]>
Commit | Line | Data |
---|---|---|
cfe50f04 JMF |
1 | # encoding: utf-8 |
2 | ||
9b122384 PH |
3 | import os |
4 | import re | |
5 | ||
6 | from .common import InfoExtractor | |
7 | from ..utils import ( | |
8 | compat_urllib_error, | |
9 | compat_urllib_parse, | |
10 | compat_urllib_request, | |
a5caba1e | 11 | compat_urlparse, |
9b122384 PH |
12 | |
13 | ExtractorError, | |
9d4660ca PH |
14 | smuggle_url, |
15 | unescapeHTML, | |
9b122384 | 16 | ) |
cfe50f04 | 17 | from .brightcove import BrightcoveIE |
9b122384 | 18 | |
0838239e | 19 | |
9b122384 | 20 | class GenericIE(InfoExtractor): |
0f818663 | 21 | IE_DESC = u'Generic downloader that works on some sites' |
9b122384 PH |
22 | _VALID_URL = r'.*' |
23 | IE_NAME = u'generic' | |
cfe50f04 JMF |
24 | _TESTS = [ |
25 | { | |
26 | u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', | |
27 | u'file': u'13601338388002.mp4', | |
aa929c37 | 28 | u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd', |
cfe50f04 | 29 | u'info_dict': { |
0838239e | 30 | u"uploader": u"www.hodiho.fr", |
cfe50f04 JMF |
31 | u"title": u"R\u00e9gis plante sa Jeep" |
32 | } | |
33 | }, | |
9d4660ca PH |
34 | # embedded vimeo video |
35 | { | |
9ee2b5f6 | 36 | u'add_ie': ['Vimeo'], |
9d4660ca PH |
37 | u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references', |
38 | u'file': u'22444065.mp4', | |
39 | u'md5': u'2903896e23df39722c33f015af0666e2', | |
40 | u'info_dict': { | |
41 | u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011', | |
42 | u"uploader_id": u"skillsmatter", | |
43 | u"uploader": u"Skills Matter", | |
44 | } | |
c19f7764 JMF |
45 | }, |
46 | # bandcamp page with custom domain | |
47 | { | |
9ee2b5f6 | 48 | u'add_ie': ['Bandcamp'], |
c19f7764 JMF |
49 | u'url': u'http://bronyrock.com/track/the-pony-mash', |
50 | u'file': u'3235767654.mp3', | |
51 | u'info_dict': { | |
52 | u'title': u'The Pony Mash', | |
53 | u'uploader': u'M_Pallante', | |
54 | }, | |
55 | u'skip': u'There is a limit of 200 free downloads / month for the test song', | |
56 | }, | |
eeb165e6 | 57 | # embedded brightcove video |
dd5bcdc4 JMF |
58 | # it also tests brightcove videos that need to set the 'Referer' in the |
59 | # http requests | |
eeb165e6 JMF |
60 | { |
61 | u'add_ie': ['Brightcove'], | |
dd5bcdc4 | 62 | u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', |
eeb165e6 | 63 | u'info_dict': { |
dd5bcdc4 | 64 | u'id': u'2765128793001', |
eeb165e6 | 65 | u'ext': u'mp4', |
dd5bcdc4 JMF |
66 | u'title': u'Le cours de bourse : l’analyse technique', |
67 | u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9', | |
68 | u'uploader': u'BFM BUSINESS', | |
eeb165e6 JMF |
69 | }, |
70 | u'params': { | |
71 | u'skip_download': True, | |
72 | }, | |
73 | }, | |
cfe50f04 | 74 | ] |
9b122384 PH |
75 | |
76 | def report_download_webpage(self, video_id): | |
77 | """Report webpage download.""" | |
78 | if not self._downloader.params.get('test', False): | |
79 | self._downloader.report_warning(u'Falling back on generic information extractor.') | |
80 | super(GenericIE, self).report_download_webpage(video_id) | |
81 | ||
82 | def report_following_redirect(self, new_url): | |
83 | """Report information extraction.""" | |
84 | self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url) | |
85 | ||
86 | def _test_redirect(self, url): | |
87 | """Check if it is a redirect, like url shorteners, in case return the new url.""" | |
88 | class HeadRequest(compat_urllib_request.Request): | |
89 | def get_method(self): | |
90 | return "HEAD" | |
91 | ||
92 | class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): | |
93 | """ | |
94 | Subclass the HTTPRedirectHandler to make it use our | |
95 | HeadRequest also on the redirected URL | |
96 | """ | |
97 | def redirect_request(self, req, fp, code, msg, headers, newurl): | |
98 | if code in (301, 302, 303, 307): | |
99 | newurl = newurl.replace(' ', '%20') | |
100 | newheaders = dict((k,v) for k,v in req.headers.items() | |
101 | if k.lower() not in ("content-length", "content-type")) | |
102 | return HeadRequest(newurl, | |
103 | headers=newheaders, | |
104 | origin_req_host=req.get_origin_req_host(), | |
105 | unverifiable=True) | |
106 | else: | |
