]>
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 | } | |
251 | } | |
cfe50f04 | 252 | ] |
9b122384 PH |
253 | |
254 | def report_download_webpage(self, video_id): | |
255 | """Report webpage download.""" | |
256 | if not self._downloader.params.get('test', False): | |
79649588 | 257 | self._downloader.report_warning('Falling back on generic information extractor.') |
9b122384 PH |
258 | super(GenericIE, self).report_download_webpage(video_id) |
259 | ||
260 | def report_following_redirect(self, new_url): | |
261 | """Report information extraction.""" | |
79649588 | 262 | self._downloader.to_screen('[redirect] Following redirect to %s' % new_url) |
9b122384 | 263 | |
42393ce2 | 264 | def _send_head(self, url): |
9b122384 | 265 | """Check if it is a redirect, like url shorteners, in case return the new url.""" |
9b122384 PH |
266 | |
267 | class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): | |
268 | """ | |
269 | Subclass the HTTPRedirectHandler to make it use our | |
aa94a6d3 | 270 | HEADRequest also on the redirected URL |
9b122384 PH |
271 | """ |
272 | def redirect_request(self, req, fp, code, msg, headers, newurl): | |
273 | if code in (301, 302, 303, 307): | |
274 | newurl = newurl.replace(' ', '%20') | |
275 | newheaders = dict((k,v) for k,v in req.headers.items() | |
276 | if k.lower() not in ("content-length", "content-type")) | |
ecbe1ad2 JMF |
277 | try: |
278 | # This function was deprecated in python 3.3 and removed in 3.4 | |
279 | origin_req_host = req.get_origin_req_host() | |
280 | except AttributeError: | |
281 | origin_req_host = req.origin_req_host | |
aa94a6d3 | 282 | return HEADRequest(newurl, |
9b122384 | 283 | headers=newheaders, |
ecbe1ad2 | 284 | origin_req_host=origin_req_host, |
9b122384 PH |
285 | unverifiable=True) |
286 | else: | |
287 | raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) | |
288 | ||
289 | class HTTPMethodFallback(compat_urllib_request.BaseHandler): | |
290 | """ | |
291 | Fallback to GET if HEAD is not allowed (405 HTTP error) | |
292 | """ | |
293 | def http_error_405(self, req, fp, code, msg, headers): | |
294 | fp.read() | |
295 | fp.close() | |
296 | ||
297 | newheaders = dict((k,v) for k,v in req.headers.items() | |
298 | if k.lower() not in ("content-length", "content-type")) | |
299 | return self.parent.open(compat_urllib_request.Request(req.get_full_url(), | |
300 | headers=newheaders, | |
301 | origin_req_host=req.get_origin_req_host(), | |
302 | unverifiable=True)) | |
303 | ||
304 | # Build our opener | |
305 | opener = compat_urllib_request.OpenerDirector() | |
306 | for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, | |
307 | HTTPMethodFallback, HEADRedirectHandler, | |
308 | compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: | |
309 | opener.add_handler(handler()) | |
310 | ||
aa94a6d3 | 311 | response = opener.open(HEADRequest(url)) |
9b122384 | 312 | if response is None: |
79649588 | 313 | raise ExtractorError('Invalid URL protocol') |
42393ce2 | 314 | return response |
9b122384 | 315 | |
4fc946b5 PH |
316 | def _extract_rss(self, url, video_id, doc): |
317 | playlist_title = doc.find('./channel/title').text | |
318 | playlist_desc_el = doc.find('./channel/description') | |
319 | playlist_desc = None if playlist_desc_el is None else playlist_desc_el.text | |
320 | ||
321 | entries = [{ | |
322 | '_type': 'url', | |
323 | 'url': e.find('link').text, | |
324 | 'title': e.find('title').text, | |
325 | } for e in doc.findall('./channel/item')] | |
326 | ||
327 | return { | |
328 | '_type': 'playlist', | |
329 | 'id': url, | |
330 | 'title': playlist_title, | |
331 | 'description': playlist_desc, | |
332 | 'entries': entries, | |
333 | } | |
334 | ||
9b122384 | 335 | def _real_extract(self, url): |
a7130543 JMF |
336 | parsed_url = compat_urlparse.urlparse(url) |
