X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/46113edab215c2211a604c06245c16d5d4e57dcf..47d80ec0b18245caeb97018d4c1af18d0b5b972b:/youtube_dl/extractor/npo.py?ds=sidebyside diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py index f36d446..c91f584 100644 --- a/youtube_dl/extractor/npo.py +++ b/youtube_dl/extractor/npo.py @@ -3,17 +3,59 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_HTTPError from ..utils import ( - unified_strdate, + fix_xml_ampersands, + orderedSet, parse_duration, qualities, - url_basename, + strip_jsonp, + unified_strdate, + ExtractorError, ) -class NPOIE(InfoExtractor): - IE_NAME = 'npo.nl' - _VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?P[^/?]+)' +class NPOBaseIE(InfoExtractor): + def _get_token(self, video_id): + token_page = self._download_webpage( + 'http://ida.omroep.nl/npoplayer/i.js', + video_id, note='Downloading token') + token = self._search_regex( + r'npoplayer\.token = "(.+?)"', token_page, 'token') + # Decryption algorithm extracted from http://npoplayer.omroep.nl/csjs/npoplayer-min.js + token_l = list(token) + first = second = None + for i in range(5, len(token_l) - 4): + if token_l[i].isdigit(): + if first is None: + first = i + elif second is None: + second = i + if first is None or second is None: + first = 12 + second = 13 + + token_l[first], token_l[second] = token_l[second], token_l[first] + + return ''.join(token_l) + + +class NPOIE(NPOBaseIE): + IE_NAME = 'npo' + IE_DESC = 'npo.nl and ntr.nl' + _VALID_URL = r'''(?x) + (?: + npo:| + https?:// + (?:www\.)? + (?: + npo\.nl/(?!live|radio)(?:[^/]+/){2}| + ntr\.nl/(?:[^/]+/){2,}| + omroepwnl\.nl/video/fragment/[^/]+__ + ) + ) + (?P[^/?#]+) + ''' _TESTS = [ { @@ -33,7 +75,7 @@ class NPOIE(InfoExtractor): 'info_dict': { 'id': 'VARA_101191800', 'ext': 'm4v', - 'title': 'De Mega Mike & Mega Thomas show', + 'title': 'De Mega Mike & Mega Thomas show: The best of.', 'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4', 'upload_date': '20090227', 'duration': 2400, @@ -45,75 +87,391 @@ class NPOIE(InfoExtractor): 'info_dict': { 'id': 'VPWON_1169289', 'ext': 'm4v', - 'title': 'Tegenlicht', - 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1', + 'title': 'Tegenlicht: De toekomst komt uit Afrika', + 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea', 'upload_date': '20130225', 'duration': 3000, }, + }, + { + 'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706', + 'info_dict': { + 'id': 'WO_VPRO_043706', + 'ext': 'wmv', + 'title': 'De nieuwe mens - Deel 1', + 'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b', + 'duration': 4680, + }, + 'params': { + # mplayer mms download + 'skip_download': True, + } + }, + # non asf in streams + { + 'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771', + 'md5': 'b3da13de374cbe2d5332a7e910bef97f', + 'info_dict': { + 'id': 'WO_NOS_762771', + 'ext': 'mp4', + 'title': 'Hoe gaat Europa verder na Parijs?', + }, + }, + { + 'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content', + 'md5': '01c6a2841675995da1f0cf776f03a9c3', + 'info_dict': { + 'id': 'VPWON_1233944', + 'ext': 'm4v', + 'title': 'Aap, poot, pies', + 'description': 'md5:c9c8005d1869ae65b858e82c01a91fde', + 'upload_date': '20150508', + 'duration': 599, + }, + }, + { + 'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698', + 'md5': 'd30cd8417b8b9bca1fdff27428860d08', + 'info_dict': { + 'id': 'POW_00996502', + 'ext': 'm4v', + 'title': '''"Dit is wel een 'landslide'..."''', + 'description': 'md5:f8d66d537dfb641380226e31ca57b8e8', + 'upload_date': '20150508', + 'duration': 462, + }, } ] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) return self._get_info(video_id) def _get_info(self, video_id): metadata = self._download_json( - 'http://e.omroep.nl/metadata/aflevering/%s' % video_id, + 'http://e.omroep.nl/metadata/%s' % video_id, video_id, # We have to remove the javascript callback - transform_source=lambda j: re.sub(r'parseMetadata\((.*?)\);\n//.*$', r'\1', j) - ) - token_page = self._download_webpage( - 'http://ida.omroep.nl/npoplayer/i.js', - video_id, - note='Downloading token' + transform_source=strip_jsonp, ) - token = self._search_regex(r'npoplayer\.token = "(.