]>
Commit | Line | Data |
---|---|---|
cfe50f04 JMF |
1 | # encoding: utf-8 |
2 | ||
79649588 PH |
3 | from __future__ import unicode_literals |
4 | ||
9b122384 PH |
5 | import os |
6 | import re | |
7 | ||
8 | from .common import InfoExtractor | |
fc9713a1 | 9 | from .youtube import YoutubeIE |
9b122384 PH |
10 | from ..utils import ( |
11 | compat_urllib_error, | |
12 | compat_urllib_parse, | |
13 | compat_urllib_request, | |
a5caba1e | 14 | compat_urlparse, |
f7300c5c | 15 | compat_xml_parse_error, |
9b122384 PH |
16 | |
17 | ExtractorError, | |
aa94a6d3 | 18 | HEADRequest, |
bcf89ce6 | 19 | parse_xml, |
9d4660ca PH |
20 | smuggle_url, |
21 | unescapeHTML, | |
42393ce2 PH |
22 | unified_strdate, |
23 | url_basename, | |
9b122384 | 24 | ) |
cfe50f04 | 25 | from .brightcove import BrightcoveIE |
c0d0b01f | 26 | from .ooyala import OoyalaIE |
93d020dd | 27 | from .rutv import RUTVIE |
cb3ac1c6 | 28 | from .smotri import SmotriIE |
9b122384 | 29 | |
0838239e | 30 | |
9b122384 | 31 | class GenericIE(InfoExtractor): |
79649588 | 32 | IE_DESC = 'Generic downloader that works on some sites' |
9b122384 | 33 | _VALID_URL = r'.*' |
79649588 | 34 | IE_NAME = 'generic' |
cfe50f04 JMF |
35 | _TESTS = [ |
36 | { | |
79649588 | 37 | 'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', |
d360a146 | 38 | 'md5': '85b90ccc9d73b4acd9138d3af4c27f89', |
79649588 | 39 | 'info_dict': { |
d360a146 S |
40 | 'id': '13601338388002', |
41 | 'ext': 'mp4', | |
79649588 PH |
42 | 'uploader': 'www.hodiho.fr', |
43 | 'title': 'R\u00e9gis plante sa Jeep', | |
cfe50f04 JMF |
44 | } |
45 | }, | |
c19f7764 JMF |
46 | # bandcamp page with custom domain |
47 | { | |
79649588 PH |
48 | 'add_ie': ['Bandcamp'], |
49 | 'url': 'http://bronyrock.com/track/the-pony-mash', | |
79649588 | 50 | 'info_dict': { |
fd50bf62 S |
51 | 'id': '3235767654', |
52 | 'ext': 'mp3', | |
79649588 PH |
53 | 'title': 'The Pony Mash', |
54 | 'uploader': 'M_Pallante', | |
c19f7764 | 55 | }, |
79649588 | 56 | 'skip': 'There is a limit of 200 free downloads / month for the test song', |
c19f7764 | 57 | }, |
eeb165e6 | 58 | # embedded brightcove video |
dd5bcdc4 JMF |
59 | # it also tests brightcove videos that need to set the 'Referer' in the |
60 | # http requests | |
eeb165e6 | 61 | { |
79649588 PH |
62 | 'add_ie': ['Brightcove'], |
63 | 'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', | |
64 | 'info_dict': { | |
65 | 'id': '2765128793001', | |
66 | 'ext': 'mp4', | |
67 | 'title': 'Le cours de bourse : l’analyse technique', | |
68 | 'description': 'md5:7e9ad046e968cb2d1114004aba466fd9', | |
69 | 'uploader': 'BFM BUSINESS', | |
eeb165e6 | 70 | }, |
79649588 PH |
71 | 'params': { |
72 | 'skip_download': True, | |
eeb165e6 JMF |
73 | }, |
74 | }, | |
17ab4d3b PH |
75 | { |
76 | # https://github.com/rg3/youtube-dl/issues/2253 | |
77 | 'url': 'http://bcove.me/i6nfkrc3', | |
17ab4d3b PH |
78 | 'md5': '0ba9446db037002366bab3b3eb30c88c', |
79 | 'info_dict': { | |
fd50bf62 S |
80 | 'id': '3101154703001', |
81 | 'ext': 'mp4', | |
17ab4d3b PH |
82 | 'title': 'Still no power', |
83 | 'uploader': 'thestar.com', | |
84 | '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.', | |
85 | }, | |
86 | 'add_ie': ['Brightcove'], | |
87 | }, | |
0479c625 S |
88 | { |
89 | 'url': 'http://www.championat.com/video/football/v/87/87499.html', | |
90 | 'md5': 'fb973ecf6e4a78a67453647444222983', | |
91 | 'info_dict': { | |
92 | 'id': '3414141473001', | |
93 | 'ext': 'mp4', | |
94 | 'title': 'Видео. Удаление Дзагоева (ЦСКА)', | |
95 | 'description': 'Онлайн-трансляция матча ЦСКА - "Волга"', | |
96 | 'uploader': 'Championat', | |
97 | }, | |
98 | }, | |
42393ce2 PH |
99 | # Direct link to a video |
100 | { | |
79649588 | 101 | 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4', |
79649588 PH |
102 | 'md5': '67d406c2bcb6af27fa886f31aa934bbe', |
103 | 'info_dict': { | |
104 | 'id': 'trailer', | |
89ef304b | 105 | 'ext': 'mp4', |
79649588 PH |
106 | 'title': 'trailer', |
107 | 'upload_date': '20100513', | |
42393ce2 | 108 | } |
c0d0b01f JMF |
109 | }, |
110 | # ooyala video | |
111 | { | |
79649588 PH |
112 | 'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219', |
113 | 'md5': '5644c6ca5d5782c1d0d350dad9bd840c', | |
