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

from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..compat import (
    compat_urllib_request,
    compat_urllib_parse,
)
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'''(?x)
                    https?://
                        (?:www\.)?
                        (?:
                            npo\.nl/(?!live|radio)(?:[^/]+/){2}|
                            ntr\.nl/(?:[^/]+/){2,}|
                            omroepwnl\.nl/video/fragment/[^/]+__
                        )
                        (?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': 'md5:52cf4eefbc96fffcbdc06d024147abea',
                '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):
        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,
        )

        # 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

        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
525daedd S	3	import re
525daedd S	4
171ca612	5	from .common import InfoExtractor
525daedd S	6	from ..compat import (
	7	compat_urllib_request,
	8	compat_urllib_parse,
	9	)
331ae266	10	from ..utils import (
5c4a81d9	11	fix_xml_ampersands,
944a3de2	12	parse_duration,
c9ea760e	13	qualities,
609a61e3	14	strip_jsonp,
5c4a81d9	15	unified_strdate,
d0df9292	16	url_basename,
331ae266 JMF	17	)
	18
	19
b9b42f2e	20	class NPOBaseIE(InfoExtractor):
04e0bac2 S	21	def _get_token(self, video_id):
	22	token_page = self._download_webpage(
	23	'http://ida.omroep.nl/npoplayer/i.js',
	24	video_id, note='Downloading token')
8d06a624	25	token = self._search_regex(
04e0bac2	26	r'npoplayer\.token = "(.+?)"', token_page, 'token')
f5ca97e3	27	# Decryption algorithm extracted from http://npoplayer.omroep.nl/csjs/npoplayer-min.js
8d06a624 YCH	28	token_l = list(token)
	29	first = second = None
	30	for i in range(5, len(token_l) - 4):
	31	if token_l[i].isdigit():
	32	if first is None:
	33	first = i
	34	elif second is None:
	35	second = i
	36	if first is None or second is None:
	37	first = 12
	38	second = 13
	39
	40	token_l[first], token_l[second] = token_l[second], token_l[first]
	41
	42	return ''.join(token_l)
04e0bac2 S	43
	44
	45	class NPOIE(NPOBaseIE):
e118031e S	46	IE_NAME = 'npo'
e118031e S	47	IE_DESC = 'npo.nl and ntr.nl'
525daedd S	48	_VALID_URL = r'''(?x)
	49	https?://
	50	(?:www\.)?
	51	(?:
	52	npo\.nl/(?!live\|radio)(?:[^/]+/){2}\|
	53	ntr\.nl/(?:[^/]+/){2,}\|
	54	omroepwnl\.nl/video/fragment/[^/]+__
	55	)
	56	(?P<id>[^/?#]+)
	57	'''
331ae266	58
944a3de2 S	59	_TESTS = [
	60	{
	61	'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
	62	'md5': '4b3f9c429157ec4775f2c9cb7b911016',
	63	'info_dict': {
	64	'id': 'VPWON_1220719',
	65	'ext': 'm4v',
	66	'title': 'Nieuwsuur',
	67	'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
	68	'upload_date': '20140622',
	69	},
331ae266	70	},
944a3de2 S	71	{
	72	'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
	73	'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
	74	'info_dict': {
	75	'id': 'VARA_101191800',
	76	'ext': 'm4v',
	77	'title': 'De Mega Mike & Mega Thomas show',
	78	'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
	79	'upload_date': '20090227',
	80	'duration': 2400,
	81	},
	82	},
	83	{
	84	'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
	85	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	86	'info_dict': {
	87	'id': 'VPWON_1169289',
	88	'ext': 'm4v',
	89	'title': 'Tegenlicht',
50ea2bb2	90	'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
944a3de2 S	91	'upload_date': '20130225',
	92	'duration': 3000,
	93	},
1ff30d7b S	94	},
	95	{
	96	'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
	97	'info_dict': {
	98	'id': 'WO_VPRO_043706',
	99	'ext': 'wmv',
	100	'title': 'De nieuwe mens - Deel 1',
	101	'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
	102	'duration': 4680,
	103	},
	104	'params': {
	105	# mplayer mms download
	106	'skip_download': True,
	107	}
	108	},
c85f3683 S	109	# non asf in streams
	110	{
	111	'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
	112	'md5': 'b3da13de374cbe2d5332a7e910bef97f',
	113	'info_dict': {
	114	'id': 'WO_NOS_762771',
	115	'ext': 'mp4',
	116	'title': 'Hoe gaat Europa verder na Parijs?',
	117	},
	118	},
e118031e S	119	{
	120	'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
	121	'md5': '01c6a2841675995da1f0cf776f03a9c3',
	122	'info_dict': {
	123	'id': 'VPWON_1233944',
	124	'ext': 'm4v',
	125	'title': 'Aap, poot, pies',
	126	'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
	127	'upload_date': '20150508',