+?)"', token_page, 'token') + + # For some videos actual video id (prid) is different (e.g. for + # http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698 + # video id is POMS_WNL_853698 but prid is POW_00996502) + video_id = metadata.get('prid') or video_id + + # titel is too generic in some cases so utilize aflevering_titel as well + # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html) + title = metadata['titel'] + sub_title = metadata.get('aflevering_titel') + if sub_title and sub_title != title: + title += ': %s' % sub_title + + token = self._get_token(video_id) formats = [] - quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std']) - for format_id in metadata['pubopties']: - format_info = self._download_json( - 'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s' % (video_id, format_id, token), - video_id, 'Downloading %s JSON' % format_id) - if format_info.get('error_code', 0) or format_info.get('errorcode', 0): - continue - streams = format_info.get('streams') - if streams: - video_info = self._download_json( - streams[0] + '&type=json', - video_id, 'Downloading %s stream JSON' % format_id) - else: - video_info = format_info - video_url = video_info.get('url') - if not video_url: - continue - if format_id == 'adaptive': - formats.extend(self._extract_m3u8_formats(video_url, video_id)) - else: + + pubopties = metadata.get('pubopties') + if pubopties: + quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std']) + for format_id in pubopties: + format_info = self._download_json( + 'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s' + % (video_id, format_id, token), + video_id, 'Downloading %s JSON' % format_id) + if format_info.get('error_code', 0) or format_info.get('errorcode', 0): + continue + streams = format_info.get('streams') + if streams: + try: + video_info = self._download_json( + streams[0] + '&type=json', + video_id, 'Downloading %s stream JSON' % format_id) + except ExtractorError as ee: + if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 404: + error = (self._parse_json(ee.cause.read().decode(), video_id, fatal=False) or {}).get('errorstring') + if error: + raise ExtractorError(error, expected=True) + raise + else: + video_info = format_info + video_url = video_info.get('url') + if not video_url: + continue + if format_id == 'adaptive': + formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4')) + else: + formats.append({ + 'url': video_url, + 'format_id': format_id, + 'quality': quality(format_id), + }) + + streams = metadata.get('streams') + if streams: + for i, stream in enumerate(streams): + stream_url = stream.get('url') + if not stream_url: + continue + if '.asf' not in stream_url: + formats.append({ + 'url': stream_url, + 'quality': stream.get('kwaliteit'), + }) + continue + asx = self._download_xml( + stream_url, video_id, + 'Downloading stream %d ASX playlist' % i, + transform_source=fix_xml_ampersands) + ref = asx.find('./ENTRY/Ref') + if ref is None: + continue + video_url = ref.get('href') + if not video_url: + continue formats.append({ 'url': video_url, - 'format_id': format_id, - 'quality': quality(format_id), + 'ext': stream.get('formaat', 'asf'), + 'quality': stream.get('kwaliteit'), }) + self._sort_formats(formats) + subtitles = {} + if metadata.get('tt888') == 'ja': + subtitles['nl'] = [{ + 'ext': 'vtt', + 'url': 'http://e.omroep.nl/tt888/%s' % video_id, + }] + return { 'id': video_id, - 'title': metadata['titel'], - 'description': metadata['info'], + 'title': title, + 'description': metadata.get('info'), 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'], 'upload_date': unified_strdate(metadata.get('gidsdatum')), 'duration': parse_duration(metadata.get('tijdsduur')), 'formats': formats, + 'subtitles': subtitles, + } + + +class NPOLiveIE(NPOBaseIE): + IE_NAME = 'npo.nl:live' + _VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P.