114 | 'info_dict': { | |
115 | 'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ', | |
116 | 'ext': 'mp4', | |
3486df38 | 117 | 'title': '2cc213299525360.mov', # that's what we get |
c0d0b01f JMF |
118 | }, |
119 | }, | |
89ef304b PH |
120 | # google redirect |
121 | { | |
122 | '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', | |
123 | 'info_dict': { | |
124 | 'id': 'cmQHVoWB5FY', | |
125 | 'ext': 'mp4', | |
126 | 'upload_date': '20130224', | |
127 | 'uploader_id': 'TheVerge', | |
128 | '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.', | |
129 | 'uploader': 'The Verge', | |
130 | 'title': 'First Firefox OS phones side-by-side', | |
131 | }, | |
132 | 'params': { | |
133 | 'skip_download': False, | |
134 | } | |
f55a1f0a | 135 | }, |
1b86cc41 | 136 | # embed.ly video |
137 | { | |
138 | 'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/', | |
139 | 'info_dict': { | |
140 | 'id': '9ODmcdjQcHQ', | |
141 | 'ext': 'mp4', | |
0a5bce56 PH |
142 | 'title': 'Tested: Grinding Coffee at 2000 Frames Per Second', |
143 | 'upload_date': '20140225', | |
144 | 'description': 'md5:06a40fbf30b220468f1e0957c0f558ff', | |
145 | 'uploader': 'Tested', | |
146 | 'uploader_id': 'testedcom', | |
1b86cc41 | 147 | }, |
148 | # No need to test YoutubeIE here | |
149 | 'params': { | |
150 | 'skip_download': True, | |
151 | }, | |
152 | }, | |
60cc4dc4 PH |
153 | # funnyordie embed |
154 | { | |
155 | 'url': 'http://www.theguardian.com/world/2014/mar/11/obama-zach-galifianakis-between-two-ferns', | |
156 | 'md5': '7cf780be104d40fea7bae52eed4a470e', | |
157 | 'info_dict': { | |
158 | 'id': '18e820ec3f', | |
159 | 'ext': 'mp4', | |
160 | 'title': 'Between Two Ferns with Zach Galifianakis: President Barack Obama', | |
161 | 'description': 'Episode 18: President Barack Obama sits down with Zach Galifianakis for his most memorable interview yet.', | |
93d020dd | 162 | }, |
60cc4dc4 | 163 | }, |
93d020dd S |
164 | # RUTV embed |
165 | { | |
166 | 'url': 'http://www.rg.ru/2014/03/15/reg-dfo/anklav-anons.html', | |
167 | 'info_dict': { | |
168 | 'id': '776940', | |
169 | 'ext': 'mp4', | |
170 | 'title': 'Охотское море стало целиком российским', | |
171 | 'description': 'md5:5ed62483b14663e2a95ebbe115eb8f43', | |
172 | }, | |
173 | 'params': { | |
174 | # m3u8 download | |
175 | 'skip_download': True, | |
176 | }, | |
aab74fa1 PH |
177 | }, |
178 | # Embedded TED video | |
179 | { | |
180 | 'url': 'http://en.support.wordpress.com/videos/ted-talks/', | |
181 | 'md5': 'deeeabcc1085eb2ba205474e7235a3d5', | |
182 | 'info_dict': { | |
183 | 'id': '981', | |
184 | 'ext': 'mp4', | |
185 | 'title': 'My web playroom', | |
186 | 'uploader': 'Ze Frank', | |
187 | 'description': 'md5:ddb2a40ecd6b6a147e400e535874947b', | |
188 | } | |
60cc4dc4 | 189 | }, |
5c386252 | 190 | # Embeded Ustream video |
191 | { | |
192 | 'url': 'http://www.american.edu/spa/pti/nsa-privacy-janus-2014.cfm', | |
193 | 'md5': '27b99cdb639c9b12a79bca876a073417', | |
194 | 'info_dict': { | |
ca6aada4 | 195 | 'id': '45734260', |
196 | 'ext': 'flv', | |
197 | 'uploader': 'AU SPA: The NSA and Privacy', | |
5c386252 | 198 | 'title': 'NSA and Privacy Forum Debate featuring General Hayden and Barton Gellman' |
199 | } | |
200 | }, | |
d95e35d6 S |
201 | # nowvideo embed hidden behind percent encoding |
202 | { | |
203 | 'url': 'http://www.waoanime.tv/the-super-dimension-fortress-macross-episode-1/', | |
204 | 'md5': '2baf4ddd70f697d94b1c18cf796d5107', | |
205 | 'info_dict': { | |
206 | 'id': '06e53103ca9aa', | |
207 | 'ext': 'flv', | |
208 | 'title': 'Macross Episode 001 Watch Macross Episode 001 onl', | |
209 | 'description': 'No description', | |
210 | }, | |
0f2a2ba1 | 211 | }, |
893f8832 PH |
212 | # arte embed |
213 | { | |
214 | 'url': 'http://www.tv-replay.fr/redirection/20-03-14/x-enius-arte-10753389.html', | |
215 | 'md5': '7653032cbb25bf6c80d80f217055fa43', | |
216 | 'info_dict': { | |
217 | 'id': '048195-004_PLUS7-F', | |
218 | 'ext': 'flv', | |
219 | 'title': 'X:enius', | |
220 | 'description': 'md5:d5fdf32ef6613cdbfd516ae658abf168', | |
221 | 'upload_date': '20140320', | |
222 | }, | |
223 | 'params': { | |
224 | 'skip_download': 'Requires rtmpdump' | |
225 | } | |
226 | }, | |
cb3ac1c6 S |
