From a97bcd80bac769707b7a9138418794aaf883b34d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Mon, 23 Dec 2013 16:41:35 +0100 Subject: [PATCH] Add an extractor for syfy.com It uses theplatfrom.com, which has been updated to work with f4m manifests --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/syfy.py | 30 +++++++++++++ youtube_dl/extractor/theplatform.py | 66 ++++++++++++++++++----------- 3 files changed, 73 insertions(+), 24 deletions(-) create mode 100644 youtube_dl/extractor/syfy.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 4678cf06c..97b53900c 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -203,6 +203,7 @@ from .stanfordoc import StanfordOpenClassroomIE from .statigram import StatigramIE from .steam import SteamIE from .streamcloud import StreamcloudIE +from .syfy import SyfyIE from .sztvhu import SztvHuIE from .teamcoco import TeamcocoIE from .techtalks import TechTalksIE diff --git a/youtube_dl/extractor/syfy.py b/youtube_dl/extractor/syfy.py new file mode 100644 index 000000000..502d43ec4 --- /dev/null +++ b/youtube_dl/extractor/syfy.py @@ -0,0 +1,30 @@ +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor + + +class SyfyIE(InfoExtractor): + _VALID_URL = r'https?://www\.syfy\.com/videos/.+?vid:(?P\d+)' + + _TEST = { + 'url': 'http://www.syfy.com/videos/Robot%20Combat%20League/Behind%20the%20Scenes/vid:2631458', + 'info_dict': { + 'id': 'NmqMrGnXvmO1', + 'ext': 'flv', + 'title': 'George Lucas has Advice for his Daughter', + 'description': 'Listen to what insights George Lucas give his daughter Amanda.', + }, + 'params': { + # f4m download + 'skip_download': True, + }, + 'add_ie': ['ThePlatform'], + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + webpage = self._download_webpage(url, video_id) + return self.url_result(self._og_search_video_url(webpage)) diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 23172143e..d60702325 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)/select/)? + |theplatform:)(?P[^/\?&]+)''' _TEST = { # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/ @@ -29,9 +32,7 @@ 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: @@ -50,26 +51,34 @@ 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 - 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) + f4m_node = body.find(_x('smil:seq/smil:video')) + if f4m_node is not None: + formats = [{ + 'ext': 'flv', + # the parameters are from syfy.com, other sites may use others + 'url': f4m_node.attrib['src'] + '?g=UXWGVKRWHFSP&hdcore=3.0.3', + }] + 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, @@ -83,4 +92,13 @@ 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_json = self._download_webpage(config_url, video_id, u'Downloading config') + config = json.loads(config_json) + smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4' + 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)