X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/0cf0312991a54458a07e903da2e47e9f3c8855ae..3fb6b1d5630c7291c5e2c0ff40b2a6b3b63ee5d9:/youtube_dl/extractor/theplatform.py diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index e2653d6..92731ad 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -2,13 +2,22 @@ from __future__ import unicode_literals import re import json +import time +import hmac +import binascii +import hashlib + from .common import InfoExtractor -from ..utils import ( +from ..compat import ( compat_str, +) +from ..utils import ( determine_ext, ExtractorError, xpath_with_ns, + unsmuggle_url, + int_or_none, ) _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'}) @@ -16,11 +25,11 @@ _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language class ThePlatformIE(InfoExtractor): _VALID_URL = r'''(?x) - (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/ + (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P[^/]+)/ (?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': { @@ -34,40 +43,95 @@ 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): + 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 mobj.group('config'): + + if not provider_id: + provider_id = 'dJ5BDC' + + if smuggled_data.get('force_smil_url', False): + smil_url = url + elif 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)) + 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: 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: 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, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type') + subtitles[lang] = [{ + 'ext': 'srt' if mime == 'text/srt' else 'ttml', + 'url': src, + }] + head = meta.find(_x('smil:head')) 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: @@ -79,13 +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, @@ -98,9 +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' @@ -115,8 +187,9 @@ class ThePlatformIE(InfoExtractor): return { 'id': video_id, 'title': info['title'], + 'subtitles': subtitles, 'formats': formats, 'description': info['description'], 'thumbnail': info['defaultThumbnailUrl'], - 'duration': info['duration'] // 1000, + 'duration': int_or_none(info.get('duration'), 1000), }