227 | # smotri embed |
228 | { | |
229 | 'url': 'http://rbctv.rbc.ru/archive/news/562949990879132.shtml', | |
230 | 'md5': 'ec40048448e9284c9a1de77bb188108b', | |
231 | 'info_dict': { | |
232 | 'id': 'v27008541fad', | |
233 | 'ext': 'mp4', | |
234 | 'title': 'Крым и Севастополь вошли в состав России', | |
235 | 'description': 'md5:fae01b61f68984c7bd2fa741e11c3175', | |
236 | 'duration': 900, | |
237 | 'upload_date': '20140318', | |
238 | 'uploader': 'rbctv_2012_4', | |
239 | 'uploader_id': 'rbctv_2012_4', | |
240 | }, | |
241 | }, | |
fa35cdad PH |
242 | # Condé Nast embed |
243 | { | |
244 | 'url': 'http://www.wired.com/2014/04/honda-asimo/', | |
245 | 'md5': 'ba0dfe966fa007657bd1443ee672db0f', | |
246 | 'info_dict': { | |
247 | 'id': '53501be369702d3275860000', | |
248 | 'ext': 'mp4', | |
249 | 'title': 'Honda’s New Asimo Robot Is More Human Than Ever', | |
250 | } | |
ebd3c7b3 PH |
251 | }, |
252 | # Dailymotion embed | |
253 | { | |
254 | 'url': 'http://www.spi0n.com/zap-spi0n-com-n216/', | |
255 | 'md5': '441aeeb82eb72c422c7f14ec533999cd', | |
256 | 'info_dict': { | |
257 | 'id': 'k2mm4bCdJ6CQ2i7c8o2', | |
258 | 'ext': 'mp4', | |
259 | 'title': 'Le Zap de Spi0n n°216 - Zapping du Web', | |
260 | 'uploader': 'Spi0n', | |
261 | }, | |
262 | 'add_ie': ['Dailymotion'], | |
2b88feed PH |
263 | }, |
264 | # YouTube embed | |
265 | { | |
266 | 'url': 'http://www.badzine.de/ansicht/datum/2014/06/09/so-funktioniert-die-neue-englische-badminton-liga.html', | |
267 | 'info_dict': { | |
268 | 'id': 'FXRb4ykk4S0', | |
269 | 'ext': 'mp4', | |
270 | 'title': 'The NBL Auction 2014', | |
271 | 'uploader': 'BADMINTON England', | |
272 | 'uploader_id': 'BADMINTONEvents', | |
273 | 'upload_date': '20140603', | |
274 | 'description': 'md5:9ef128a69f1e262a700ed83edb163a73', | |
275 | }, | |
276 | 'add_ie': ['Youtube'], | |
277 | 'params': { | |
278 | 'skip_download': True, | |
279 | } | |
280 | }, | |
c5cd249e JMF |
281 | # MTVSercices embed |
282 | { | |
283 | 'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too', | |
284 | 'md5': '35727f82f58c76d996fc188f9755b0d5', | |
285 | 'info_dict': { | |
286 | 'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9', | |
287 | 'ext': 'mp4', | |
288 | 'title': 'Review', | |
289 | 'description': 'Mario\'s life in the fast lane has never looked so good.', | |
290 | }, | |
291 | }, | |
61013473 | 292 | # YouTube embed via <data-embed-url=""> |
293 | { | |
294 | 'url': 'https://play.google.com/store/apps/details?id=com.gameloft.android.ANMP.GloftA8HM', | |
295 | 'md5': 'c267b1ab6d736057d64babaa37e07a66', | |
296 | 'info_dict': { | |
297 | 'id': 'Ybd-qmqYYpA', | |
298 | 'ext': 'mp4', | |
299 | 'title': 'Asphalt 8: Airborne - Chinese Great Wall - Android Game Trailer', | |
300 | 'uploader': 'gameloftandroid', | |
301 | 'uploader_id': 'gameloftandroid', | |
302 | 'upload_date': '20140321', | |
303 | 'description': 'md5:9c6dca5dd75b7131ce482ccf080749d6' | |
304 | } | |
fa35cdad | 305 | } |
cfe50f04 | 306 | ] |
9b122384 PH |
307 | |
308 | def report_download_webpage(self, video_id): | |
309 | """Report webpage download.""" | |
310 | if not self._downloader.params.get('test', False): | |
79649588 | 311 | self._downloader.report_warning('Falling back on generic information extractor.') |
9b122384 PH |
312 | super(GenericIE, self).report_download_webpage(video_id) |
313 | ||
314 | def report_following_redirect(self, new_url): | |
315 | """Report information extraction.""" | |
79649588 | 316 | self._downloader.to_screen('[redirect] Following redirect to %s' % new_url) |
9b122384 | 317 | |
42393ce2 | 318 | def _send_head(self, url): |
9b122384 | 319 | """Check if it is a redirect, like url shorteners, in case return the new url.""" |
9b122384 PH |
320 | |
321 | class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): | |
322 | """ | |
323 | Subclass the HTTPRedirectHandler to make it use our | |
aa94a6d3 | 324 | HEADRequest also on the redirected URL |
9b122384 PH |
325 | """ |
326 | def redirect_request(self, req, fp, code, msg, headers, newurl): | |
327 | if code in (301, 302, 303, 307): | |
328 | newurl = newurl.replace(' ', '%20') | |
329 | newheaders = dict((k,v) for k,v in req.headers.items() | |
330 | if k.lower() not in ("content-length", "content-type")) | |