	128	'duration': 599,
	129	},
525daedd S	130	},
	131	{
	132	'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
	133	'md5': 'd30cd8417b8b9bca1fdff27428860d08',
	134	'info_dict': {
	135	'id': 'POW_00996502',
	136	'ext': 'm4v',
	137	'title': '''"Dit is wel een 'landslide'..."''',
	138	'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
	139	'upload_date': '20150508',
	140	'duration': 462,
	141	},
e118031e	142	}
944a3de2	143	]
331ae266 JMF	144
331ae266 JMF	145	def _real_extract(self, url):
04e0bac2	146	video_id = self._match_id(url)
d0df9292	147	return self._get_info(video_id)
331ae266	148
d0df9292	149	def _get_info(self, video_id):
331ae266	150	metadata = self._download_json(
8d06a624	151	'http://e.omroep.nl/metadata/%s' % video_id,
331ae266 JMF	152	video_id,
331ae266 JMF	153	# We have to remove the javascript callback
609a61e3	154	transform_source=strip_jsonp,
331ae266	155	)
04e0bac2	156
525daedd S	157	# For some videos actual video id (prid) is different (e.g. for
	158	# http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
	159	# video id is POMS_WNL_853698 but prid is POW_00996502)
	160	video_id = metadata.get('prid') or video_id
	161
04e0bac2	162	token = self._get_token(video_id)
331ae266	163
c9ea760e	164	formats = []
1ff30d7b S	165
	166	pubopties = metadata.get('pubopties')
	167	if pubopties:
	168	quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
	169	for format_id in pubopties:
	170	format_info = self._download_json(
	171	'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
	172	% (video_id, format_id, token),
	173	video_id, 'Downloading %s JSON' % format_id)
	174	if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
	175	continue
	176	streams = format_info.get('streams')
	177	if streams:
	178	video_info = self._download_json(
	179	streams[0] + '&type=json',
	180	video_id, 'Downloading %s stream JSON' % format_id)
	181	else:
	182	video_info = format_info
	183	video_url = video_info.get('url')
	184	if not video_url:
	185	continue
	186	if format_id == 'adaptive':
	187	formats.extend(self._extract_m3u8_formats(video_url, video_id))
	188	else:
	189	formats.append({
	190	'url': video_url,
	191	'format_id': format_id,
	192	'quality': quality(format_id),
	193	})
	194
	195	streams = metadata.get('streams')
	196	if streams:
	197	for i, stream in enumerate(streams):
	198	stream_url = stream.get('url')
	199	if not stream_url:
	200	continue
5c4a81d9	201	if '.asf' not in stream_url:
a0977064 S	202	formats.append({
	203	'url': stream_url,
	204	'quality': stream.get('kwaliteit'),
	205	})
	206	continue
1ff30d7b S	207	asx = self._download_xml(
	208	stream_url, video_id,
	209	'Downloading stream %d ASX playlist' % i,
	210	transform_source=fix_xml_ampersands)
	211	ref = asx.find('./ENTRY/Ref')
	212	if ref is None:
	213	continue
	214	video_url = ref.get('href')
	215	if not video_url:
	216	continue
c9ea760e	217	formats.append({
944a3de2	218	'url': video_url,
1ff30d7b S	219	'ext': stream.get('formaat', 'asf'),
1ff30d7b S	220	'quality': stream.get('kwaliteit'),
c9ea760e	221	})
1ff30d7b	222
9e91449c	223	self._sort_formats(formats)
25e5ebf3	224
9e91449c S	225	subtitles = {}
9e91449c S	226	if metadata.get('tt888') == 'ja':
b9b42f2e JMF	227	subtitles['nl'] = [{
	228	'ext': 'vtt',
	229	'url': 'http://e.omroep.nl/tt888/%s' % video_id,
	230	}]
331ae266 JMF	231
	232	return {
	233	'id': video_id,
	234	'title': metadata['titel'],
331ae266	235	'description': metadata['info'],
944a3de2 S	236	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	237	'upload_date': unified_strdate(metadata.get('gidsdatum')),
	238	'duration': parse_duration(metadata.get('tijdsduur')),
c9ea760e	239	'formats': formats,
25e5ebf3	240	'subtitles': subtitles,
331ae266	241	}
d0df9292 JMF	242
d0df9292 JMF	243
04e0bac2 S	244	class NPOLiveIE(NPOBaseIE):
04e0bac2 S	245	IE_NAME = 'npo.nl:live'
171ca612	246	_VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>.+)'
04e0bac2 S	247
	248	_TEST = {
	249	'url': 'http://www.npo.nl/live/npo-1',
	250	'info_dict': {
	251	'id': 'LI_NEDERLAND1_136692',
	252	'display_id': 'npo-1',
	253	'ext': 'mp4',
	254	'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	255	'description': 'Livestream',
	256	'is_live': True,
	257	},
	258	'params': {
	259	'skip_download': True,
	260	}
	261	}
	262
	263	def _real_extract(self, url):
	264	display_id = self._match_id(url)
	265
	266	webpage = self._download_webpage(url, display_id)
	267
	268	live_id = self._search_regex(
	269	r'data-prid="([^"]+)"', webpage, 'live id')
	270
	271	metadata = self._download_json(
