From 10e3d734723def2d91bdd3d02b85a149c056dcfb Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Mon, 27 Oct 2014 01:14:17 +0100 Subject: [PATCH] [nbc] Fix ThePlatform embedded videos --- youtube_dl/extractor/nbc.py | 7 +++--- youtube_dl/extractor/theplatform.py | 34 +++++++++++++++-------------- 2 files changed, 22 insertions(+), 19 deletions(-) diff --git a/youtube_dl/extractor/nbc.py b/youtube_dl/extractor/nbc.py index e75ab7c39..7b5449031 100644 --- a/youtube_dl/extractor/nbc.py +++ b/youtube_dl/extractor/nbc.py @@ -26,8 +26,7 @@ class NBCIE(InfoExtractor): } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) theplatform_url = self._search_regex('class="video-player video-player-full" data-mpx-url="(.*?)"', webpage, 'theplatform url') if theplatform_url.startswith('//'): @@ -57,7 +56,7 @@ class NBCNewsIE(InfoExtractor): 'md5': 'b2421750c9f260783721d898f4c42063', 'info_dict': { 'id': 'I1wpAI_zmhsQ', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'How Twitter Reacted To The Snowden Interview', 'description': 'md5:65a0bd5d76fe114f3c2727aa3a81fe64', }, @@ -97,6 +96,8 @@ def _real_extract(self, url): ] for base_url in base_urls: + if not base_url: + continue playlist_url = base_url + '?form=MPXNBCNewsAPI' all_videos = self._download_json(playlist_url, title)['videos'] diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 0be793b1c..a04925633 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -6,6 +6,7 @@ from .common import InfoExtractor from ..utils import ( compat_str, + determine_ext, ExtractorError, xpath_with_ns, ) @@ -34,10 +35,21 @@ class ThePlatformIE(InfoExtractor): 'skip_download': True, }, } + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('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 = self._download_json(config_url, video_id, 'Downloading config') + 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)) + - def _get_info(self, video_id, smil_url): meta = self._download_xml(smil_url, video_id) - try: error_msg = next( n.attrib['abstract'] @@ -89,10 +101,14 @@ def _get_info(self, video_id, smil_url): for f in switch.findall(_x('smil:video')): attr = f.attrib vbr = int(attr['system-bitrate']) // 1000 + ext = determine_ext(attr['src']) + if ext == 'once': + ext = 'mp4' formats.append({ 'format_id': compat_str(vbr), 'url': attr['src'], 'vbr': vbr, + 'ext': ext, }) self._sort_formats(formats) @@ -104,17 +120,3 @@ def _get_info(self, video_id, smil_url): 'thumbnail': info['defaultThumbnailUrl'], 'duration': info['duration']//1000, } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('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 = self._download_json(config_url, video_id, 'Downloading config') - 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)