ecbe1ad2 JMF |
331 | try: |
332 | # This function was deprecated in python 3.3 and removed in 3.4 | |
333 | origin_req_host = req.get_origin_req_host() | |
334 | except AttributeError: | |
335 | origin_req_host = req.origin_req_host | |
aa94a6d3 | 336 | return HEADRequest(newurl, |
9b122384 | 337 | headers=newheaders, |
ecbe1ad2 | 338 | origin_req_host=origin_req_host, |
9b122384 PH |
339 | unverifiable=True) |
340 | else: | |
341 | raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) | |
342 | ||
343 | class HTTPMethodFallback(compat_urllib_request.BaseHandler): | |
344 | """ | |
345 | Fallback to GET if HEAD is not allowed (405 HTTP error) | |
346 | """ | |
347 | def http_error_405(self, req, fp, code, msg, headers): | |
348 | fp.read() | |
349 | fp.close() | |
350 | ||
351 | newheaders = dict((k,v) for k,v in req.headers.items() | |
352 | if k.lower() not in ("content-length", "content-type")) | |
353 | return self.parent.open(compat_urllib_request.Request(req.get_full_url(), | |
354 | headers=newheaders, | |
355 | origin_req_host=req.get_origin_req_host(), | |
356 | unverifiable=True)) | |
357 | ||
358 | # Build our opener | |
359 | opener = compat_urllib_request.OpenerDirector() | |
360 | for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, | |
361 | HTTPMethodFallback, HEADRedirectHandler, | |
362 | compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: | |
363 | opener.add_handler(handler()) | |
364 | ||
aa94a6d3 | 365 | response = opener.open(HEADRequest(url)) |
9b122384 | 366 | if response is None: |
79649588 | 367 | raise ExtractorError('Invalid URL protocol') |
42393ce2 | 368 | return response |
9b122384 | 369 | |
4fc946b5 PH |
370 | def _extract_rss(self, url, video_id, doc): |
371 | playlist_title = doc.find('./channel/title').text | |
372 | playlist_desc_el = doc.find('./channel/description') | |
373 | playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text | |
374 | ||
375 | entries = [{ | |
376 | '_type': 'url', | |
377 | 'url': e.find('link').text, | |
378 | 'title': e.find('title').text, | |
379 | } for e in doc.findall('./channel/item')] | |
380 | ||
381 | return { | |
382 | '_type': 'playlist', | |
383 | 'id': url, | |
384 | 'title': playlist_title, | |
385 | 'description': playlist_desc, | |
386 | 'entries': entries, | |
387 | } | |
388 | ||
9b122384 | 389 | def _real_extract(self, url): |
ebd3c7b3 PH |
390 | if url.startswith('//'): |
391 | return { | |
392 | '_type': 'url', | |
20991253 | 393 | 'url': self.http_scheme() + url, |
ebd3c7b3 PH |
394 | } |
395 | ||
a7130543 JMF |
396 | parsed_url = compat_urlparse.urlparse(url) |
397 | if not parsed_url.scheme: | |
04b4d394 PH |
398 | default_search = self._downloader.params.get('default_search') |
399 | if default_search is None: | |
1f7ccb90 | 400 | default_search = 'fixup_error' |
04b4d394 | 401 | |
1f7ccb90 | 402 | if default_search in ('auto', 'auto_warning', 'fixup_error'): |
04b4d394 PH |
403 | if '/' in url: |
404 | self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') | |
405 | return self.url_result('http://' + url) | |
1f7ccb90 | 406 | elif default_search != 'fixup_error': |
9c1fc022 | 407 | if default_search == 'auto_warning': |
0e67ab0d PH |
408 | if re.match(r'^(?:url|URL)$', url): |
409 | raise ExtractorError( | |
410 | 'Invalid URL: %r . Call youtube-dl like this: youtube-dl -v "https://www.youtube.com/watch?v=BaW_jenozKc" ' % url, | |
411 | expected=True) | |
412 | else: | |
413 | self._downloader.report_warning( | |
7571c02c | 414 | 'Falling back to youtube search for %s . Set --default-search "auto" to suppress this warning.' % url) |
04b4d394 | 415 | return self.url_result('ytsearch:' + url) |
1f7ccb90 PH |
416 | |
417 | if default_search in ('error', 'fixup_error'): | |
7571c02c PH |
418 | raise ExtractorError( |
419 | ('%r is not a valid URL. ' | |
eef4a7a3 | 420 | 'Set --default-search "ytsearch" (or run youtube-dl "ytsearch:%s" ) to search YouTube' |
7571c02c | 421 | ) % (url, url), expected=True) |
04b4d394 PH |
422 | else: |
423 | assert ':' in default_search | |
424 | return self.url_result(default_search + url) | |
8012bd24 | 425 | video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0] |