107 | raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) | |
108 | ||
109 | class HTTPMethodFallback(compat_urllib_request.BaseHandler): | |
110 | """ | |
111 | Fallback to GET if HEAD is not allowed (405 HTTP error) | |
112 | """ | |
113 | def http_error_405(self, req, fp, code, msg, headers): | |
114 | fp.read() | |
115 | fp.close() | |
116 | ||
117 | newheaders = dict((k,v) for k,v in req.headers.items() | |
118 | if k.lower() not in ("content-length", "content-type")) | |
119 | return self.parent.open(compat_urllib_request.Request(req.get_full_url(), | |
120 | headers=newheaders, | |
121 | origin_req_host=req.get_origin_req_host(), | |
122 | unverifiable=True)) | |
123 | ||
124 | # Build our opener | |
125 | opener = compat_urllib_request.OpenerDirector() | |
126 | for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, | |
127 | HTTPMethodFallback, HEADRedirectHandler, | |
128 | compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: | |
129 | opener.add_handler(handler()) | |
130 | ||
131 | response = opener.open(HeadRequest(url)) | |
132 | if response is None: | |
133 | raise ExtractorError(u'Invalid URL protocol') | |
134 | new_url = response.geturl() | |
135 | ||
136 | if url == new_url: | |
137 | return False | |
138 | ||
139 | self.report_following_redirect(new_url) | |
140 | return new_url | |
141 | ||
142 | def _real_extract(self, url): | |
a7130543 JMF |
143 | parsed_url = compat_urlparse.urlparse(url) |
144 | if not parsed_url.scheme: | |
145 | self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') | |
146 | return self.url_result('http://' + url) | |
147 | ||
30934689 PH |
148 | try: |
149 | new_url = self._test_redirect(url) | |
150 | if new_url: | |
cecaaf3f | 151 | return self.url_result(new_url) |
30934689 PH |
152 | except compat_urllib_error.HTTPError: |
153 | # This may be a stupid server that doesn't like HEAD, our UA, or so | |
154 | pass | |
9b122384 PH |
155 | |
156 | video_id = url.split('/')[-1] | |
157 | try: | |
158 | webpage = self._download_webpage(url, video_id) | |
159 | except ValueError: | |
160 | # since this is the last-resort InfoExtractor, if | |
161 | # this error is thrown, it'll be thrown here | |
e484c81f | 162 | raise ExtractorError(u'Failed to download URL: %s' % url) |
9b122384 PH |
163 | |
164 | self.report_extraction(video_id) | |
887c6acd PH |
165 | |
166 | # it's tempting to parse this further, but you would | |
167 | # have to take into account all the variations like | |
168 | # Video Title - Site Name | |
169 | # Site Name | Video Title | |
170 | # Video Title - Tagline | Site Name | |
171 | # and so on and so forth; it's just not practical | |
ef4fd848 PH |
172 | video_title = self._html_search_regex( |
173 | r'(?s)<title>(.*?)</title>', webpage, u'video title', | |
174 | default=u'video') | |
175 | ||
176 | # video uploader is domain name | |
177 | video_uploader = self._search_regex( | |
178 | r'^(?:https?://)?([^/]*)/.*', url, u'video uploader') | |
887c6acd | 179 | |
627a91a9 | 180 | # Look for BrightCove: |
eeb165e6 JMF |
181 | bc_url = BrightcoveIE._extract_brightcove_url(webpage) |
182 | if bc_url is not None: | |
cfe50f04 | 183 | self.to_screen(u'Brightcove video detected.') |
cfe50f04 JMF |
184 | return self.url_result(bc_url, 'Brightcove') |
185 | ||
9d4660ca PH |
186 | # Look for embedded Vimeo player |
187 | mobj = re.search( | |
53c1d3ef | 188 | r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage) |
9d4660ca PH |
189 | if mobj: |
190 | player_url = unescapeHTML(mobj.group(1)) | |
191 | surl = smuggle_url(player_url, {'Referer': url}) | |
192 | return self.url_result(surl, 'Vimeo') | |
193 | ||
53c1d3ef | 194 | # Look for embedded YouTube player |
887c6acd | 195 | matches = re.findall( |
ef4fd848 | 196 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/embed/.+?)\1', webpage) |
887c6acd PH |
197 | if matches: |
198 | urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube') | |
199 | for tuppl in matches] | |
200 | return self.playlist_result( | |
201 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
53c1d3ef | 202 | |
355e4fd0 PH |
203 | # Look for embedded Dailymotion player |
204 | matches = re.findall( | |
ef4fd848 | 205 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage) |