337 | if not parsed_url.scheme: | |
04b4d394 PH |
338 | default_search = self._downloader.params.get('default_search') |
339 | if default_search is None: | |
9c1fc022 | 340 | default_search = 'auto_warning' |
04b4d394 | 341 | |
9c1fc022 | 342 | if default_search in ('auto', 'auto_warning'): |
04b4d394 PH |
343 | if '/' in url: |
344 | self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') | |
345 | return self.url_result('http://' + url) | |
346 | else: | |
9c1fc022 PH |
347 | if default_search == 'auto_warning': |
348 | self._downloader.report_warning( | |
349 | 'Falling back to youtube search for %s . Set --default-search to "auto" to suppress this warning.' % url) | |
04b4d394 PH |
350 | return self.url_result('ytsearch:' + url) |
351 | else: | |
352 | assert ':' in default_search | |
353 | return self.url_result(default_search + url) | |
8012bd24 | 354 | video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0] |
a7130543 | 355 | |
79649588 | 356 | self.to_screen('%s: Requesting header' % video_id) |
c1d1facd | 357 | |
30934689 | 358 | try: |
42393ce2 PH |
359 | response = self._send_head(url) |
360 | ||
361 | # Check for redirect | |
362 | new_url = response.geturl() | |
363 | if url != new_url: | |
364 | self.report_following_redirect(new_url) | |
cecaaf3f | 365 | return self.url_result(new_url) |
42393ce2 PH |
366 | |
367 | # Check for direct link to a video | |
368 | content_type = response.headers.get('Content-Type', '') | |
3e785145 | 369 | m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type) |
42393ce2 PH |
370 | if m: |
371 | upload_date = response.headers.get('Last-Modified') | |
372 | if upload_date: | |
373 | upload_date = unified_strdate(upload_date) | |
42393ce2 PH |
374 | return { |
375 | 'id': video_id, | |
376 | 'title': os.path.splitext(url_basename(url))[0], | |
377 | 'formats': [{ | |
378 | 'format_id': m.group('format_id'), | |
379 | 'url': url, | |
79649588 | 380 | 'vcodec': 'none' if m.group('type') == 'audio' else None |
42393ce2 PH |
381 | }], |
382 | 'upload_date': upload_date, | |
383 | } | |
384 | ||
30934689 PH |
385 | except compat_urllib_error.HTTPError: |
386 | # This may be a stupid server that doesn't like HEAD, our UA, or so | |
387 | pass | |
9b122384 | 388 | |
9b122384 PH |
389 | try: |
390 | webpage = self._download_webpage(url, video_id) | |
391 | except ValueError: | |
392 | # since this is the last-resort InfoExtractor, if | |
393 | # this error is thrown, it'll be thrown here | |
79649588 | 394 | raise ExtractorError('Failed to download URL: %s' % url) |
9b122384 PH |
395 | |
396 | self.report_extraction(video_id) | |
887c6acd | 397 | |
4fc946b5 PH |
398 | # Is it an RSS feed? |
399 | try: | |
bcf89ce6 | 400 | doc = parse_xml(webpage) |
4fc946b5 PH |
401 | if doc.tag == 'rss': |
402 | return self._extract_rss(url, video_id, doc) | |
f7300c5c | 403 | except compat_xml_parse_error: |
4fc946b5 PH |
404 | pass |
405 | ||
14390730 S |
406 | # Sometimes embedded video player is hidden behind percent encoding |
407 | # (e.g. https://github.com/rg3/youtube-dl/issues/2448) | |
408 | # Unescaping the whole page allows to handle those cases in a generic way | |
1f7659db S |
409 | webpage = compat_urllib_parse.unquote(webpage) |
410 | ||
887c6acd PH |
411 | # it's tempting to parse this further, but you would |
412 | # have to take into account all the variations like | |
413 | # Video Title - Site Name | |
414 | # Site Name | Video Title | |
415 | # Video Title - Tagline | Site Name | |
416 | # and so on and so forth; it's just not practical | |
ef4fd848 | 417 | video_title = self._html_search_regex( |
79649588 PH |
418 | r'(?s)<title>(.*?)</title>', webpage, 'video title', |
419 | default='video') | |
ef4fd848 PH |
420 | |
421 | # video uploader is domain name | |