a7130543 | 426 | |
79649588 | 427 | self.to_screen('%s: Requesting header' % video_id) |
c1d1facd | 428 | |
30934689 | 429 | try: |
42393ce2 PH |
430 | response = self._send_head(url) |
431 | ||
432 | # Check for redirect | |
433 | new_url = response.geturl() | |
434 | if url != new_url: | |
435 | self.report_following_redirect(new_url) | |
cecaaf3f | 436 | return self.url_result(new_url) |
42393ce2 PH |
437 | |
438 | # Check for direct link to a video | |
439 | content_type = response.headers.get('Content-Type', '') | |
3e785145 | 440 | m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type) |
42393ce2 PH |
441 | if m: |
442 | upload_date = response.headers.get('Last-Modified') | |
443 | if upload_date: | |
444 | upload_date = unified_strdate(upload_date) | |
42393ce2 PH |
445 | return { |
446 | 'id': video_id, | |
447 | 'title': os.path.splitext(url_basename(url))[0], | |
448 | 'formats': [{ | |
449 | 'format_id': m.group('format_id'), | |
450 | 'url': url, | |
79649588 | 451 | 'vcodec': 'none' if m.group('type') == 'audio' else None |
42393ce2 PH |
452 | }], |
453 | 'upload_date': upload_date, | |
454 | } | |
455 | ||
30934689 PH |
456 | except compat_urllib_error.HTTPError: |
457 | # This may be a stupid server that doesn't like HEAD, our UA, or so | |
458 | pass | |
9b122384 | 459 | |
9b122384 PH |
460 | try: |
461 | webpage = self._download_webpage(url, video_id) | |
462 | except ValueError: | |
463 | # since this is the last-resort InfoExtractor, if | |
464 | # this error is thrown, it'll be thrown here | |
79649588 | 465 | raise ExtractorError('Failed to download URL: %s' % url) |
9b122384 PH |
466 | |
467 | self.report_extraction(video_id) | |
887c6acd | 468 | |
4fc946b5 PH |
469 | # Is it an RSS feed? |
470 | try: | |
bcf89ce6 | 471 | doc = parse_xml(webpage) |
4fc946b5 PH |
472 | if doc.tag == 'rss': |
473 | return self._extract_rss(url, video_id, doc) | |
f7300c5c | 474 | except compat_xml_parse_error: |
4fc946b5 PH |
475 | pass |
476 | ||
14390730 S |
477 | # Sometimes embedded video player is hidden behind percent encoding |
478 | # (e.g. https://github.com/rg3/youtube-dl/issues/2448) | |
479 | # Unescaping the whole page allows to handle those cases in a generic way | |
1f7659db S |
480 | webpage = compat_urllib_parse.unquote(webpage) |
481 | ||
887c6acd PH |
482 | # it's tempting to parse this further, but you would |
483 | # have to take into account all the variations like | |
484 | # Video Title - Site Name | |
485 | # Site Name | Video Title | |
486 | # Video Title - Tagline | Site Name | |
487 | # and so on and so forth; it's just not practical | |
ef4fd848 | 488 | video_title = self._html_search_regex( |
79649588 PH |
489 | r'(?s)<title>(.*?)</title>', webpage, 'video title', |
490 | default='video') | |
ef4fd848 PH |
491 | |
492 | # video uploader is domain name | |
493 | video_uploader = self._search_regex( | |
79649588 | 494 | r'^(?:https?://)?([^/]*)/.*', url, 'video uploader') |
887c6acd | 495 | |
627a91a9 | 496 | # Look for BrightCove: |
99877772 PH |
497 | bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) |
498 | if bc_urls: | |
79649588 | 499 | self.to_screen('Brightcove video detected.') |
99877772 PH |
500 | entries = [{ |
501 | '_type': 'url', | |
502 | 'url': smuggle_url(bc_url, {'Referer': url}), | |
503 | 'ie_key': 'Brightcove' | |
504 | } for bc_url in bc_urls] | |
505 | ||
506 | return { | |
507 | '_type': 'playlist', | |
508 | 'title': video_title, | |
509 | 'id': video_id, | |
510 | 'entries': entries, | |
511 | } | |
cfe50f04 | 512 | |
7115ca84 | 513 | # Look for embedded (iframe) Vimeo player |
9d4660ca | 514 | mobj = re.search( |
15fd51b3 | 515 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage) |
9d4660ca | 516 | if mobj: |
15fd51b3 | 517 | player_url = unescapeHTML(mobj.group('url')) |
9d4660ca PH |
518 | surl = smuggle_url(player_url, {'Referer': url}) |
519 | return self.url_result(surl, 'Vimeo') | |
520 | ||
7115ca84 PH |
521 | # Look for embedded (swf embed) Vimeo player |
522 | mobj = re.search( | |
c3f51436 | 523 | r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) |
7115ca84 PH |
524 | if mobj: |
525 | return self.url_result(mobj.group(1), 'Vimeo') | |
526 | ||
