[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)
                    (?:
                        npo:|
                        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 VPROIE(NPOIE):
    _VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html'

    _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': 'md5:52cf4eefbc96fffcbdc06d024147abea',
                'upload_date': '20130225',
            },
        },
        {
            'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
            'info_dict': {
                'id': 'sergio-herman',
                'title': 'Sergio Herman: Fucking perfect',
            },
            'playlist_count': 2,
        }
    ]

    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, 'NPO')
            for video_id in re.findall(r'data-media-id="([^"]+)"', webpage)
        ]

        playlist_title = self._search_regex(
            r'<title>\s*([^>]+?)\s*-\s*Teledoc\s*-\s*VPRO\s*</title>',
            webpage, 'playlist title', default=None) or self._og_search_title(webpage)

        return self.playlist_result(entries, playlist_id, playlist_title)


class WNLIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'

    _TEST = {
        '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,
    }

    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, 'NPO')
            for video_id, part in re.findall(
                r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>(Deel \d+)', webpage)
        ]

        playlist_title = self._html_search_regex(
            r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>',
            webpage, 'playlist title')

        return self.playlist_result(entries, playlist_id, playlist_title)
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	48	_VALID_URL = r'''(?x)
03f32a7e S	49	(?:
	50	npo:\|
	51	https?://
	52	(?:www\.)?
	53	(?:
	54	npo\.nl/(?!live\|radio)(?:[^/]+/){2}\|
	55	ntr\.nl/(?:[^/]+/){2,}\|
	56	omroepwnl\.nl/video/fragment/[^/]+__
	57	)
525daedd S	58	)
	59	(?P<id>[^/?#]+)
	60	'''
331ae266	61
944a3de2 S	62	_TESTS = [
	63	{
	64	'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
	65	'md5': '4b3f9c429157ec4775f2c9cb7b911016',
	66	'info_dict': {
	67	'id': 'VPWON_1220719',
	68	'ext': 'm4v',
	69	'title': 'Nieuwsuur',
	70	'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
	71	'upload_date': '20140622',
	72	},
331ae266	73	},
944a3de2 S	74	{
	75	'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
	76	'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
	77	'info_dict': {
	78	'id': 'VARA_101191800',
	79	'ext': 'm4v',
	80	'title': 'De Mega Mike & Mega Thomas show',
	81	'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
	82	'upload_date': '20090227',
	83	'duration': 2400,
	84	},
	85	},
	86	{
	87	'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
	88	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	89	'info_dict': {
	90	'id': 'VPWON_1169289',
	91	'ext': 'm4v',
	92	'title': 'Tegenlicht',
50ea2bb2	93	'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
944a3de2 S	94	'upload_date': '20130225',
	95	'duration': 3000,
	96	},
1ff30d7b S	97	},
	98	{
	99	'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
	100	'info_dict': {
	101	'id': 'WO_VPRO_043706',
	102	'ext': 'wmv',
	103	'title': 'De nieuwe mens - Deel 1',
	104	'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
	105	'duration': 4680,
	106	},
	107	'params': {
	108	# mplayer mms download
	109	'skip_download': True,
	110	}
	111	},
c85f3683 S	112	# non asf in streams
	113	{
	114	'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
	115	'md5': 'b3da13de374cbe2d5332a7e910bef97f',
	116	'info_dict': {
	117	'id': 'WO_NOS_762771',
	118	'ext': 'mp4',
	119	'title': 'Hoe gaat Europa verder na Parijs?',
	120	},
	121	},
e118031e S	122	{
	123	'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
	124	'md5': '01c6a2841675995da1f0cf776f03a9c3',
	125	'info_dict': {
	126	'id': 'VPWON_1233944',
	127	'ext': 'm4v',
	128	'title': 'Aap, poot, pies',
	129	'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
	130	'upload_date': '20150508',
	131	'duration': 599,
	132	},
525daedd S	133	},
	134	{
	135	'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
	136	'md5': 'd30cd8417b8b9bca1fdff27428860d08',
	137	'info_dict': {
	138	'id': 'POW_00996502',
	139	'ext': 'm4v',
	140	'title': '''"Dit is wel een 'landslide'..."''',
	141	'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
	142	'upload_date': '20150508',
	143	'duration': 462,
	144	},
e118031e	145	}
944a3de2	146	]
331ae266 JMF	147
331ae266 JMF	148	def _real_extract(self, url):
04e0bac2	149	video_id = self._match_id(url)
d0df9292	150	return self._get_info(video_id)
331ae266	151
d0df9292	152	def _get_info(self, video_id):
331ae266	153	metadata = self._download_json(
8d06a624	154	'http://e.omroep.nl/metadata/%s' % video_id,
