[yt-dlp.git] / youtube_dl / extractor / npo.py

from __future__ import unicode_literals

from .common import InfoExtractor
from ..utils import (
    fix_xml_ampersands,
    parse_duration,
    qualities,
    strip_jsonp,
    unified_strdate,
    url_basename,
)


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'https?://(?:www\.)?(?:npo|ntr)\.nl/(?!live|radio)(?:[^/]+/){2,}(?P<id>[^/?#]+)'

    _TESTS = [
        {
            'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
            'md5': '4b3f9c429157ec4775f2c9cb7b911016',
            'info_dict': {
                'id': 'VPWON_1220719',
                'ext': 'm4v',
                'title': 'Nieuwsuur',
                'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
                'upload_date': '20140622',
            },
        },
        {
            'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
            'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
            'info_dict': {
                'id': 'VARA_101191800',
                'ext': 'm4v',
                'title': 'De Mega Mike & Mega Thomas show',
                'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
                'upload_date': '20090227',
                'duration': 2400,
            },
        },
        {
            'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
            'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
            'info_dict': {
                'id': 'VPWON_1169289',
                'ext': 'm4v',
                'title': 'Tegenlicht',
                'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
                '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,
            },
        }
    ]

    def _real_extract(self, url):
        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/%s' % video_id,
            video_id,
            # We have to remove the javascript callback
            transform_source=strip_jsonp,
        )

        token = self._get_token(video_id)

        formats = []

        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:
                    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:
                    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,
                    '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'],
            '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<id>.+)'

    _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 NPORadioIE(InfoExtractor):
    IE_NAME = 'npo.nl:radio'
    _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'

    _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<id>\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 TegenlichtVproIE(NPOIE):
    IE_NAME = 'tegenlicht.vpro.nl'
    _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'

    _TESTS = [
        {
            'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
            'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
            'info_dict': {
                'id': 'VPWON_1169289',
                'ext': 'm4v',
                'title': 'Tegenlicht',
                'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
                'upload_date': '20130225',
            },
        },
    ]