53c1d3ef | 527 | # Look for embedded YouTube player |
1f9da904 | 528 | matches = re.findall(r'''(?x) |
2b88feed PH |
529 | (?: |
530 | <iframe[^>]+?src=| | |
c71dfccc | 531 | data-video-url=| |
2b88feed PH |
532 | <embed[^>]+?src=| |
533 | embedSWF\(?:\s* | |
534 | ) | |
535 | (["\']) | |
536 | (?P<url>(?:https?:)?//(?:www\.)?youtube\.com/ | |
1f9da904 PH |
537 | (?:embed|v)/.+?) |
538 | \1''', webpage) | |
887c6acd PH |
539 | if matches: |
540 | urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') | |
541 | for tuppl in matches] | |
37e3cbe2 | 542 | # First, ensure we have a duplicate free list of entries |
543 | seen = set() | |
544 | new_list = [] | |
545 | theurl = tuple(url.items()) | |
546 | if theurl not in seen: | |
547 | seen.add(theurl) | |
548 | new_list.append(url) | |
549 | urlrs = new_list | |
887c6acd PH |
550 | return self.playlist_result( |
551 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
53c1d3ef | 552 | |
355e4fd0 PH |
553 | # Look for embedded Dailymotion player |
554 | matches = re.findall( | |
ef4fd848 | 555 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) |
355e4fd0 | 556 | if matches: |
ebd3c7b3 | 557 | urlrs = [self.url_result(unescapeHTML(tuppl[1])) |
355e4fd0 | 558 | for tuppl in matches] |
37e3cbe2 | 559 | # First, ensure we have a duplicate free list of entries |
560 | seen = set() | |
561 | new_list = [] | |
562 | theurl = tuple(url.items()) | |
563 | if theurl not in seen: | |
564 | seen.add(theurl) | |
565 | new_list.append(url) | |
566 | urlrs = new_list | |
355e4fd0 PH |
567 | return self.playlist_result( |
568 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
569 | ||
ef4fd848 PH |
570 | # Look for embedded Wistia player |
571 | match = re.search( | |
572 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) | |
573 | if match: | |
574 | return { | |
575 | '_type': 'url_transparent', | |
576 | 'url': unescapeHTML(match.group('url')), | |
577 | 'ie_key': 'Wistia', | |
578 | 'uploader': video_uploader, | |
579 | 'title': video_title, | |
580 | 'id': video_id, | |
581 | } | |
582 | ||
ee3e63e4 | 583 | # Look for embedded blip.tv player |
19dab5e6 | 584 | mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage) |
ee3e63e4 | 585 | if mobj: |
19dab5e6 | 586 | return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV') |
587 | mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage) | |
ee3e63e4 | 588 | if mobj: |
19dab5e6 | 589 | return self.url_result(mobj.group(1), 'BlipTV') |
ee3e63e4 | 590 | |
fa35cdad PH |
591 | # Look for embedded condenast player |
592 | matches = re.findall( | |
593 | r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")', | |
594 | webpage) | |
595 | if matches: | |
596 | return { | |
597 | '_type': 'playlist', | |
598 | 'entries': [{ | |
599 | '_type': 'url', | |
600 | 'ie_key': 'CondeNast', | |
601 | 'url': ma, | |
602 | } for ma in matches], | |
603 | 'title': video_title, | |
604 | 'id': video_id, | |
605 | } | |
606 | ||
c19f7764 JMF |
607 | # Look for Bandcamp pages with custom domain |
608 | mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) | |
609 | if mobj is not None: | |
610 | burl = unescapeHTML(mobj.group(1)) | |
09804265 JMF |
611 | # Don't set the extractor because it can be a track url or an album |
612 | return self.url_result(burl) | |
c19f7764 | 613 | |
f25571ff PH |
614 | # Look for embedded Vevo player |
615 | mobj = re.search( | |
616 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage) | |
617 | if mobj is not None: | |
618 | return self.url_result(mobj.group('url')) | |
619 | ||
c0d0b01f | 620 | # Look for Ooyala videos |
750f9020 JMF |
621 | mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or |
622 | re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage)) | |
c0d0b01f | 623 | if mobj is not None: |
750f9020 | 624 | return OoyalaIE._build_url_result(mobj.group('ec')) |
c0d0b01f | 625 | |
aa94a6d3 | 626 | # Look for Aparat videos |
48099643 | 627 | mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage) |
aa94a6d3 PH |
628 | if mobj is not None: |
629 | return self.url_result(mobj.group(1), 'Aparat') | |
630 | ||
c93c2ab1 | 631 | # Look for MPORA videos |
c3f51436 | 632 | mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage) |
