X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Ftheplatform.py;h=91f2453ebf237db086f18f1ca2b30f2ce73a3728;hb=aba3231de18dbb8cdfcc653cbba137844637e65e;hp=650d7517f83ee09d4467d4338cbd7c97620121a5;hpb=f8b56e95b82d9cb783c5a789b6b2770efff57b73;p=youtube-dl diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 650d7517f..91f2453eb 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -11,7 +11,10 @@ _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language class ThePlatformIE(InfoExtractor): - _VALID_URL = r'(?:https?://link\.theplatform\.com/s/[^/]+/|theplatform:)(?P[^/\?]+)' + _VALID_URL = r'''(?x) + (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/ + (?P(?:[^/\?]+/(?:swf|config)|onsite)/select/)? + |theplatform:)(?P[^/\?&]+)''' _TEST = { # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/ @@ -29,16 +32,14 @@ class ThePlatformIE(InfoExtractor): }, } - def _get_info(self, video_id): - smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?' - 'format=smil&mbr=true'.format(video_id)) + def _get_info(self, video_id, smil_url): meta = self._download_xml(smil_url, video_id) try: error_msg = next( n.attrib['abstract'] for n in meta.findall(_x('.//smil:ref')) - if n.attrib['title'] == u'Geographic Restriction') + if n.attrib.get('title') == u'Geographic Restriction') except StopIteration: pass else: @@ -50,20 +51,39 @@ class ThePlatformIE(InfoExtractor): head = meta.find(_x('smil:head')) body = meta.find(_x('smil:body')) - base_url = head.find(_x('smil:meta')).attrib['base'] - switch = body.find(_x('smil:switch')) - formats = [] - for f in switch.findall(_x('smil:video')): - attr = f.attrib - formats.append({ - 'url': base_url, - 'play_path': 'mp4:' + attr['src'], + + f4m_node = body.find(_x('smil:seq/smil:video')) + if f4m_node is not None: + f4m_url = f4m_node.attrib['src'] + if 'manifest.f4m?' not in f4m_url: + f4m_url += '?' + # the parameters are from syfy.com, other sites may use others, + # they also work for nbc.com + f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3' + formats = [{ 'ext': 'flv', - 'width': int(attr['width']), - 'height': int(attr['height']), - 'vbr': int(attr['system-bitrate']), - }) - formats.sort(key=lambda f: (f['height'], f['width'], f['vbr'])) + 'url': f4m_url, + }] + else: + base_url = head.find(_x('smil:meta')).attrib['base'] + switch = body.find(_x('smil:switch')) + formats = [] + for f in switch.findall(_x('smil:video')): + attr = f.attrib + width = int(attr['width']) + height = int(attr['height']) + vbr = int(attr['system-bitrate']) // 1000 + format_id = '%dx%d_%dk' % (width, height, vbr) + formats.append({ + 'format_id': format_id, + 'url': base_url, + 'play_path': 'mp4:' + attr['src'], + 'ext': 'flv', + 'width': width, + 'height': height, + 'vbr': vbr, + }) + self._sort_formats(formats) return { 'id': video_id, @@ -77,4 +97,14 @@ class ThePlatformIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('id') - return self._get_info(video_id) + if mobj.group('config'): + config_url = url+ '&form=json' + config_url = config_url.replace('swf/', 'config/') + config_url = config_url.replace('onsite/', 'onsite/config/') + config_json = self._download_webpage(config_url, video_id, u'Downloading config') + config = json.loads(config_json) + smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m' + else: + smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?' + 'format=smil&mbr=true'.format(video_id)) + return self._get_info(video_id, smil_url)