_result(url, ie='MTVServicesEmbedded') | |
793 | ||
794 | # Look for embedded yahoo player | |
795 | mobj = re.search( | |
796 | r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1', | |
797 | webpage) | |
798 | if mobj is not None: | |
799 | return self.url_result(mobj.group('url'), 'Yahoo') | |
800 | ||
801 | # Look for embedded sbs.com.au player | |
802 | mobj = re.search( | |
803 | r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1', | |
804 | webpage) | |
805 | if mobj is not None: | |
806 | return self.url_result(mobj.group('url'), 'SBS') | |
807 | ||
808 | # Start with something easy: JW Player in SWFObject | |
809 | found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) | |
810 | if not found: | |
811 | # Look for gorilla-vid style embedding | |
812 | found = re.findall(r'''(?sx) | |
813 | (?: | |
814 | jw_plugins| | |
815 | JWPlayerOptions| | |
816 | jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup | |
817 | ) | |
818 | .*?file\s*:\s*["\'](.*?)["\']''', webpage) | |
819 | if not found: | |
820 | # Broaden the search a little bit | |
821 | found = re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage) | |
822 | if not found: | |
823 | # Broaden the findall a little bit: JWPlayer JS loader | |
824 | found = re.findall(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage) | |
825 | if not found: | |
826 | # Try to find twitter cards info | |
827 | found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage) | |
828 | if not found: | |
829 | # We look for Open Graph info: | |
830 | # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am) | |
831 | m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage) | |
832 | # We only look in og:video if the MIME type is a video, don't try if it's a Flash player: | |
833 | if m_video_type is not None: | |
834 | def check_video(vurl): | |
835 | vpath = compat_urlparse.urlparse(vurl).path | |
836 | return '.' in vpath and not vpath.endswith('.swf') | |
837 | found = list(filter( | |
838 | check_video, | |
839 | re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))) | |
840 | if not found: | |
841 | # HTML5 video | |
842 | found = re.findall(r'(?s)<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage) | |
843 | if not found: | |
844 | found = re.search( | |
845 | r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")' | |
846 | r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"', | |
847 | webpage) | |
848 | if found: | |
849 | new_url = found.group(1) | |
850 | self.report_following_redirect(new_url) | |
851 | return { | |
852 | '_type': 'url', | |
853 | 'url': new_url, | |
854 | } | |
855 | if not found: | |
856 | raise ExtractorError('Unsupported URL: %s' % url) | |
857 | ||
858 | entries = [] | |
859 | for video_url in found: | |
860 | video_url = compat_urlparse.urljoin(url, video_url) | |
861 | video_id = compat_urllib_parse.unquote(os.path.basename(video_url)) | |
862 | ||
863 | # Sometimes, jwplayer extraction will result in a YouTube URL | |
864 | if YoutubeIE.suitable(video_url): | |
865 | entries.append(self.url_result(video_url, 'Youtube')) | |
866 | continue | |
867 | ||
868 | # here's a fun little line of code for you: | |
869 | video_id = os.path.splitext(video_id)[0] | |
870 | ||
871 | entries.append({ | |
872 | 'id': video_id, | |
873 | 'url': video_url, | |
874 | 'uploader': video_uploader, | |
875 | 'title': video_title, | |
876 | }) | |
877 | ||
878 | if len(entries) == 1: | |
879 | return entries[0] | |
880 | else: | |
881 | for num, e in enumerate(entries, start=1): | |
882 | e['title'] = '%s (%d)' % (e['title'], num) | |
883 | return { | |
884 | '_type': 'playlist', | |
885 | 'entries': entries, | |
886 | } | |
887 |