X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/03342304420e5daeb428ffdcc7bbd2bbfecfa61a..9865587cf7b579793ba274067c3e44065d5ff77d:/youtube_dl/extractor/theplatform.py diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py index 93d8715..0705551 100644 --- a/youtube_dl/extractor/theplatform.py +++ b/youtube_dl/extractor/theplatform.py @@ -1,4 +1,4 @@ -# -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals import re @@ -8,7 +8,8 @@ import binascii import hashlib -from .common import InfoExtractor +from .once import OnceIE +from .adobepass import AdobePassIE from ..compat import ( compat_parse_qs, compat_urllib_parse_urlparse, @@ -20,72 +21,113 @@ from ..utils import ( int_or_none, sanitized_Request, unsmuggle_url, + update_url_query, xpath_with_ns, mimetype2ext, + find_xpath_attr, ) default_ns = 'http://www.w3.org/2005/SMIL21/Language' _x = lambda p: xpath_with_ns(p, {'smil': default_ns}) -class ThePlatformBaseIE(InfoExtractor): - def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'): - meta = self._download_xml(smil_url, video_id, note=note) - try: - error_msg = next( - n.attrib['abstract'] - for n in meta.findall(_x('.//smil:ref')) - if n.attrib.get('title') == 'Geographic Restriction' or n.attrib.get('title') == 'Expired') - except StopIteration: - pass - else: - raise ExtractorError(error_msg, expected=True) +class ThePlatformBaseIE(OnceIE): + _TP_TLD = 'com' - formats = self._parse_smil_formats( + def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'): + meta = self._download_xml( + smil_url, video_id, note=note, query={'format': 'SMIL'}, + headers=self.geo_verification_headers()) + error_element = find_xpath_attr(meta, _x('.//smil:ref'), 'src') + if error_element is not None: + exception = find_xpath_attr( + error_element, _x('.//smil:param'), 'name', 'exception') + if exception is not None: + if exception.get('value') == 'GeoLocationBlocked': + self.raise_geo_restricted(error_element.attrib['abstract']) + elif error_element.attrib['src'].startswith( + 'http://link.theplatform.%s/s/errorFiles/Unavailable.' + % self._TP_TLD): + raise ExtractorError( + error_element.attrib['abstract'], expected=True) + + smil_formats = self._parse_smil_formats( meta, smil_url, video_id, namespace=default_ns, # the parameters are from syfy.com, other sites may use others, # they also work for nbc.com f4m_params={'g': 'UXWGVKRWHFSP', 'hdcore': '3.0.3'}, transform_rtmp_url=lambda streamer, src: (streamer, 'mp4:' + src)) - for _format in formats: - ext = determine_ext(_format['url']) - if ext == 'once': - _format['ext'] = 'mp4' + formats = [] + for _format in smil_formats: + if OnceIE.suitable(_format['url']): + formats.extend(self._extract_once_formats(_format['url'])) + else: + media_url = _format['url'] + if determine_ext(media_url) == 'm3u8': + hdnea2 = self._get_cookies(media_url).get('hdnea2') + if hdnea2: + _format['url'] = update_url_query(media_url, {'hdnea3': hdnea2.value}) - self._sort_formats(formats) + formats.append(_format) subtitles = self._parse_smil_subtitles(meta, default_ns) return formats, subtitles - def get_metadata(self, path, video_id): - info_url = 'http://link.theplatform.com/s/%s?format=preview' % path - info = self._download_json(info_url, video_id) + def _download_theplatform_metadata(self, path, video_id): + info_url = 'http://link.theplatform.