422 | video_uploader = self._search_regex( | |
79649588 | 423 | r'^(?:https?://)?([^/]*)/.*', url, 'video uploader') |
887c6acd | 424 | |
627a91a9 | 425 | # Look for BrightCove: |
99877772 PH |
426 | bc_urls = BrightcoveIE._extract_brightcove_urls(webpage) |
427 | if bc_urls: | |
79649588 | 428 | self.to_screen('Brightcove video detected.') |
99877772 PH |
429 | entries = [{ |
430 | '_type': 'url', | |
431 | 'url': smuggle_url(bc_url, {'Referer': url}), | |
432 | 'ie_key': 'Brightcove' | |
433 | } for bc_url in bc_urls] | |
434 | ||
435 | return { | |
436 | '_type': 'playlist', | |
437 | 'title': video_title, | |
438 | 'id': video_id, | |
439 | 'entries': entries, | |
440 | } | |
cfe50f04 | 441 | |
7115ca84 | 442 | # Look for embedded (iframe) Vimeo player |
9d4660ca | 443 | mobj = re.search( |
15fd51b3 | 444 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//player\.vimeo\.com/video/.+?)\1', webpage) |
9d4660ca | 445 | if mobj: |
15fd51b3 | 446 | player_url = unescapeHTML(mobj.group('url')) |
9d4660ca PH |
447 | surl = smuggle_url(player_url, {'Referer': url}) |
448 | return self.url_result(surl, 'Vimeo') | |
449 | ||
7115ca84 PH |
450 | # Look for embedded (swf embed) Vimeo player |
451 | mobj = re.search( | |
c3f51436 | 452 | r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage) |
7115ca84 PH |
453 | if mobj: |
454 | return self.url_result(mobj.group(1), 'Vimeo') | |
455 | ||
53c1d3ef | 456 | # Look for embedded YouTube player |
1f9da904 PH |
457 | matches = re.findall(r'''(?x) |
458 | (?:<iframe[^>]+?src=|embedSWF\(\s*) | |
459 | (["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/ | |
460 | (?:embed|v)/.+?) | |
461 | \1''', webpage) | |
887c6acd PH |
462 | if matches: |
463 | urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') | |
464 | for tuppl in matches] | |
465 | return self.playlist_result( | |
466 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
53c1d3ef | 467 | |
355e4fd0 PH |
468 | # Look for embedded Dailymotion player |
469 | matches = re.findall( | |
ef4fd848 | 470 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) |
355e4fd0 PH |
471 | if matches: |
472 | urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion') | |
473 | for tuppl in matches] | |
474 | return self.playlist_result( | |
475 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
476 | ||
ef4fd848 PH |
477 | # Look for embedded Wistia player |
478 | match = re.search( | |
479 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) | |
480 | if match: | |
481 | return { | |
482 | '_type': 'url_transparent', | |
483 | 'url': unescapeHTML(match.group('url')), | |
484 | 'ie_key': 'Wistia', | |
485 | 'uploader': video_uploader, | |
486 | 'title': video_title, | |
487 | 'id': video_id, | |
488 | } | |
489 | ||
ee3e63e4 | 490 | # Look for embedded blip.tv player |
19dab5e6 | 491 | mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage) |
ee3e63e4 | 492 | if mobj: |
19dab5e6 | 493 | return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV') |
494 | mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage) | |
ee3e63e4 | 495 | if mobj: |
19dab5e6 | 496 | return self.url_result(mobj.group(1), 'BlipTV') |
ee3e63e4 | 497 | |
fa35cdad PH |
498 | # Look for embedded condenast player |
499 | matches = re.findall( | |
500 | r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="(https?://player\.cnevids\.com/embed/[^"]+")', | |
501 | webpage) | |
502 | if matches: | |
503 | return { | |
504 | '_type': 'playlist', | |
505 | 'entries': [{ | |
506 | '_type': 'url', | |
507 | 'ie_key': 'CondeNast', | |
508 | 'url': ma, | |
509 | } for ma in matches], | |
510 | 'title': video_title, | |
511 | 'id': video_id, | |
512 | } | |
513 | ||
c19f7764 JMF |
514 | # Look for Bandcamp pages with custom domain |
515 | mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) | |
