diff options
Diffstat (limited to 'youtube_dl/extractor/mtv.py')
| -rw-r--r-- | youtube_dl/extractor/mtv.py | 115 |
1 files changed, 64 insertions, 51 deletions
diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index e96d3952c..ed11f521a 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -10,34 +10,8 @@ from ..utils import ( def _media_xml_tag(tag): return '{http://search.yahoo.com/mrss/}%s' % tag -class MTVIE(InfoExtractor): - _VALID_URL = r'^https?://(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$' - - _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/' - - _TESTS = [ - { - u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml', - u'file': u'853555.mp4', - u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8', - u'info_dict': { - u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"', - u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.', - }, - }, - { - u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml', - u'file': u'USCJY1331283.mp4', - u'md5': u'73b4e7fcadd88929292fe52c3ced8caf', - u'info_dict': { - u'title': u'Everything Has Changed', - u'upload_date': u'20130606', - u'uploader': u'Taylor Swift', - }, - u'skip': u'VEVO is only available in some countries', - }, - ] +class MTVServicesInfoExtractor(InfoExtractor): @staticmethod def _id_from_uri(uri): return uri.split(':')[-1] @@ -47,18 +21,22 @@ class MTVIE(InfoExtractor): def _transform_rtmp_url(rtmp_video_url): m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\..+?/.*)$', rtmp_video_url) if not m: - raise ExtractorError(u'Cannot transform RTMP url') + return rtmp_video_url base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/' return base + m.group('finalid') def _get_thumbnail_url(self, uri, itemdoc): - return 'http://mtv.mtvnimages.com/uri/' + uri + search_path = '%s/%s' % (_media_xml_tag('group'), _media_xml_tag('thumbnail')) + thumb_node = itemdoc.find(search_path) + if thumb_node is None: + return None + else: + return thumb_node.attrib['url'] def _extract_video_formats(self, metadataXml): if '/error_country_block.swf' in metadataXml: raise ExtractorError(u'This video is not available from your country.', expected=True) mdoc = xml.etree.ElementTree.fromstring(metadataXml.encode('utf-8')) - renditions = mdoc.findall('.//rendition') formats = [] for rendition in mdoc.findall('.//rendition'): @@ -93,7 +71,7 @@ class MTVIE(InfoExtractor): else: description = None - info = { + return { 'title': itemdoc.find('title').text, 'formats': self._extract_video_formats(mediagen_page), 'id': video_id, @@ -101,32 +79,67 @@ class MTVIE(InfoExtractor): 'description': description, } - # TODO: Remove when #980 has been merged - info.update(info['formats'][-1]) - - return info - def _get_videos_info(self, uri): video_id = self._id_from_uri(uri) data = compat_urllib_parse.urlencode({'uri': uri}) - infoXml = self._download_webpage(self._FEED_URL +'?' + data, video_id, - u'Downloading info') - idoc = xml.etree.ElementTree.fromstring(infoXml.encode('utf-8')) + + def fix_ampersand(s): + """ Fix unencoded ampersand in XML """ + return s.replace(u'& ', '& ') + idoc = self._download_xml( + self._FEED_URL + '?' + data, video_id, + u'Downloading info', transform_source=fix_ampersand) return [self._get_video_info(item) for item in idoc.findall('.//item')] - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') - webpage = self._download_webpage(url, video_id) +class MTVIE(MTVServicesInfoExtractor): + _VALID_URL = r'''(?x)^https?:// + (?:(?:www\.)?mtv\.com/videos/.+?/(?P<videoid>[0-9]+)/[^/]+$| + m\.mtv\.com/videos/video\.rbml\?.*?id=(?P<mgid>[^&]+))''' - # Some videos come from Vevo.com - m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";', - webpage, re.DOTALL) - if m_vevo: - vevo_id = m_vevo.group(1); - self.to_screen(u'Vevo video detected: %s' % vevo_id) - return self.url_result('vevo:%s' % vevo_id, ie='Vevo') + _FEED_URL = 'http://www.mtv.com/player/embed/AS3/rss/' - uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri') + _TESTS = [ + { + u'url': u'http://www.mtv.com/videos/misc/853555/ours-vh1-storytellers.jhtml', + u'file': u'853555.mp4', + u'md5': u'850f3f143316b1e71fa56a4edfd6e0f8', + u'info_dict': { + u'title': u'Taylor Swift - "Ours (VH1 Storytellers)"', + u'description': u'Album: Taylor Swift performs "Ours" for VH1 Storytellers at Harvey Mudd College.', + }, + }, + { + u'add_ie': ['Vevo'], + u'url': u'http://www.mtv.com/videos/taylor-swift/916187/everything-has-changed-ft-ed-sheeran.jhtml', + u'file': u'USCJY1331283.mp4', + u'md5': u'73b4e7fcadd88929292fe52c3ced8caf', + u'info_dict': { + u'title': u'Everything Has Changed', + u'upload_date': u'20130606', + u'uploader': u'Taylor Swift', + }, + u'skip': u'VEVO is only available in some countries', + }, + ] + + def _get_thumbnail_url(self, uri, itemdoc): + return 'http://mtv.mtvnimages.com/uri/' + uri + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('videoid') + uri = mobj.group('mgid') + if uri is None: + webpage = self._download_webpage(url, video_id) + + # Some videos come from Vevo.com + m_vevo = re.search(r'isVevoVideo = true;.*?vevoVideoId = "(.*?)";', + webpage, re.DOTALL) + if m_vevo: + vevo_id = m_vevo.group(1); + self.to_screen(u'Vevo video detected: %s' % vevo_id) + return self.url_result('vevo:%s' % vevo_id, ie='Vevo') + + uri = self._html_search_regex(r'/uri/(.*?)\?', webpage, u'uri') return self._get_videos_info(uri) |