%s/s/%s?format=preview' % (self._TP_TLD, path) + return self._download_json(info_url, video_id) + def _parse_theplatform_metadata(self, info): 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] = [{ + subtitles.setdefault(lang, []).append({ 'ext': mimetype2ext(mime), 'url': src, - }] + }) + + duration = info.get('duration') + tp_chapters = info.get('chapters', []) + chapters = [] + if tp_chapters: + def _add_chapter(start_time, end_time): + start_time = float_or_none(start_time, 1000) + end_time = float_or_none(end_time, 1000) + if start_time is None or end_time is None: + return + chapters.append({ + 'start_time': start_time, + 'end_time': end_time, + }) + + for chapter in tp_chapters[:-1]: + _add_chapter(chapter.get('startTime'), chapter.get('endTime')) + _add_chapter(tp_chapters[-1].get('startTime'), tp_chapters[-1].get('endTime') or duration) return { 'title': info['title'], 'subtitles': subtitles, 'description': info['description'], 'thumbnail': info['defaultThumbnailUrl'], - 'duration': int_or_none(info.get('duration'), 1000), + 'duration': float_or_none(duration, 1000), + 'timestamp': int_or_none(info.get('pubDate'), 1000) or None, + 'uploader': info.get('billingCode'), + 'chapters': chapters, } + def _extract_theplatform_metadata(self, path, video_id): + info = self._download_theplatform_metadata(path, video_id) + return self._parse_theplatform_metadata(info) + -class ThePlatformIE(ThePlatformBaseIE): +class ThePlatformIE(ThePlatformBaseIE, AdobePassIE): _VALID_URL = r'''(?x) (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P[^/]+)/ - (?:(?P(?:(?:[^/]+/)+select/)?media/)|(?P(?:[^/\?]+/(?:swf|config)|onsite)/select/))? + (?:(?:(?:[^/]+/)+select/)?(?Pmedia/(?:guid/\d+/)?)?|(?P(?:[^/\?]+/(?:swf|config)|onsite)/select/))? |theplatform:)(?P[^/\?&]+)''' _TESTS = [{ @@ -97,11 +139,15 @@ class ThePlatformIE(ThePlatformBaseIE): 'title': 'Blackberry\'s big, bold Z30', 'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.', 'duration': 247, + 'timestamp': 1383239700, + 'upload_date': '20131031', + 'uploader': 'CBSI-NEW', }, 'params': { # rtmp download 'skip_download': True, }, + 'skip': '404 Not Found', }, { # 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', @@ -110,6 +156,9 @@ class ThePlatformIE(ThePlatformBaseIE): 'ext': 'flv', 'description': 'md5:ac330c9258c04f9d7512cf26b9595409', 'title': 'Tesla Model S: A second step towards a cleaner motoring future', + 'timestamp': 1426176191, + 'upload_date': '20150312', + 'uploader': 'CBSI-NEW', }, 'params': { # rtmp download @@ -122,23 +171,24 @@ class ThePlatformIE(ThePlatformBaseIE): 'ext': 'mp4', 'description': 'md5:644ad9188d655b742f942bf2e06b002d', 'title': 'HIGHLIGHTS: USA bag first ever series Cup win', + 'uploader': 'EGSM', } }, { 'url': 'http://player.theplatform.com/p/NnzsPC/widget/select/media/4Y0TlYUr_ZT7', 'only_matching': True, }, { 'url': 'http://player.theplatform.com/p/2E2eJC/nbcNewsOffsite?guid=tdy_or_siri_150701', - 'md5': '734f3790fb5fc4903da391beeebc4836', + 'md5': 'fb96bb3d85118930a5b055783a3bd992', 'info_dict': { 'id': 'tdy_or_siri_150701', 'ext': 'mp4', 'title': 'iPhone Siri’s sassy response to a math question has people talking', 'description': 'md5:a565d1deadd5086f3331d57298ec6333', 'duration': 83.0, - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'timestamp': 1435752600, 'upload_date': '20150701', - 'categories': ['Today/Shows/Orange Room', 'Today/Sections/Money', 'Today/Topics/Tech', "Today/Topics/Editor's picks"], + 'uploader': 'NBCU-NEWS', }, }, { # From http://www.nbc.com/the-blacklist/video/sir-crispin-crandall/2928790?onid=137781#vc137781=1 @@ -147,6 +197,24 @@ class ThePlatformIE(ThePlatformBaseIE): 'only_matching': True, }] + @classmethod + def _extract_urls(cls, webpage): + m = re.search( + r'''(?x) + https?://player\.theplatform\.com/p/.+?)