516 | if mobj is not None: | |
517 | burl = unescapeHTML(mobj.group(1)) | |
09804265 JMF |
518 | # Don't set the extractor because it can be a track url or an album |
519 | return self.url_result(burl) | |
c19f7764 | 520 | |
f25571ff PH |
521 | # Look for embedded Vevo player |
522 | mobj = re.search( | |
523 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage) | |
524 | if mobj is not None: | |
525 | return self.url_result(mobj.group('url')) | |
526 | ||
c0d0b01f | 527 | # Look for Ooyala videos |
750f9020 JMF |
528 | mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or |
529 | re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage)) | |
c0d0b01f | 530 | if mobj is not None: |
750f9020 | 531 | return OoyalaIE._build_url_result(mobj.group('ec')) |
c0d0b01f | 532 | |
aa94a6d3 | 533 | # Look for Aparat videos |
48099643 | 534 | mobj = re.search(r'<iframe .*?src="(http://www\.aparat\.com/video/[^"]+)"', webpage) |
aa94a6d3 PH |
535 | if mobj is not None: |
536 | return self.url_result(mobj.group(1), 'Aparat') | |
537 | ||
c93c2ab1 | 538 | # Look for MPORA videos |
c3f51436 | 539 | mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage) |
c93c2ab1 PH |
540 | if mobj is not None: |
541 | return self.url_result(mobj.group(1), 'Mpora') | |
5f59ee79 | 542 | |
15c0e8e7 | 543 | # Look for embedded NovaMov-based player |
8f89e687 | 544 | mobj = re.search( |
15c0e8e7 S |
545 | r'''(?x)<iframe[^>]+?src=(["\']) |
546 | (?P<url>http://(?:(?:embed|www)\.)? | |
547 | (?:novamov\.com| | |
548 | nowvideo\.(?:ch|sx|eu|at|ag|co)| | |
549 | videoweed\.(?:es|com)| | |
550 | movshare\.(?:net|sx|ag)| | |
551 | divxstage\.(?:eu|net|ch|co|at|ag)) | |
552 | /embed\.php.+?)\1''', webpage) | |
8f89e687 | 553 | if mobj is not None: |
15c0e8e7 | 554 | return self.url_result(mobj.group('url')) |
50f56607 | 555 | |
9834872b PH |
556 | # Look for embedded Facebook player |
557 | mobj = re.search( | |
db1f3888 | 558 | r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage) |
9834872b PH |
559 | if mobj is not None: |
560 | return self.url_result(mobj.group('url'), 'Facebook') | |
561 | ||
ca97a56e S |
562 | # Look for embedded VK player |
563 | mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage) | |
564 | if mobj is not None: | |
565 | return self.url_result(mobj.group('url'), 'VK') | |
566 | ||
db1f3888 PH |
567 | # Look for embedded Huffington Post player |
568 | mobj = re.search( | |
c3f51436 | 569 | r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage) |
db1f3888 PH |
570 | if mobj is not None: |
571 | return self.url_result(mobj.group('url'), 'HuffPost') | |
572 | ||
1b86cc41 | 573 | # Look for embed.ly |
574 | mobj = re.search(r'class=["\']embedly-card["\'][^>]href=["\'](?P<url>[^"\']+)', webpage) | |
575 | if mobj is not None: | |
576 | return self.url_result(mobj.group('url')) | |
577 | mobj = re.search(r'class=["\']embedly-embed["\'][^>]src=["\'][^"\']*url=(?P<url>[^&]+)', webpage) | |
578 | if mobj is not None: | |
579 | return self.url_result(compat_urllib_parse.unquote(mobj.group('url'))) | |
580 | ||
60cc4dc4 PH |
581 | # Look for funnyordie embed |
582 | matches = re.findall(r'<iframe[^>]+?src="(https?://(?:www\.)?funnyordie\.com/embed/[^"]+)"', webpage) | |
583 | if matches: | |
584 | urlrs = [self.url_result(unescapeHTML(eurl), 'FunnyOrDie') | |
585 | for eurl in matches] | |
586 | return self.playlist_result( | |
587 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
588 | ||
93d020dd S |
589 | # Look for embedded RUTV player |
590 | rutv_url = RUTVIE._extract_url(webpage) | |
591 | if rutv_url: | |
592 | return self.url_result(rutv_url, 'RUTV') | |
593 | ||
7e2ede98 JMF |
594 | # Look for embedded TED player |
595 | mobj = re.search( | |
596 | r'<iframe[^>]+?src=(["\'])(?P<url>http://embed\.ted\.com/.+?)\1', webpage) | |