331ae266 JMF	155	video_id,
331ae266 JMF	156	# We have to remove the javascript callback
609a61e3	157	transform_source=strip_jsonp,
331ae266	158	)
04e0bac2	159
525daedd S	160	# For some videos actual video id (prid) is different (e.g. for
	161	# http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
	162	# video id is POMS_WNL_853698 but prid is POW_00996502)
	163	video_id = metadata.get('prid') or video_id
	164
04e0bac2	165	token = self._get_token(video_id)
331ae266	166
c9ea760e	167	formats = []
1ff30d7b S	168
	169	pubopties = metadata.get('pubopties')
	170	if pubopties:
	171	quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
	172	for format_id in pubopties:
	173	format_info = self._download_json(
	174	'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
	175	% (video_id, format_id, token),
	176	video_id, 'Downloading %s JSON' % format_id)
	177	if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
	178	continue
	179	streams = format_info.get('streams')
	180	if streams:
	181	video_info = self._download_json(
	182	streams[0] + '&type=json',
	183	video_id, 'Downloading %s stream JSON' % format_id)
	184	else:
	185	video_info = format_info
	186	video_url = video_info.get('url')
	187	if not video_url:
	188	continue
	189	if format_id == 'adaptive':
	190	formats.extend(self._extract_m3u8_formats(video_url, video_id))
	191	else:
	192	formats.append({
	193	'url': video_url,
	194	'format_id': format_id,
	195	'quality': quality(format_id),
	196	})
	197
	198	streams = metadata.get('streams')
	199	if streams:
	200	for i, stream in enumerate(streams):
	201	stream_url = stream.get('url')
	202	if not stream_url:
	203	continue
5c4a81d9	204	if '.asf' not in stream_url:
a0977064 S	205	formats.append({
	206	'url': stream_url,
	207	'quality': stream.get('kwaliteit'),
	208	})
	209	continue
1ff30d7b S	210	asx = self._download_xml(
	211	stream_url, video_id,
	212	'Downloading stream %d ASX playlist' % i,
	213	transform_source=fix_xml_ampersands)
	214	ref = asx.find('./ENTRY/Ref')
	215	if ref is None:
	216	continue
	217	video_url = ref.get('href')
	218	if not video_url:
	219	continue
c9ea760e	220	formats.append({
944a3de2	221	'url': video_url,
1ff30d7b S	222	'ext': stream.get('formaat', 'asf'),
1ff30d7b S	223	'quality': stream.get('kwaliteit'),
c9ea760e	224	})
1ff30d7b	225
9e91449c	226	self._sort_formats(formats)
25e5ebf3	227
9e91449c S	228	subtitles = {}
9e91449c S	229	if metadata.get('tt888') == 'ja':
b9b42f2e JMF	230	subtitles['nl'] = [{
	231	'ext': 'vtt',
	232	'url': 'http://e.omroep.nl/tt888/%s' % video_id,
	233	}]
331ae266 JMF	234
	235	return {
	236	'id': video_id,
	237	'title': metadata['titel'],
331ae266	238	'description': metadata['info'],
944a3de2 S	239	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	240	'upload_date': unified_strdate(metadata.get('gidsdatum')),
	241	'duration': parse_duration(metadata.get('tijdsduur')),
c9ea760e	242	'formats': formats,
25e5ebf3	243	'subtitles': subtitles,
331ae266	244	}
d0df9292 JMF	245
d0df9292 JMF	246
04e0bac2 S	247	class NPOLiveIE(NPOBaseIE):
04e0bac2 S	248	IE_NAME = 'npo.nl:live'
171ca612	249	_VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>.+)'
04e0bac2 S	250
	251	_TEST = {
	252	'url': 'http://www.npo.nl/live/npo-1',
	253	'info_dict': {
	254	'id': 'LI_NEDERLAND1_136692',
	255	'display_id': 'npo-1',
	256	'ext': 'mp4',
	257	'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	258	'description': 'Livestream',
	259	'is_live': True,
	260	},
	261	'params': {
	262	'skip_download': True,
	263	}
	264	}
	265
	266	def _real_extract(self, url):
	267	display_id = self._match_id(url)
	268
	269	webpage = self._download_webpage(url, display_id)
	270
	271	live_id = self._search_regex(
	272	r'data-prid="([^"]+)"', webpage, 'live id')
	273
	274	metadata = self._download_json(
	275	'http://e.omroep.nl/metadata/%s' % live_id,
	276	display_id, transform_source=strip_jsonp)
	277
	278	token = self._get_token(display_id)
	279
	280	formats = []
	281
	282	streams = metadata.get('streams')
	283	if streams:
	284	for stream in streams:
	285	stream_type = stream.get('type').lower()
32aaeca7 S	286	# smooth streaming is not supported
32aaeca7 S	287	if stream_type in ['ss', 'ms']:
04e0bac2 S	288	continue
	289	stream_info = self._download_json(
	290	'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
	291	% (stream.get('url'), token),
	292	display_id, 'Downloading %s JSON' % stream_type)
	293	if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
	294	continue
	295	stream_url = self._download_json(
	296	stream_info['stream'], display_id,
	297	'Downloading %s URL' % stream_type,
b0872c19 S	298	'Unable to download %s URL' % stream_type,
	299	transform_source=strip_jsonp, fatal=False)
	300	if not stream_url:
	301	continue
04e0bac2 S	302	if stream_type == 'hds':
	303	f4m_formats = self._extract_f4m_formats(stream_url, display_id)
