[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.nl'
    _VALID_URL = r'https?://(?:www\.)?npo\.nl/(?!live|radio)[^/]+/[^/]+/(?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?',
            },
        },
    ]

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