X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/f46044c66663049e286c20ee015db99d47d9dd8a..7df4343e553a9942a290e5e1dffe931261ab844e:/youtube_dl/extractor/theplatform.py diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index feac666..92731ad 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -17,6 +17,7 @@ from ..utils import ( ExtractorError, xpath_with_ns, unsmuggle_url, + int_or_none, ) _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'}) @@ -28,7 +29,7 @@ class ThePlatformIE(InfoExtractor): (?P(?:[^/\?]+/(?:swf|config)|onsite)/select/)? |theplatform:)(?P[^/\?&]+)''' - _TEST = { + _TESTS = [{ # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/ 'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true', 'info_dict': { @@ -42,7 +43,20 @@ class ThePlatformIE(InfoExtractor): # rtmp download 'skip_download': True, }, - } + }, { + # from http://www.cnet.com/videos/tesla-model-s-a-second-step-towards-a-cleaner-motoring-future/ + 'url': 'http://link.theplatform.com/s/kYEXFC/22d_qsQ6MIRT', + 'info_dict': { + 'id': '22d_qsQ6MIRT', + 'ext': 'flv', + 'description': 'md5:ac330c9258c04f9d7512cf26b9595409', + 'title': 'Tesla Model S: A second step towards a cleaner motoring future', + }, + 'params': { + # rtmp download + 'skip_download': True, + } + }] @staticmethod def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False): @@ -92,7 +106,7 @@ class ThePlatformIE(InfoExtractor): error_msg = next( n.attrib['abstract'] for n in meta.findall(_x('.//smil:ref')) - if n.attrib.get('title') == 'Geographic Restriction') + if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired') except StopIteration: pass else: @@ -116,6 +130,8 @@ class ThePlatformIE(InfoExtractor): body = meta.find(_x('smil:body')) f4m_node = body.find(_x('smil:seq//smil:video')) + if f4m_node is None: + f4m_node = body.find(_x('smil:seq/smil:video')) if f4m_node is not None and '.f4m' in f4m_node.attrib['src']: f4m_url = f4m_node.attrib['src'] if 'manifest.f4m?' not in f4m_url: @@ -127,13 +143,19 @@ class ThePlatformIE(InfoExtractor): else: formats = [] switch = body.find(_x('smil:switch')) + if switch is None: + switch = body.find(_x('smil:par//smil:switch')) + if switch is None: + switch = body.find(_x('smil:par/smil:switch')) + if switch is None: + switch = body.find(_x('smil:par')) if switch is not None: base_url = head.find(_x('smil:meta')).attrib['base'] 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 + width = int_or_none(attr.get('width')) + height = int_or_none(attr.get('height')) + vbr = int_or_none(attr.get('system-bitrate'), 1000) format_id = '%dx%d_%dk' % (width, height, vbr) formats.append({ 'format_id': format_id, @@ -146,9 +168,11 @@ class ThePlatformIE(InfoExtractor): }) else: switch = body.find(_x('smil:seq//smil:switch')) + if switch is None: + switch = body.find(_x('smil:seq/smil:switch')) for f in switch.findall(_x('smil:video')): attr = f.attrib - vbr = int(attr['system-bitrate']) // 1000 + vbr = int_or_none(attr.get('system-bitrate'), 1000) ext = determine_ext(attr['src']) if ext == 'once': ext = 'mp4' @@ -167,5 +191,5 @@ class ThePlatformIE(InfoExtractor): 'formats': formats, 'description': info['description'], 'thumbnail': info['defaultThumbnailUrl'], - 'duration': info['duration'] // 1000, + 'duration': int_or_none(info.get('duration'), 1000), }