355e4fd0 PH |
206 | if matches: |
207 | urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion') | |
208 | for tuppl in matches] | |
209 | return self.playlist_result( | |
210 | urlrs, playlist_id=video_id, playlist_title=video_title) | |
211 | ||
ef4fd848 PH |
212 | # Look for embedded Wistia player |
213 | match = re.search( | |
214 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage) | |
215 | if match: | |
216 | return { | |
217 | '_type': 'url_transparent', | |
218 | 'url': unescapeHTML(match.group('url')), | |
219 | 'ie_key': 'Wistia', | |
220 | 'uploader': video_uploader, | |
221 | 'title': video_title, | |
222 | 'id': video_id, | |
223 | } | |
224 | ||
ee3e63e4 | 225 | # Look for embedded blip.tv player |
226 | mobj = re.search(r'<meta\s[^>]*https?://api.blip.tv/\w+/redirect/\w+/(\d+)', webpage) | |
227 | if mobj: | |
228 | return self.url_result('http://blip.tv/seo/-'+mobj.group(1), 'BlipTV') | |
229 | mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*https?://(?:\w+\.)?blip.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', webpage) | |
230 | if mobj: | |
231 | player_url = 'http://blip.tv/play/%s.x?p=1' % mobj.group(1) | |
232 | player_page = self._download_webpage(player_url, mobj.group(1)) | |
233 | blip_video_id = self._search_regex(r'data-episode-id="(\d+)', player_page, u'blip_video_id', fatal=False) | |
234 | if blip_video_id: | |
235 | return self.url_result('http://blip.tv/seo/-'+blip_video_id, 'BlipTV') | |
236 | ||
c19f7764 JMF |
237 | # Look for Bandcamp pages with custom domain |
238 | mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage) | |
239 | if mobj is not None: | |
240 | burl = unescapeHTML(mobj.group(1)) | |
09804265 JMF |
241 | # Don't set the extractor because it can be a track url or an album |
242 | return self.url_result(burl) | |
c19f7764 | 243 | |
f25571ff PH |
244 | # Look for embedded Vevo player |
245 | mobj = re.search( | |
246 | r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage) | |
247 | if mobj is not None: | |
248 | return self.url_result(mobj.group('url')) | |
249 | ||
9b122384 PH |
250 | # Start with something easy: JW Player in SWFObject |
251 | mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage) | |
252 | if mobj is None: | |
253 | # Broaden the search a little bit | |
254 | mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage) | |
255 | if mobj is None: | |
256 | # Broaden the search a little bit: JWPlayer JS loader | |
113577e1 | 257 | mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"]*)', webpage) |
9b122384 PH |
258 | if mobj is None: |
259 | # Try to find twitter cards info | |
260 | mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage) | |
261 | if mobj is None: | |
262 | # We look for Open Graph info: | |
263 | # We have to match any number spaces between elements, some sites try to align them (eg.: statigr.am) | |
264 | m_video_type = re.search(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage) | |
265 | # We only look in og:video if the MIME type is a video, don't try if it's a Flash player: | |
266 | if m_video_type is not None: | |
267 | mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage) | |
7fea7156 PH |
268 | if mobj is None: |
269 | # HTML5 video | |
08e291b5 | 270 | mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL) |
9b122384 | 271 | if mobj is None: |
e484c81f | 272 | raise ExtractorError(u'Unsupported URL: %s' % url) |
9b122384 PH |
273 | |
274 | # It's possible that one of the regexes | |
275 | # matched, but returned an empty group: | |
276 | if mobj.group(1) is None: | |
e484c81f | 277 | raise ExtractorError(u'Did not find a valid video URL at %s' % url) |
9b122384 | 278 | |
08e291b5 | 279 | video_url = mobj.group(1) |
a5caba1e | 280 | video_url = compat_urlparse.urljoin(url, video_url) |
08e291b5 | 281 | video_id = compat_urllib_parse.unquote(os.path.basename(video_url)) |
9b122384 PH |
282 | |
283 | # here's a fun little line of code for you: | |
9b122384 PH |
284 | video_id = os.path.splitext(video_id)[0] |
285 | ||
113577e1 | 286 | return { |
9b122384 PH |
287 | 'id': video_id, |
288 | 'url': video_url, | |
289 | 'uploader': video_uploader, | |
9b122384 | 290 | 'title': video_title, |
113577e1 | 291 | } |