diff options
Diffstat (limited to 'youtube_dl/extractor/generic.py')
| -rw-r--r-- | youtube_dl/extractor/generic.py | 47 |
1 files changed, 26 insertions, 21 deletions
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index b633e896c..d48c84f8d 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -8,11 +8,13 @@ from ..utils import ( compat_urllib_error, compat_urllib_parse, compat_urllib_request, + compat_urlparse, ExtractorError, ) from .brightcove import BrightcoveIE + class GenericIE(InfoExtractor): IE_DESC = u'Generic downloader that works on some sites' _VALID_URL = r'.*' @@ -23,21 +25,10 @@ class GenericIE(InfoExtractor): u'file': u'13601338388002.mp4', u'md5': u'85b90ccc9d73b4acd9138d3af4c27f89', u'info_dict': { - u"uploader": u"www.hodiho.fr", + u"uploader": u"www.hodiho.fr", u"title": u"R\u00e9gis plante sa Jeep" } }, - { - u'url': u'http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/', - u'file': u'2371591881001.mp4', - u'md5': u'9e80619e0a94663f0bdc849b4566af19', - u'note': u'Test Brightcove downloads and detection in GenericIE', - u'info_dict': { - u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”', - u'uploader': u'8TV', - u'description': u'md5:a950cc4285c43e44d763d036710cd9cd', - } - }, ] def report_download_webpage(self, video_id): @@ -107,8 +98,18 @@ class GenericIE(InfoExtractor): return new_url def _real_extract(self, url): - new_url = self._test_redirect(url) - if new_url: return [self.url_result(new_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: @@ -116,11 +117,11 @@ class GenericIE(InfoExtractor): except ValueError: # since this is the last-resort InfoExtractor, if # this error is thrown, it'll be thrown here - raise ExtractorError(u'Invalid URL: %s' % url) + raise ExtractorError(u'Failed to download URL: %s' % url) self.report_extraction(video_id) - # Look for BrigthCove: - m_brightcove = re.search(r'<object.+?class=([\'"]).*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL) + # Look for BrightCove: + m_brightcove = re.search(r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>', webpage, re.DOTALL) if m_brightcove is not None: self.to_screen(u'Brightcove video detected.') bc_url = BrightcoveIE._build_brighcove_url(m_brightcove.group()) @@ -145,15 +146,19 @@ class GenericIE(InfoExtractor): if m_video_type is not None: mobj = re.search(r'<meta.*?property="og:video".*?content="(.*?)"', webpage) if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) + # HTML5 video + mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL) + if mobj is None: + raise ExtractorError(u'Unsupported URL: %s' % url) # It's possible that one of the regexes # matched, but returned an empty group: if mobj.group(1) is None: - raise ExtractorError(u'Invalid URL: %s' % url) + raise ExtractorError(u'Did not find a valid video URL at %s' % url) - video_url = compat_urllib_parse.unquote(mobj.group(1)) - video_id = os.path.basename(video_url) + video_url = mobj.group(1) + video_url = compat_urlparse.urljoin(url, video_url) + video_id = compat_urllib_parse.unquote(os.path.basename(video_url)) # here's a fun little line of code for you: video_extension = os.path.splitext(video_id)[1][1:] |