\2 + ''', webpage) + if m: + return [m.group('url')] + + # Are whitesapces ignored in URLs? + # https://github.com/ytdl-org/youtube-dl/issues/12044 + matches = re.findall( + r'(?s)<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage) + if matches: + return [re.sub(r'\s', '', list(zip(*matches))[1][0])] + @staticmethod def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False): flags = '10' if include_qs else '00' @@ -155,11 +223,11 @@ class ThePlatformIE(ThePlatformBaseIE): def str_to_hex(str): return binascii.b2a_hex(str.encode('ascii')).decode('ascii') - def hex_to_str(hex): - return binascii.a2b_hex(hex) + def hex_to_bytes(hex): + return binascii.a2b_hex(hex.encode('ascii')) - 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)) + relative_path = re.match(r'https?://link\.theplatform\.com/s/([^?]+)', url).group(1) + clear_text = hex_to_bytes(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) @@ -174,10 +242,10 @@ class ThePlatformIE(ThePlatformBaseIE): if not provider_id: provider_id = 'dJ5BDC' - path = provider_id + path = provider_id + '/' if mobj.group('media'): - path += '/media' - path += '/' + video_id + path += mobj.group('media') + path += video_id qs_dict = compat_parse_qs(compat_urllib_parse_urlparse(url).query) if 'guid' in qs_dict: @@ -203,7 +271,7 @@ class ThePlatformIE(ThePlatformBaseIE): if smuggled_data.get('force_smil_url', False): smil_url = url - # Explicitly specified SMIL (see https://github.com/rg3/youtube-dl/issues/7385) + # Explicitly specified SMIL (see https://github.com/ytdl-org/youtube-dl/issues/7385) elif '/guid/' in url: headers = {} source_url = smuggled_data.get('source_url') @@ -216,7 +284,7 @@ class ThePlatformIE(ThePlatformBaseIE): webpage, 'smil url', group='url') path = self._search_regex( r'link\.theplatform\.com/s/((?:[^/?#&]+/)+[^/?#&]+)', smil_url, 'path') - smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4&format=SMIL' + smil_url += '?' if '?' not in smil_url else '&' + 'formats=m3u,mpeg4' elif mobj.group('config'): config_url = url + '&form=json' config_url = config_url.replace('swf/', 'config/') @@ -226,17 +294,18 @@ class ThePlatformIE(ThePlatformBaseIE): release_url = config['releaseUrl'] else: release_url = 'http://link.theplatform.com/s/%s?mbr=true' % path - smil_url = release_url + '&format=SMIL&formats=MPEG4&manifest=f4m' + smil_url = release_url + '&formats=MPEG4&manifest=f4m' else: - smil_url = 'http://link.theplatform.com/s/%s/meta.smil?format=smil&mbr=true' % path + smil_url = 'http://link.theplatform.com/s/%s?mbr=true' % path sig = smuggled_data.get('sig') if sig: smil_url = self._sign_url(smil_url, sig['key'], sig['secret']) formats, subtitles = self._extract_theplatform_smil(smil_url, video_id) + self._sort_formats(formats) - ret = self.get_metadata(path, video_id) + ret = self._extract_theplatform_metadata(path, video_id) combined_subtitles = self._merge_subtitles(ret.get('subtitles', {}), subtitles) ret.update({ 'id': video_id, @@ -248,49 +317,61 @@ class ThePlatformIE(ThePlatformBaseIE): class ThePlatformFeedIE(ThePlatformBaseIE): - _URL_TEMPLATE = '%s//feed.theplatform.com/f/%s/%s?form=json&byGuid=%s' - _VALID_URL = r'https?://feed\.theplatform\.com/f/(?P[^/]+)/(?P[^?/]+)\?(?:[^&]+&)*byGuid=(?P[a-zA-Z0-9_]+)' - _TEST = { + _URL_TEMPLATE = '%s//feed.theplatform.com/f/%s/%s?form=json&%s' + _VALID_URL = r'https?://feed\.theplatform\.com/f/(?P[^/]+)/(?P[^?/]+)\?