c93c2ab1 PH |
633 | if mobj is not None: |
634 | return self.url_result(mobj.group(1), 'Mpora') | |
5f59ee79 | 635 | |
15c0e8e7 | 636 | # Look for embedded NovaMov-based player |
8f89e687 | 637 | mobj = re.search( |
8dfa187b | 638 | r'''(?x)<(?:pagespeed_)?iframe[^>]+?src=(["\']) |
15c0e8e7 S |
639 | (?P<url>http://(?:(?:embed|www)\.)? |
640 | (?:novamov\.com| | |
641 | nowvideo\.(?:ch|sx|eu|at|ag|co)| | |
642 | videoweed\.(?:es|com)| | |
643 | movshare\.(?:net|sx|ag)| | |
644 | divxstage\.(?:eu|net|ch|co|at|ag)) | |
645 | /embed\.php.+?)\1''', webpage) | |
8f89e687 | 646 | if mobj is not None: |
15c0e8e7 | 647 | return self.url_result(mobj.group('url')) |
50f56607 | 648 | |
9834872b PH |
649 | # Look for embedded Facebook player |
650 | mobj = re.search( | |
db1f3888 | 651 | r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage) |
9834872b PH |
652 | if mobj is not None: |
653 | return self.url_result(mobj.group('url'), 'Facebook') | |
654 | ||
ca97a56e S |
655 | # Look for embedded VK player |
656 | mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage) | |
657 | if mobj is not None: | |
658 | return self.url_result(mobj.group('url'), 'VK') | |
659 | ||
0364fa8b S |
660 | # Look for embedded ivi player |
661 | mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage) | |
662 | if mobj is not None: | |
663 | return self.url_result(mobj.group('url'), 'Ivi') | |
664 | ||
db1f3888 PH |
665 | # Look for embedded Huffington Post player |
666 | mobj = re.search( | |
c3f51436 | 667 | r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage) |
db1f3888 PH |
668 | if mobj is not None: |
669 | return self.url_result(mobj.group('url'), 'HuffPost') | |
670 | ||
1b86cc41 | 671 | # Look for embed.ly |
672 | mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage) | |
673 | if mobj is not None: | |
674 | return self.url_result(mobj.group('url')) | |
675 | mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage) | |
676 | if mobj is not None: | |
677 | return self.url_result(compat_urllib_parse.unquote(mobj.group('url'))) | |
678 | ||
60cc4dc4 PH |
679 | # Look for funnyordie embed |
680 | matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage) | |
681 | if matches: | |
682 | urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie') | |
683 | for eurl in matches] | |
37e3cbe2 | 684 | # First, ensure we have a duplicate free list of entries |
685 | seen = set() | |
686 | new_list = [] | |
687 | theurl = tuple(url.items()) | |
688 | if theurl not in seen: | |
689 | seen.add(theurl) | |
690 | new_list.append(url) | |
691 | urlrs = new_list | |
60cc4dc4 PH |
692 | return self.playlist_result( |
693 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
694 | ||
93d020dd S |
695 | # Look for embedded RUTV player |
696 | rutv_url = RUTVIE._extract_url(webpage) | |
697 | if rutv_url: | |
698 | return self.url_result(rutv_url, 'RUTV') | |
699 | ||
7e2ede98 JMF |
700 | # Look for embedded TED player |
701 | mobj = re.search( | |
702 | r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage) | |
703 | if mobj is not None: | |
704 | return self.url_result(mobj.group('url'), 'TED') | |
705 | ||
5c386252 | 706 | # Look for embedded Ustream videos |
707 | mobj = re.search( | |
708 | r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage) | |
709 | if mobj is not None: | |
710 | return self.url_result(mobj.group('url'), 'Ustream') | |
711 | ||
893f8832 PH |
712 | # Look for embedded arte.tv player |
713 | mobj = re.search( | |
714 | r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"', | |
715 | webpage) | |
716 | if mobj is not None: | |
717 | return self.url_result(mobj.group('url'), 'ArteTVEmbed') | |
718 | ||
cb3ac1c6 S |
719 | # Look for embedded smotri.com player |
720 | smotri_url = SmotriIE._extract_url(webpage) | |
721 | if smotri_url: | |
722 | return self.url_result(smotri_url, 'Smotri') | |
723 | ||
20991253 PH |
724 | # Look for embeded soundcloud player |
725 | mobj = re.search( | |
726 | r'<iframe src="(?P<url>https?://(?:w\.)?soundcloud\.com/player[^"]+)"', | |
727 | webpage) | |
728 | if mobj is not None: | |
729 | url = unescapeHTML(mobj.group('url')) | |
730 | return self.url_result(url) | |
731 | ||
826ec77f PH |
732 | # Look for embedded vulture.com player |