597 | if mobj is not None: | |
598 | return self.url_result(mobj.group('url'), 'TED') | |
599 | ||
5c386252 | 600 | # Look for embedded Ustream videos |
601 | mobj = re.search( | |
602 | r'<iframe[^>]+?src=(["\'])(?P<url>http://www\.ustream\.tv/embed/.+?)\1', webpage) | |
603 | if mobj is not None: | |
604 | return self.url_result(mobj.group('url'), 'Ustream') | |
605 | ||
893f8832 PH |
606 | # Look for embedded arte.tv player |
607 | mobj = re.search( | |
608 | r'<script [^>]*?src="(?P<url>http://www\.arte\.tv/playerv2/embed[^"]+)"', | |
609 | webpage) | |
610 | if mobj is not None: | |
611 | return self.url_result(mobj.group('url'), 'ArteTVEmbed') | |
612 | ||
cb3ac1c6 S |
613 | # Look for embedded smotri.com player |
614 | smotri_url = SmotriIE._extract_url(webpage) | |
615 | if smotri_url: | |
616 | return self.url_result(smotri_url, 'Smotri') | |
617 | ||
9b122384 PH |
618 | # Start with something easy: JW Player in SWFObject |
619 | mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) | |
d981cef6 PH |
620 | if mobj is None: |
621 | # Look for gorilla-vid style embedding | |
9f8928d0 | 622 | mobj = re.search(r'(?s)(?:jw_plugins|JWPlayerOptions).*?file\s*:\s*["\'](.*?)["\']', webpage) |
9b122384 PH |
623 | if mobj is None: |
624 | # Broaden the search a little bit | |
625 | mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage) | |
626 | if mobj is None: | |
627 | # Broaden the search a little bit: JWPlayer JS loader | |
2f21eb2d | 628 | mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage) |
8d7c0cca | 629 | |
9b122384 PH |
630 | if mobj is None: |
631 | # Try to find twitter cards info | |
632 | mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage) | |
633 | if mobj is None: | |
634 | # We look for Open Graph info: | |
635 | # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am) | |
636 | m_video_type = re.search(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage) | |
637 | # We only look in og:video if the MIME type is a video, don't try if it's a Flash player: | |
638 | if m_video_type is not None: | |
639 | mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage) | |
7fea7156 PH |
640 | if mobj is None: |
641 | # HTML5 video | |
08e291b5 | 642 | mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL) |
89ef304b PH |
643 | if mobj is None: |
644 | mobj = re.search( | |
645 | r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")' | |
646 | r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"', | |
647 | webpage) | |
648 | if mobj: | |
649 | new_url = mobj.group(1) | |
650 | self.report_following_redirect(new_url) | |
651 | return { | |
652 | '_type': 'url', | |
653 | 'url': new_url, | |
654 | } | |
9b122384 | 655 | if mobj is None: |
79649588 | 656 | raise ExtractorError('Unsupported URL: %s' % url) |
9b122384 PH |
657 | |
658 | # It's possible that one of the regexes | |
659 | # matched, but returned an empty group: | |
660 | if mobj.group(1) is None: | |
79649588 | 661 | raise ExtractorError('Did not find a valid video URL at %s' % url) |
9b122384 | 662 | |
08e291b5 | 663 | video_url = mobj.group(1) |
a5caba1e | 664 | video_url = compat_urlparse.urljoin(url, video_url) |
08e291b5 | 665 | video_id = compat_urllib_parse.unquote(os.path.basename(video_url)) |
9b122384 | 666 | |
fc9713a1 PH |
667 | # Sometimes, jwplayer extraction will result in a YouTube URL |
668 | if YoutubeIE.suitable(video_url): | |
669 | return self.url_result(video_url, 'Youtube') | |
670 | ||
9b122384 | 671 | # here's a fun little line of code for you: |
9b122384 PH |
672 | video_id = os.path.splitext(video_id)[0] |
673 | ||
113577e1 | 674 | return { |
fc9713a1 PH |
675 | 'id': video_id, |
676 | 'url': video_url, | |
9b122384 | 677 | 'uploader': video_uploader, |
fc9713a1 | 678 | 'title': video_title, |
113577e1 | 679 | } |