(?:[^&]+&)*(?Pby(?:Gui|I)d=(?P[^&]+))' + _TESTS = [{ # From http://player.theplatform.com/p/7wvmTC/MSNBCEmbeddedOffSite?guid=n_hardball_5biden_140207 'url': 'http://feed.theplatform.com/f/7wvmTC/msnbc_video-p-test?form=json&pretty=true&range=-40&byGuid=n_hardball_5biden_140207', - 'md5': '22d2b84f058d3586efcd99e57d59d314', + 'md5': '6e32495b5073ab414471b615c5ded394', 'info_dict': { 'id': 'n_hardball_5biden_140207', 'ext': 'mp4', 'title': 'The Biden factor: will Joe run in 2016?', 'description': 'Could Vice President Joe Biden be preparing a 2016 campaign? Mark Halperin and Sam Stein weigh in.', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'upload_date': '20140208', 'timestamp': 1391824260, 'duration': 467.0, 'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'], + 'uploader': 'NBCU-NEWS', }, - } - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - - video_id = mobj.group('id') - provider_id = mobj.group('provider_id') - feed_id = mobj.group('feed_id') + }, { + 'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byGuid=nn_netcast_180306.Copy.01', + 'only_matching': True, + }] - real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, video_id) - feed = self._download_json(real_url, video_id) - entry = feed['entries'][0] + def _extract_feed_info(self, provider_id, feed_id, filter_query, video_id, custom_fields=None, asset_types_query={}, account_id=None): + real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, filter_query) + entry = self._download_json(real_url, video_id)['entries'][0] + main_smil_url = 'http://link.theplatform.com/s/%s/media/guid/%d/%s' % (provider_id, account_id, entry['guid']) if account_id else entry.get('plmedia$publicUrl') formats = [] subtitles = {} first_video_id = None duration = None + asset_types = [] for item in entry['media$content']: - smil_url = item['plfile$url'] + '&format=SMIL&mbr=true' + smil_url = item['plfile$url'] cur_video_id = ThePlatformIE._match_id(smil_url) if first_video_id is None: first_video_id = cur_video_id duration = float_or_none(item.get('plfile$duration')) - cur_formats, cur_subtitles = self._extract_theplatform_smil(smil_url, video_id, 'Downloading SMIL data for %s' % cur_video_id) - formats.extend(cur_formats) - subtitles = self._merge_subtitles(subtitles, cur_subtitles) + file_asset_types = item.get('plfile$assetTypes') or compat_parse_qs(compat_urllib_parse_urlparse(smil_url).query)['assetTypes'] + for asset_type in file_asset_types: + if asset_type in asset_types: + continue + asset_types.append(asset_type) + query = { + 'mbr': 'true', + 'formats': item['plfile$format'], + 'assetTypes': asset_type, + } + if asset_type in asset_types_query: + query.update(asset_types_query[asset_type]) + cur_formats, cur_subtitles = self._extract_theplatform_smil(update_url_query( + main_smil_url or smil_url, query), video_id, 'Downloading SMIL data for %s' % asset_type) + formats.extend(cur_formats) + subtitles = self._merge_subtitles(subtitles, cur_subtitles) self._sort_formats(formats) @@ -303,7 +384,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE): timestamp = int_or_none(entry.get('media$availableDate'), scale=1000) categories = [item['media$name'] for item in entry.get('media$categories', [])] - ret = self.get_metadata('%s/%s' % (provider_id, first_video_id), video_id) + ret = self._extract_theplatform_metadata('%s/%s' % (provider_id, first_video_id), video_id) subtitles = self._merge_subtitles(subtitles, ret['subtitles']) ret.update({ 'id': video_id, @@ -314,5 +395,17 @@ class ThePlatformFeedIE(ThePlatformBaseIE): 'timestamp': timestamp, 'categories': categories, }) + if custom_fields: + ret.update(custom_fields(entry)) return ret + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + + video_id = mobj.group('id') + provider_id = mobj.group('provider_id') + feed_id = mobj.group('feed_id') + filter_query = mobj.group('filter') + + return self._extract_feed_info(provider_id, feed_id, filter_query, video_id)