	272	'http://e.omroep.nl/metadata/%s' % live_id,
	273	display_id, transform_source=strip_jsonp)
	274
	275	token = self._get_token(display_id)
	276
	277	formats = []
	278
	279	streams = metadata.get('streams')
	280	if streams:
	281	for stream in streams:
	282	stream_type = stream.get('type').lower()
32aaeca7 S	283	# smooth streaming is not supported
32aaeca7 S	284	if stream_type in ['ss', 'ms']:
04e0bac2 S	285	continue
	286	stream_info = self._download_json(
	287	'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
	288	% (stream.get('url'), token),
	289	display_id, 'Downloading %s JSON' % stream_type)
	290	if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
	291	continue
	292	stream_url = self._download_json(
	293	stream_info['stream'], display_id,
	294	'Downloading %s URL' % stream_type,
b0872c19 S	295	'Unable to download %s URL' % stream_type,
	296	transform_source=strip_jsonp, fatal=False)
	297	if not stream_url:
	298	continue
04e0bac2 S	299	if stream_type == 'hds':
	300	f4m_formats = self._extract_f4m_formats(stream_url, display_id)
	301	# f4m downloader downloads only piece of live stream
	302	for f4m_format in f4m_formats:
	303	f4m_format['preference'] = -1
	304	formats.extend(f4m_formats)
	305	elif stream_type == 'hls':
	306	formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
	307	else:
	308	formats.append({
	309	'url': stream_url,
32aaeca7	310	'preference': -10,
04e0bac2 S	311	})
	312
	313	self._sort_formats(formats)
	314
	315	return {
	316	'id': live_id,
	317	'display_id': display_id,
	318	'title': self._live_title(metadata['titel']),
	319	'description': metadata['info'],
	320	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	321	'formats': formats,
	322	'is_live': True,
	323	}
	324
	325
171ca612 S	326	class NPORadioIE(InfoExtractor):
	327	IE_NAME = 'npo.nl:radio'
	328	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
	329
	330	_TEST = {
	331	'url': 'http://www.npo.nl/radio/radio-1',
	332	'info_dict': {
	333	'id': 'radio-1',
	334	'ext': 'mp3',
	335	'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	336	'is_live': True,
	337	},
	338	'params': {
	339	'skip_download': True,
	340	}
	341	}
	342
	343	@staticmethod
	344	def _html_get_attribute_regex(attribute):
	345	return r'{0}\s=\s\'([^\']+)\''.format(attribute)
	346
	347	def _real_extract(self, url):
	348	video_id = self._match_id(url)
	349
	350	webpage = self._download_webpage(url, video_id)
	351
	352	title = self._html_search_regex(
	353	self._html_get_attribute_regex('data-channel'), webpage, 'title')
	354
	355	stream = self._parse_json(
	356	self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
	357	video_id)
	358
	359	codec = stream.get('codec')
	360
	361	return {
	362	'id': video_id,
	363	'url': stream['url'],
	364	'title': self._live_title(title),
	365	'acodec': codec,
	366	'ext': codec,
	367	'is_live': True,
	368	}
	369
	370
	371	class NPORadioFragmentIE(InfoExtractor):
	372	IE_NAME = 'npo.nl:radio:fragment'
	373	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
	374
	375	_TEST = {
	376	'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
	377	'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
	378	'info_dict': {
	379	'id': '174356',
	380	'ext': 'mp3',
	381	'title': 'Jubileumconcert Willeke Alberti',
	382	},
	383	}
	384
	385	def _real_extract(self, url):
	386	audio_id = self._match_id(url)
	387
	388	webpage = self._download_webpage(url, audio_id)
	389
390	title = self._html_search_regex(
391	r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
392	webpage, 'title')
393
394	audio_url = self._search_regex(
395	r"data-streams='([^']+)'", webpage, 'audio url')
396
397	return {
398	'id': audio_id,
399	'url': audio_url,
400	'title': title,
401	}
402
403
d0df9292 JMF	404	class TegenlichtVproIE(NPOIE):
	405	IE_NAME = 'tegenlicht.vpro.nl'
	406	_VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'
	407
	408	_TESTS = [
	409	{
	410	'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
	411	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	412	'info_dict': {
	413	'id': 'VPWON_1169289',
	414	'ext': 'm4v',
	415	'title': 'Tegenlicht',
	416	'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
	417	'upload_date': '20130225',
	418	},
	419	},
	420	]
	421
	422	def _real_extract(self, url):
	423	name = url_basename(url)
	424	webpage = self._download_webpage(url, name)
	425	urn = self._html_search_meta('mediaurn', webpage)
	426	info_page = self._download_json(
	427	'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name)
	428	return self._get_info(info_page['mid'])