    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'])
Commit	Line	Data
331ae266 JMF	1	from __future__ import unicode_literals
331ae266 JMF	2
171ca612	3	from .common import InfoExtractor
331ae266	4	from ..utils import (
5c4a81d9	5	fix_xml_ampersands,
944a3de2	6	parse_duration,
c9ea760e	7	qualities,
609a61e3	8	strip_jsonp,
5c4a81d9	9	unified_strdate,
d0df9292	10	url_basename,
331ae266 JMF	11	)
	12
	13
b9b42f2e	14	class NPOBaseIE(InfoExtractor):
04e0bac2 S	15	def _get_token(self, video_id):
	16	token_page = self._download_webpage(
	17	'http://ida.omroep.nl/npoplayer/i.js',
	18	video_id, note='Downloading token')
8d06a624	19	token = self._search_regex(
04e0bac2	20	r'npoplayer\.token = "(.+?)"', token_page, 'token')
f5ca97e3	21	# Decryption algorithm extracted from http://npoplayer.omroep.nl/csjs/npoplayer-min.js
8d06a624 YCH	22	token_l = list(token)
	23	first = second = None
	24	for i in range(5, len(token_l) - 4):
	25	if token_l[i].isdigit():
	26	if first is None:
	27	first = i
	28	elif second is None:
	29	second = i
	30	if first is None or second is None:
	31	first = 12
	32	second = 13
	33
	34	token_l[first], token_l[second] = token_l[second], token_l[first]
	35
	36	return ''.join(token_l)
04e0bac2 S	37
	38
	39	class NPOIE(NPOBaseIE):
e118031e S	40	IE_NAME = 'npo'
	41	IE_DESC = 'npo.nl and ntr.nl'
	42	_VALID_URL = r'https?://(?:www\.)?(?:npo\|ntr)\.nl/(?!live\|radio)(?:[^/]+/){2,}(?P<id>[^/?#]+)'
331ae266	43
944a3de2 S	44	_TESTS = [
	45	{
	46	'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
	47	'md5': '4b3f9c429157ec4775f2c9cb7b911016',
	48	'info_dict': {
	49	'id': 'VPWON_1220719',
	50	'ext': 'm4v',
	51	'title': 'Nieuwsuur',
	52	'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
	53	'upload_date': '20140622',
	54	},
331ae266	55	},
944a3de2 S	56	{
	57	'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
	58	'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
	59	'info_dict': {
	60	'id': 'VARA_101191800',
	61	'ext': 'm4v',
	62	'title': 'De Mega Mike & Mega Thomas show',
	63	'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
	64	'upload_date': '20090227',
	65	'duration': 2400,
	66	},
	67	},
	68	{
	69	'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
	70	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	71	'info_dict': {
	72	'id': 'VPWON_1169289',
	73	'ext': 'm4v',
	74	'title': 'Tegenlicht',
	75	'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
	76	'upload_date': '20130225',
	77	'duration': 3000,
	78	},
1ff30d7b S	79	},
	80	{
	81	'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
	82	'info_dict': {
	83	'id': 'WO_VPRO_043706',
	84	'ext': 'wmv',
	85	'title': 'De nieuwe mens - Deel 1',
	86	'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
	87	'duration': 4680,
	88	},
	89	'params': {
	90	# mplayer mms download
	91	'skip_download': True,
	92	}
	93	},
c85f3683 S	94	# non asf in streams
	95	{
	96	'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
	97	'md5': 'b3da13de374cbe2d5332a7e910bef97f',
	98	'info_dict': {
	99	'id': 'WO_NOS_762771',
	100	'ext': 'mp4',
	101	'title': 'Hoe gaat Europa verder na Parijs?',
	102	},
	103	},
e118031e S	104	{
	105	'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
	106	'md5': '01c6a2841675995da1f0cf776f03a9c3',
	107	'info_dict': {
	108	'id': 'VPWON_1233944',
	109	'ext': 'm4v',
	110	'title': 'Aap, poot, pies',
	111	'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
	112	'upload_date': '20150508',
	113	'duration': 599,
	114	},
	115	}
944a3de2	116	]
331ae266 JMF	117
331ae266 JMF	118	def _real_extract(self, url):
04e0bac2	119	video_id = self._match_id(url)
d0df9292	120	return self._get_info(video_id)
331ae266	121
d0df9292	122	def _get_info(self, video_id):
331ae266	123	metadata = self._download_json(
8d06a624	124	'http://e.omroep.nl/metadata/%s' % video_id,
331ae266 JMF	125	video_id,
331ae266 JMF	126	# We have to remove the javascript callback
609a61e3	127	transform_source=strip_jsonp,
331ae266	128	)
04e0bac2 S	129
04e0bac2 S	130	token = self._get_token(video_id)
331ae266	131
c9ea760e	132	formats = []
1ff30d7b S	133
	134	pubopties = metadata.get('pubopties')
	135	if pubopties:
	136	quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
	137	for format_id in pubopties:
	138	format_info = self._download_json(
	139	'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
	140	% (video_id, format_id, token),
	141	video_id, 'Downloading %s JSON' % format_id)
	142	if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
	143	continue
	144	streams = format_info.get('streams')
	145	if streams:
	146	video_info = self._download_json(
	147	streams[0] + '&type=json',
	148	video_id, 'Downloading %s stream JSON' % format_id)
	149	else:
	150	video_info = format_info
	151	video_url = video_info.get('url')
	152	if not video_url:
	153	continue
	154	if format_id == 'adaptive':
	155	formats.extend(self._extract_m3u8_formats(video_url, video_id))
	156	else:
	157	formats.append({
	158	'url': video_url,
	159	'format_id': format_id,
	160	'quality': quality(format_id),
	161	})
	162
	163	streams = metadata.get('streams')
	164	if streams:
	165	for i, stream in enumerate(streams):
	166	stream_url = stream.get('url')
	167	if not stream_url:
	168	continue
5c4a81d9	169	if '.asf' not in stream_url:
a0977064 S	170	formats.append({
	171	'url': stream_url,
	172	'quality': stream.get('kwaliteit'),
	173	})
	174	continue
1ff30d7b S	175	asx = self._download_xml(
	176	stream_url, video_id,
	177	'Downloading stream %d ASX playlist' % i,
	178	transform_source=fix_xml_ampersands)
	179	ref = asx.find('./ENTRY/Ref')
	180	if ref is None:
	181	continue
	182	video_url = ref.get('href')
	183	if not video_url:
	184	continue
c9ea760e	185	formats.append({
944a3de2	186	'url': video_url,
1ff30d7b S	187	'ext': stream.get('formaat', 'asf'),
1ff30d7b S	188	'quality': stream.get('kwaliteit'),
c9ea760e	189	})
1ff30d7b	190
9e91449c	191	self._sort_formats(formats)
25e5ebf3	192
9e91449c S	193	subtitles = {}