	304	# f4m downloader downloads only piece of live stream
	305	for f4m_format in f4m_formats:
	306	f4m_format['preference'] = -1
	307	formats.extend(f4m_formats)
	308	elif stream_type == 'hls':
	309	formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
	310	else:
	311	formats.append({
	312	'url': stream_url,
32aaeca7	313	'preference': -10,
04e0bac2 S	314	})
	315
	316	self._sort_formats(formats)
	317
	318	return {
	319	'id': live_id,
	320	'display_id': display_id,
	321	'title': self._live_title(metadata['titel']),
	322	'description': metadata['info'],
	323	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	324	'formats': formats,
	325	'is_live': True,
	326	}
	327
	328
171ca612 S	329	class NPORadioIE(InfoExtractor):
	330	IE_NAME = 'npo.nl:radio'
	331	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
	332
	333	_TEST = {
	334	'url': 'http://www.npo.nl/radio/radio-1',
	335	'info_dict': {
	336	'id': 'radio-1',
	337	'ext': 'mp3',
	338	'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	339	'is_live': True,
	340	},
	341	'params': {
	342	'skip_download': True,
	343	}
	344	}
	345
	346	@staticmethod
	347	def _html_get_attribute_regex(attribute):
	348	return r'{0}\s=\s\'([^\']+)\''.format(attribute)
	349
	350	def _real_extract(self, url):
	351	video_id = self._match_id(url)
	352
	353	webpage = self._download_webpage(url, video_id)
	354
	355	title = self._html_search_regex(
	356	self._html_get_attribute_regex('data-channel'), webpage, 'title')
	357
	358	stream = self._parse_json(
	359	self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
	360	video_id)
	361
	362	codec = stream.get('codec')
	363
	364	return {
	365	'id': video_id,
	366	'url': stream['url'],
	367	'title': self._live_title(title),
	368	'acodec': codec,
	369	'ext': codec,
	370	'is_live': True,
	371	}
	372
	373
	374	class NPORadioFragmentIE(InfoExtractor):
	375	IE_NAME = 'npo.nl:radio:fragment'
	376	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
	377
	378	_TEST = {
	379	'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
	380	'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
	381	'info_dict': {
	382	'id': '174356',
	383	'ext': 'mp3',
	384	'title': 'Jubileumconcert Willeke Alberti',
	385	},
	386	}
	387
	388	def _real_extract(self, url):
	389	audio_id = self._match_id(url)
	390
	391	webpage = self._download_webpage(url, audio_id)
	392
393	title = self._html_search_regex(
394	r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
395	webpage, 'title')
396
397	audio_url = self._search_regex(
398	r"data-streams='([^']+)'", webpage, 'audio url')
399
400	return {
401	'id': audio_id,
402	'url': audio_url,
403	'title': title,
404	}
405
406
611ac379 S	407	class VPROIE(NPOIE):
611ac379 S	408	_VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html'
d0df9292 JMF	409
	410	_TESTS = [
	411	{
	412	'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
	413	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	414	'info_dict': {
	415	'id': 'VPWON_1169289',
	416	'ext': 'm4v',
	417	'title': 'Tegenlicht',
611ac379	418	'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
d0df9292 JMF	419	'upload_date': '20130225',
	420	},
	421	},
611ac379 S	422	{
	423	'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
	424	'info_dict': {
	425	'id': 'sergio-herman',
	426	'title': 'Sergio Herman: Fucking perfect',
	427	},
	428	'playlist_count': 2,
	429	}
d0df9292 JMF	430	]
	431
	432	def _real_extract(self, url):
611ac379 S	433	playlist_id = self._match_id(url)
	434
	435	webpage = self._download_webpage(url, playlist_id)
	436
	437	entries = [
	438	self.url_result('npo:%s' % video_id, 'NPO')
	439	for video_id in re.findall(r'data-media-id="([^"]+)"', webpage)
	440	]
	441
	442	playlist_title = self._search_regex(
	443	r'<title>\s([^>]+?)\s-\sTeledoc\s-\sVPRO\s</title>',
	444	webpage, 'playlist title', default=None) or self._og_search_title(webpage)
	445
	446	return self.playlist_result(entries, playlist_id, playlist_title)
03f32a7e S	447
	448
	449	class WNLIE(InfoExtractor):
	450	_VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
	451
	452	_TEST = {
	453	'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
	454	'info_dict': {
	455	'id': 'vandaag-de-dag-6-mei',
	456	'title': 'Vandaag de Dag 6 mei',
	457	},
	458	'playlist_count': 4,
	459	}
	460
	461	def _real_extract(self, url):
	462	playlist_id = self._match_id(url)
	463
	464	webpage = self._download_webpage(url, playlist_id)
	465
	466	entries = [
	467	self.url_result('npo:%s' % video_id, 'NPO')
	468	for video_id, part in re.findall(
	469	r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>(Deel \d+)', webpage)
	470	]
	471
	472	playlist_title = self._html_search_regex(
	473	r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>',
	474	webpage, 'playlist title')
	475
	476	return self.playlist_result(entries, playlist_id, playlist_title)