733 | mobj = re.search( | |
734 | r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"', | |
735 | webpage) | |
736 | if mobj is not None: | |
737 | url = unescapeHTML(mobj.group('url')) | |
738 | return self.url_result(url, ie='Vulture') | |
739 | ||
c5cd249e JMF |
740 | # Look for embedded mtvservices player |
741 | mobj = re.search( | |
742 | r'<iframe src="(?P<url>https?://media\.mtvnservices\.com/embed/[^"]+)"', | |
743 | webpage) | |
744 | if mobj is not None: | |
745 | url = unescapeHTML(mobj.group('url')) | |
746 | return self.url_result(url, ie='MTVServicesEmbedded') | |
747 | ||
49807b4a S |
748 | # Look for embedded yahoo player |
749 | mobj = re.search( | |
750 | r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:screen|movies)\.yahoo\.com/.+?\.html\?format=embed)\1', | |
751 | webpage) | |
752 | if mobj is not None: | |
753 | return self.url_result(mobj.group('url'), 'Yahoo') | |
754 | ||
9b122384 | 755 | # Start with something easy: JW Player in SWFObject |
b30b8698 PH |
756 | found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) |
757 | if not found: | |
d981cef6 | 758 | # Look for gorilla-vid style embedding |
b30b8698 | 759 | found = re.findall(r'''(?sx) |
c0292e8a PH |
760 | (?: |
761 | jw_plugins| | |
762 | JWPlayerOptions| | |
763 | jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup | |
764 | ) | |
765 | .*?file\s*:\s*["\'](.*?)["\']''', webpage) | |
b30b8698 | 766 | if not found: |
9b122384 | 767 | # Broaden the search a little bit |
b30b8698 PH |
768 | found = re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage) |
769 | if not found: | |
770 | # Broaden the findall a little bit: JWPlayer JS loader | |
771 | found = re.findall(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage) | |
772 | if not found: | |
9b122384 | 773 | # Try to find twitter cards info |
b30b8698 PH |
774 | found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage) |
775 | if not found: | |
9b122384 PH |
776 | # We look for Open Graph info: |
777 | # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am) | |
b30b8698 | 778 | m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage) |
9b122384 PH |
779 | # We only look in og:video if the MIME type is a video, don't try if it's a Flash player: |
780 | if m_video_type is not None: | |
b30b8698 PH |
781 | found = re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage) |
782 | if not found: | |
7fea7156 | 783 | # HTML5 video |
b30b8698 PH |
784 | found = re.findall(r'(?s)<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage) |
785 | if not found: | |
a5a45015 | 786 | found = re.search( |
89ef304b PH |
787 | r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")' |
788 | r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"', | |
789 | webpage) | |
b30b8698 PH |
790 | if found: |
791 | new_url = found.group(1) | |
89ef304b PH |
792 | self.report_following_redirect(new_url) |
793 | return { | |
794 | '_type': 'url', | |
795 | 'url': new_url, | |
796 | } | |
b30b8698 | 797 | if not found: |
79649588 | 798 | raise ExtractorError('Unsupported URL: %s' % url) |
9b122384 | 799 | |
b30b8698 PH |
800 | entries = [] |
801 | for video_url in found: | |
802 | video_url = compat_urlparse.urljoin(url, video_url) | |
803 | video_id = compat_urllib_parse.unquote(os.path.basename(video_url)) | |
9b122384 | 804 | |
b30b8698 PH |
805 | # Sometimes, jwplayer extraction will result in a YouTube URL |
806 | if YoutubeIE.suitable(video_url): | |
807 | entries.append(self.url_result(video_url, 'Youtube')) | |
808 | continue | |
9b122384 | 809 | |
b30b8698 PH |
810 | # here's a fun little line of code for you: |
811 | video_id = os.path.splitext(video_id)[0] | |
fc9713a1 | 812 | |
b30b8698 PH |
813 | entries.append({ |
814 | 'id': video_id, | |
815 | 'url': video_url, | |
816 | 'uploader': video_uploader, | |
817 | 'title': video_title, | |
818 | }) | |
819 | ||
820 | if len(entries) == 1: | |
669f0e7c | 821 | return entries[0] |
b30b8698 PH |
822 | else: |
823 | for num, e in enumerate(entries, start=1): | |
824 | e['title'] = '%s (%d)' % (e['title'], num) | |
825 | return { | |
826 | '_type': 'playlist', | |
827 | 'entries': entries, | |
828 | } | |
9b122384 | 829 |