9e91449c S	194	if metadata.get('tt888') == 'ja':
b9b42f2e JMF	195	subtitles['nl'] = [{
	196	'ext': 'vtt',
	197	'url': 'http://e.omroep.nl/tt888/%s' % video_id,
	198	}]
331ae266 JMF	199
	200	return {
	201	'id': video_id,
	202	'title': metadata['titel'],
331ae266	203	'description': metadata['info'],
944a3de2 S	204	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	205	'upload_date': unified_strdate(metadata.get('gidsdatum')),
	206	'duration': parse_duration(metadata.get('tijdsduur')),
c9ea760e	207	'formats': formats,
25e5ebf3	208	'subtitles': subtitles,
331ae266	209	}
d0df9292 JMF	210
d0df9292 JMF	211
04e0bac2 S	212	class NPOLiveIE(NPOBaseIE):
04e0bac2 S	213	IE_NAME = 'npo.nl:live'
171ca612	214	_VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>.+)'
04e0bac2 S	215
	216	_TEST = {
	217	'url': 'http://www.npo.nl/live/npo-1',
	218	'info_dict': {
	219	'id': 'LI_NEDERLAND1_136692',
	220	'display_id': 'npo-1',
	221	'ext': 'mp4',
	222	'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	223	'description': 'Livestream',
	224	'is_live': True,
	225	},
	226	'params': {
	227	'skip_download': True,
	228	}
	229	}
	230
	231	def _real_extract(self, url):
	232	display_id = self._match_id(url)
	233
	234	webpage = self._download_webpage(url, display_id)
	235
	236	live_id = self._search_regex(
	237	r'data-prid="([^"]+)"', webpage, 'live id')
	238
	239	metadata = self._download_json(
	240	'http://e.omroep.nl/metadata/%s' % live_id,
	241	display_id, transform_source=strip_jsonp)
	242
	243	token = self._get_token(display_id)
	244
	245	formats = []
	246
	247	streams = metadata.get('streams')
	248	if streams:
	249	for stream in streams:
	250	stream_type = stream.get('type').lower()
32aaeca7 S	251	# smooth streaming is not supported
32aaeca7 S	252	if stream_type in ['ss', 'ms']:
04e0bac2 S	253	continue
	254	stream_info = self._download_json(
	255	'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
	256	% (stream.get('url'), token),
	257	display_id, 'Downloading %s JSON' % stream_type)
	258	if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
	259	continue
	260	stream_url = self._download_json(
	261	stream_info['stream'], display_id,
	262	'Downloading %s URL' % stream_type,
b0872c19 S	263	'Unable to download %s URL' % stream_type,
	264	transform_source=strip_jsonp, fatal=False)
	265	if not stream_url:
	266	continue
04e0bac2 S	267	if stream_type == 'hds':
	268	f4m_formats = self._extract_f4m_formats(stream_url, display_id)
	269	# f4m downloader downloads only piece of live stream
	270	for f4m_format in f4m_formats:
	271	f4m_format['preference'] = -1
	272	formats.extend(f4m_formats)
	273	elif stream_type == 'hls':
	274	formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
	275	else:
	276	formats.append({
	277	'url': stream_url,
32aaeca7	278	'preference': -10,
04e0bac2 S	279	})
	280
	281	self._sort_formats(formats)
	282
	283	return {
	284	'id': live_id,
	285	'display_id': display_id,
	286	'title': self._live_title(metadata['titel']),
	287	'description': metadata['info'],
	288	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	289	'formats': formats,
	290	'is_live': True,
	291	}
	292
	293
171ca612 S	294	class NPORadioIE(InfoExtractor):
	295	IE_NAME = 'npo.nl:radio'
	296	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
	297
	298	_TEST = {
	299	'url': 'http://www.npo.nl/radio/radio-1',
	300	'info_dict': {
	301	'id': 'radio-1',
	302	'ext': 'mp3',
	303	'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	304	'is_live': True,
	305	},
	306	'params': {
	307	'skip_download': True,
	308	}
	309	}
	310
	311	@staticmethod
	312	def _html_get_attribute_regex(attribute):
	313	return r'{0}\s=\s\'([^\']+)\''.format(attribute)
	314
	315	def _real_extract(self, url):
	316	video_id = self._match_id(url)
	317
	318	webpage = self._download_webpage(url, video_id)
	319
	320	title = self._html_search_regex(
	321	self._html_get_attribute_regex('data-channel'), webpage, 'title')
	322
	323	stream = self._parse_json(
	324	self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
	325	video_id)
	326
	327	codec = stream.get('codec')
	328
	329	return {
	330	'id': video_id,
	331	'url': stream['url'],
	332	'title': self._live_title(title),
	333	'acodec': codec,
	334	'ext': codec,
	335	'is_live': True,
	336	}
	337
	338
	339	class NPORadioFragmentIE(InfoExtractor):
	340	IE_NAME = 'npo.nl:radio:fragment'
	341	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
	342
	343	_TEST = {
	344	'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
	345	'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
	346	'info_dict': {
	347	'id': '174356',
	348	'ext': 'mp3',
	349	'title': 'Jubileumconcert Willeke Alberti',
	350	},
	351	}
	352
	353	def _real_extract(self, url):
	354	audio_id = self._match_id(url)
	355
	356	webpage = self._download_webpage(url, audio_id)
	357
358	title = self._html_search_regex(
359	r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
360	webpage, 'title')
361
362	audio_url = self._search_regex(
363	r"data-streams='([^']+)'", webpage, 'audio url')
364
365	return {
366	'id': audio_id,
367	'url': audio_url,
368	'title': title,
369	}
370
371
d0df9292 JMF	372	class TegenlichtVproIE(NPOIE):
	373	IE_NAME = 'tegenlicht.vpro.nl'
	374	_VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'
	375
	376	_TESTS = [
	377	{
	378	'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
	379	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	380	'info_dict': {
	381	'id': 'VPWON_1169289',
	382	'ext': 'm4v',
	383	'title': 'Tegenlicht',
	384	'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
	385	'upload_date': '20130225',
	386	},
	387	},
	388	]
	389
	390	def _real_extract(self, url):
	391	name = url_basename(url)
	392	webpage = self._download_webpage(url, name)
	393	urn = self._html_search_meta('mediaurn', webpage)
	394	info_page = self._download_json(
	395	'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name)
	396	return self._get_info(info_page['mid'])