X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Ftheplatform.py;h=1579822f2c8611d6b9e8e464624321009d897251;hb=c84dd8a90dcc75547b343449b921b644a2119c4f;hp=af6ef0033af061713d81785734381a6beffc2c6d;hpb=1cc79574fc5df21bf35dccf61eac0e9e75ed8d20;p=youtube-dl diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index af6ef0033..1579822f2 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -2,8 +2,13 @@ from __future__ import unicode_literals import re import json +import time +import hmac +import binascii +import hashlib -from .common import InfoExtractor + +from .subtitles import SubtitlesInfoExtractor from ..compat import ( compat_str, ) @@ -11,14 +16,15 @@ from ..utils import ( determine_ext, ExtractorError, xpath_with_ns, + unsmuggle_url, ) _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'}) -class ThePlatformIE(InfoExtractor): +class ThePlatformIE(SubtitlesInfoExtractor): _VALID_URL = r'''(?x) - (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/ + (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P[^/]+)/ (?P(?:[^/\?]+/(?:swf|config)|onsite)/select/)? |theplatform:)(?P[^/\?&]+)''' @@ -38,9 +44,33 @@ class ThePlatformIE(InfoExtractor): }, } + @staticmethod + def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False): + flags = '10' if include_qs else '00' + expiration_date = '%x' % (int(time.time()) + life) + + def str_to_hex(str): + return binascii.b2a_hex(str.encode('ascii')).decode('ascii') + + def hex_to_str(hex): + return binascii.a2b_hex(hex) + + relative_path = url.split('http://link.theplatform.com/s/')[1].split('?')[0] + clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path)) + checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest() + sig = flags + expiration_date + checksum + str_to_hex(sig_secret) + return '%s&sig=%s' % (url, sig) + def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + mobj = re.match(self._VALID_URL, url) + provider_id = mobj.group('provider_id') video_id = mobj.group('id') + + if not provider_id: + provider_id = 'dJ5BDC' + if mobj.group('config'): config_url = url + '&form=json' config_url = config_url.replace('swf/', 'config/') @@ -48,8 +78,12 @@ class ThePlatformIE(InfoExtractor): 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)) + smil_url = ('http://link.theplatform.com/s/{0}/{1}/meta.smil?' + 'format=smil&mbr=true'.format(provider_id, video_id)) + + sig = smuggled_data.get('sig') + if sig: + smil_url = self._sign_url(smil_url, sig['key'], sig['secret']) meta = self._download_xml(smil_url, video_id) try: @@ -62,10 +96,24 @@ class ThePlatformIE(InfoExtractor): else: raise ExtractorError(error_msg, expected=True) - info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id) + info_url = 'http://link.theplatform.com/s/{0}/{1}?format=preview'.format(provider_id, video_id) info_json = self._download_webpage(info_url, video_id) info = json.loads(info_json) + subtitles = {} + captions = info.get('captions') + if isinstance(captions, list): + for caption in captions: + lang, src = caption.get('lang'), caption.get('src') + if lang and src: + subtitles[lang] = src + + if self._downloader.params.get('listsubtitles', False): + self._list_available_subtitles(video_id, subtitles) + return + + subtitles = self.extract_subtitles(video_id, subtitles) + head = meta.find(_x('smil:head')) body = meta.find(_x('smil:body')) @@ -117,6 +165,7 @@ class ThePlatformIE(InfoExtractor): return { 'id': video_id, 'title': info['title'], + 'subtitles': subtitles, 'formats': formats, 'description': info['description'], 'thumbnail': info['defaultThumbnailUrl'],