+)' + + _TEST = { + 'url': 'http://www.npo.nl/live/npo-1', + 'info_dict': { + 'id': 'LI_NEDERLAND1_136692', + 'display_id': 'npo-1', + 'ext': 'mp4', + 'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'description': 'Livestream', + 'is_live': True, + }, + 'params': { + 'skip_download': True, + } + } + + def _real_extract(self, url): + display_id = self._match_id(url) + + webpage = self._download_webpage(url, display_id) + + live_id = self._search_regex( + r'data-prid="([^"]+)"', webpage, 'live id') + + metadata = self._download_json( + 'http://e.omroep.nl/metadata/%s' % live_id, + display_id, transform_source=strip_jsonp) + + token = self._get_token(display_id) + + formats = [] + + streams = metadata.get('streams') + if streams: + for stream in streams: + stream_type = stream.get('type').lower() + # smooth streaming is not supported + if stream_type in ['ss', 'ms']: + continue + stream_info = self._download_json( + 'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp' + % (stream.get('url'), token), + display_id, 'Downloading %s JSON' % stream_type) + if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0): + continue + stream_url = self._download_json( + stream_info['stream'], display_id, + 'Downloading %s URL' % stream_type, + 'Unable to download %s URL' % stream_type, + transform_source=strip_jsonp, fatal=False) + if not stream_url: + continue + if stream_type == 'hds': + f4m_formats = self._extract_f4m_formats(stream_url, display_id) + # f4m downloader downloads only piece of live stream + for f4m_format in f4m_formats: + f4m_format['preference'] = -1 + formats.extend(f4m_formats) + elif stream_type == 'hls': + formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4')) + else: + formats.append({ + 'url': stream_url, + 'preference': -10, + }) + + self._sort_formats(formats) + + return { + 'id': live_id, + 'display_id': display_id, + 'title': self._live_title(metadata['titel']), + 'description': metadata['info'], + 'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'], + 'formats': formats, + 'is_live': True, } -class TegenlichtVproIE(NPOIE): - IE_NAME = 'tegenlicht.vpro.nl' - _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?' +class NPORadioIE(InfoExtractor): + IE_NAME = 'npo.nl:radio' + _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P[^/]+)/?$' + + _TEST = { + 'url': 'http://www.npo.nl/radio/radio-1', + 'info_dict': { + 'id': 'radio-1', + 'ext': 'mp3', + 'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'is_live': True, + }, + 'params': { + 'skip_download': True, + } + } + + @staticmethod + def _html_get_attribute_regex(attribute): + return r'{0}\s*=\s*\'([^\']+)\''.format(attribute) + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + title = self._html_search_regex( + self._html_get_attribute_regex('data-channel'), webpage, 'title') + + stream = self._parse_json( + self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'), + video_id) + + codec = stream.get('codec') + + return { + 'id': video_id, + 'url': stream['url'], + 'title': self._live_title(title), + 'acodec': codec, + 'ext': codec, + 'is_live': True, + } + + +class NPORadioFragmentIE(InfoExtractor): + IE_NAME = 'npo.nl:radio:fragment' + _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P\d+)' + + _TEST = { + 'url': 'http://www.npo.nl/radio/radio-5/fragment/174356', + 'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2', + 'info_dict': { + 'id': '174356', + 'ext': 'mp3', + 'title': 'Jubileumconcert Willeke Alberti', + }, + } + + def _real_extract(self, url): + audio_id = self._match_id(url) + + webpage = self._download_webpage(url, audio_id) + + title = self._html_search_regex( + r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id, + webpage, 'title') + + audio_url = self._search_regex( + r"data-streams='([^']+)'", webpage, 'audio url') + + return { + 'id': audio_id, + 'url': audio_url, + 'title': title, + } + + +class SchoolTVIE(InfoExtractor): + IE_NAME = 'schooltv' + _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P[^/?