# encoding: utf-8 import os import re from .common import InfoExtractor from ..utils import ( compat_urllib_error, compat_urllib_parse, compat_urllib_request, compat_urlparse, ExtractorError, smuggle_url, unescapeHTML, ) from .brightcove import BrightcoveIE class GenericIE(InfoExtractor): IE_DESC = u'Generic downloader that works on some sites' _VALID_URL = r'.*' IE_NAME = u'generic' _TESTS = [ { u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html', u'file': u'13601338388002.mp4', u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd', u'info_dict': { u"uploader": u"www.hodiho.fr", u"title": u"R\u00e9gis plante sa Jeep" } }, # embedded vimeo video { u'add_ie': ['Vimeo'], u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references', u'file': u'22444065.mp4', u'md5': u'2903896e23df39722c33f015af0666e2', u'info_dict': { u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011', u"uploader_id": u"skillsmatter", u"uploader": u"Skills Matter", } }, # bandcamp page with custom domain { u'add_ie': ['Bandcamp'], u'url': u'http://bronyrock.com/track/the-pony-mash', u'file': u'3235767654.mp3', u'info_dict': { u'title': u'The Pony Mash', u'uploader': u'M_Pallante', }, u'skip': u'There is a limit of 200 free downloads / month for the test song', }, # embedded brightcove video # it also tests brightcove videos that need to set the 'Referer' in the # http requests { u'add_ie': ['Brightcove'], u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/', u'info_dict': { u'id': u'2765128793001', u'ext': u'mp4', u'title': u'Le cours de bourse : l’analyse technique', u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9', u'uploader': u'BFM BUSINESS', }, u'params': { u'skip_download': True, }, }, ] def report_download_webpage(self, video_id): """Report webpage download.""" if not self._downloader.params.get('test', False): self._downloader.report_warning(u'Falling back on generic information extractor.') super(GenericIE, self).report_download_webpage(video_id) def report_following_redirect(self, new_url): """Report information extraction.""" self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url) def _test_redirect(self, url): """Check if it is a redirect, like url shorteners, in case return the new url.""" class HeadRequest(compat_urllib_request.Request): def get_method(self): return "HEAD" class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler): """ Subclass the HTTPRedirectHandler to make it use our HeadRequest also on the redirected URL """ def redirect_request(self, req, fp, code, msg, headers, newurl): if code in (301, 302, 303, 307): newurl = newurl.replace(' ', '%20') newheaders = dict((k,v) for k,v in req.headers.items() if k.lower() not in ("content-length", "content-type")) return HeadRequest(newurl, headers=newheaders, origin_req_host=req.get_origin_req_host(), unverifiable=True) else: raise compat_urllib_error.HTTPError(req.get_full_url(), code, msg, headers, fp) class HTTPMethodFallback(compat_urllib_request.BaseHandler): """ Fallback to GET if HEAD is not allowed (405 HTTP error) """ def http_error_405(self, req, fp, code, msg, headers): fp.read() fp.close() newheaders = dict((k,v) for k,v in req.headers.items() if k.lower() not in ("content-length", "content-type")) return self.parent.open(compat_urllib_request.Request(req.get_full_url(), headers=newheaders, origin_req_host=req.get_origin_req_host(), unverifiable=True)) # Build our opener opener = compat_urllib_request.OpenerDirector() for handler in [compat_urllib_request.HTTPHandler, compat_urllib_request.HTTPDefaultErrorHandler, HTTPMethodFallback, HEADRedirectHandler, compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]: opener.add_handler(handler()) response = opener.open(HeadRequest(url)) if response is None: raise ExtractorError(u'Invalid URL protocol') new_url = response.geturl() if url == new_url: return False self.report_following_redirect(new_url) return new_url def _real_extract(self, url): parsed_url = compat_urlparse.urlparse(url) if not parsed_url.scheme: self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http') return self.url_result('http://' + url) try: new_url = self._test_redirect(url) if new_url: return self.url_result(new_url) except compat_urllib_error.HTTPError: # This may be a stupid server that doesn't like HEAD, our UA, or so pass video_id = url.split('/')[-1] try: webpage = self._download_webpage(url, video_id) except ValueError: # since this is the last-resort InfoExtractor, if # this error is thrown, it'll be thrown here raise ExtractorError(u'Failed to download URL: %s' % url) self.report_extraction(video_id) # it's tempting to parse this further, but you would # have to take into account all the variations like # Video Title - Site Name # Site Name | Video Title # Video Title - Tagline | Site Name # and so on and so forth; it's just not practical video_title = self._html_search_regex( r'(?s)