#&]+)' + + _TEST = { + 'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/', + 'info_dict': { + 'id': 'WO_NTR_429477', + 'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam', + 'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?', + 'ext': 'mp4', + 'description': 'md5:abfa0ff690adb73fd0297fd033aaa631' + }, + 'params': { + # Skip because of m3u8 download + 'skip_download': True + } + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + video_id = self._search_regex( + r'data-mid=(["\'])(?P(?:(?!\1).)+)\1', webpage, 'video_id', group='id') + return { + '_type': 'url_transparent', + 'ie_key': 'NPO', + 'url': 'npo:%s' % video_id, + 'display_id': display_id + } + + +class NPOPlaylistBaseIE(NPOIE): + def _real_extract(self, url): + playlist_id = self._match_id(url) + + webpage = self._download_webpage(url, playlist_id) + + entries = [ + self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id) + for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage)) + ] + + playlist_title = self._html_search_regex( + self._PLAYLIST_TITLE_RE, webpage, 'playlist title', + default=None) or self._og_search_title(webpage) + + return self.playlist_result(entries, playlist_id, playlist_title) + + +class VPROIE(NPOPlaylistBaseIE): + IE_NAME = 'vpro' + _VALID_URL = r'https?://(?:www\.)?(?:(?:tegenlicht\.)?vpro|2doc)\.nl/(?:[^/]+/)*(?P[^/]+)\.html' + _PLAYLIST_TITLE_RE = (r']+class=["\'].*?\bmedia-platform-title\b.*?["\'][^>]*>([^<]+)', + r']+class=["\'].*?\bmedia-platform-subtitle\b.*?["\'][^>]*>([^<]+)') + _PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"' _TESTS = [ { @@ -122,17 +480,80 @@ class TegenlichtVproIE(NPOIE): 'info_dict': { 'id': 'VPWON_1169289', 'ext': 'm4v', - 'title': 'Tegenlicht', - 'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1', + 'title': 'De toekomst komt uit Afrika', + 'description': 'md5:52cf4eefbc96fffcbdc06d024147abea', 'upload_date': '20130225', }, + 'skip': 'Video gone', + }, + { + 'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html', + 'info_dict': { + 'id': 'sergio-herman', + 'title': 'sergio herman: fucking perfect', + }, + 'playlist_count': 2, + }, + { + # playlist with youtube embed + 'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html', + 'info_dict': { + 'id': 'education-education', + 'title': 'education education', + }, + 'playlist_count': 2, }, + { + 'url': 'http://www.2doc.nl/documentaires/series/2doc/2015/oktober/de-tegenprestatie.html', + 'info_dict': { + 'id': 'de-tegenprestatie', + 'title': 'De Tegenprestatie', + }, + 'playlist_count': 2, + }, { + 'url': 'http://www.2doc.nl/speel~VARA_101375237~mh17-het-verdriet-van-nederland~.html', + 'info_dict': { + 'id': 'VARA_101375237', + 'ext': 'm4v', + 'title': 'MH17: Het verdriet van Nederland', + 'description': 'md5:09e1a37c1fdb144621e22479691a9f18', + 'upload_date': '20150716', + }, + 'params': { + # Skip because of m3u8 download + 'skip_download': True + }, + } ] - def _real_extract(self, url): - name = url_basename(url) - webpage = self._download_webpage(url, name) - urn = self._html_search_meta('mediaurn', webpage) - info_page = self._download_json( - 'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name) - return self._get_info(info_page['mid']) + +class WNLIE(NPOPlaylistBaseIE): + IE_NAME = 'wnl' + _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P[^/]+)__\d+' + _PLAYLIST_TITLE_RE = r'(?s)]+class="subject"[^>]*>(.+?)' + _PLAYLIST_ENTRY_RE = r']+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+' + + _TESTS = [{ + 'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515', + 'info_dict': { + 'id': 'vandaag-de-dag-6-mei', + 'title': 'Vandaag de Dag 6 mei', + }, + 'playlist_count': 4, + }] + + +class AndereTijdenIE(NPOPlaylistBaseIE): + IE_NAME = 'anderetijden' + _VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P[^/?#&]+)' + _PLAYLIST_TITLE_RE = r'(?s)]+class=["\'].*?\bpage-title\b.*?["\'][^>]*>(.+?)' + _PLAYLIST_ENTRY_RE = r']+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']' + + _TESTS = [{ + 'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem', + 'info_dict': { + 'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem', + 'title': 'Duitse soldaten over de Slag bij Arnhem', + }, + 'playlist_count': 3, + }]