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

from __future__ import unicode_literals

import re

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


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: The best of.',
                '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: De toekomst komt uit Afrika',
                'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
                'upload_date': '20130225',
                'duration': 3000,
            },
        },
        {
            'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
            'info_dict': {
                'id': 'WO_VPRO_043706',
                'ext': 'wmv',
                'title': 'De nieuwe mens - Deel 1',
                'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
                'duration': 4680,
            },
            'params': {
                # mplayer mms download
                'skip_download': True,
            }
        },
        # non asf in streams
        {
            'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
            'md5': 'b3da13de374cbe2d5332a7e910bef97f',
            'info_dict': {
                'id': 'WO_NOS_762771',
                'ext': 'mp4',
                'title': 'Hoe gaat Europa verder na Parijs?',
            },
        },
        {
            'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
            'md5': '01c6a2841675995da1f0cf776f03a9c3',
            'info_dict': {
                'id': 'VPWON_1233944',
                'ext': 'm4v',
                'title': 'Aap, poot, pies',
                'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
                'upload_date': '20150508',
                'duration': 599,
            },
        },
        {
            'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
            'md5': 'd30cd8417b8b9bca1fdff27428860d08',
            'info_dict': {
                'id': 'POW_00996502',
                'ext': 'm4v',
                'title': '''"Dit is wel een 'landslide'..."''',
                'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
                'upload_date': '20150508',
                'duration': 462,
            },
        }
    ]

    def _real_extract(self, url):
        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

        # titel is too generic in some cases so utilize aflevering_titel as well
        # when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html)
        title = metadata['titel']
        sub_title = metadata.get('aflevering_titel')
        if sub_title and sub_title != title:
            title += ': %s' % sub_title

        token = self._get_token(video_id)

        formats = []

        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, 'mp4'))
                else:
                    formats.append({
                        'url': video_url,
                        'format_id': format_id,
                        'quality': quality(format_id),
                    })

        streams = metadata.get('streams')
        if streams:
            for i, stream in enumerate(streams):
                stream_url = stream.get('url')
                if not stream_url:
                    continue
                if '.asf' not in stream_url:
                    formats.append({
                        'url': stream_url,
                        'quality': stream.get('kwaliteit'),
                    })
                    continue
                asx = self._download_xml(
                    stream_url, video_id,
                    'Downloading stream %d ASX playlist' % i,
                    transform_source=fix_xml_ampersands)
                ref = asx.find('./ENTRY/Ref')
                if ref is None:
                    continue
                video_url = ref.get('href')
                if not video_url:
                    continue
                formats.append({
                    'url': video_url,
                    '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': title,
            'description': metadata.get('info'),
            'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
            'upload_date': unified_strdate(metadata.get('gidsdatum')),
            'duration': parse_duration(metadata.get('tijdsduur')),
            'formats': formats,
            'subtitles': subtitles,
        }


class NPOLiveIE(NPOBaseIE):
    IE_NAME = 'npo.nl:live'
    _VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<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 SchoolTVIE(InfoExtractor):
    IE_NAME = 'schooltv'
    _VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'

    _TEST = {
        'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
        'info_dict': {
            'id': 'WO_NTR_429477',
            'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
            'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
            'ext': 'mp4',
            'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
        },
        'params': {
            # Skip because of m3u8 download
            'skip_download': True
        }
    }

    def _real_extract(self, url):
        display_id = self._match_id(url)
        webpage = self._download_webpage(url, display_id)
        video_id = self._search_regex(
            r'data-mid=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video_id', group='id')
        return {
            '_type': 'url_transparent',
            'ie_key': 'NPO',
            'url': 'npo:%s' % video_id,
            'display_id': display_id
        }


class NPOPlaylistBaseIE(NPOIE):
    def _real_extract(self, url):
        playlist_id = self._match_id(url)

        webpage = self._download_webpage(url, playlist_id)

        entries = [
            self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
            for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage))
        ]

        playlist_title = self._html_search_regex(
            self._PLAYLIST_TITLE_RE, webpage, 'playlist title',
            default=None) or self._og_search_title(webpage)

        return self.playlist_result(entries, playlist_id, playlist_title)


class VPROIE(NPOPlaylistBaseIE):
    IE_NAME = 'vpro'
    _VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html'
    _PLAYLIST_TITLE_RE = r'<h1[^>]+class=["\'].*?\bmedia-platform-title\b.*?["\'][^>]*>([^<]+)'
    _PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"'

    _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': 'De toekomst komt uit Afrika',
                'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
                'upload_date': '20130225',
            },
            'skip': 'Video gone',
        },
        {
            'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
            'info_dict': {
                'id': 'sergio-herman',
                'title': 'sergio herman: fucking perfect',
            },
            'playlist_count': 2,
        },
        {
            # playlist with youtube embed
            'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
            'info_dict': {
                'id': 'education-education',
                'title': 'education education',
            },
            'playlist_count': 2,
        }
    ]


class WNLIE(NPOPlaylistBaseIE):
    _VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
    _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>'
    _PLAYLIST_ENTRY_RE = r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+'

    _TESTS = [{
        'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
        'info_dict': {
            'id': 'vandaag-de-dag-6-mei',
            'title': 'Vandaag de Dag 6 mei',
        },
        'playlist_count': 4,
    }]


class AndereTijdenIE(NPOPlaylistBaseIE):
    _VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P<id>[^/?#&]+)'
    _PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class=["\'].*?\bpage-title\b.*?["\'][^>]*>(.+?)</h1>'
    _PLAYLIST_ENTRY_RE = r'<figure[^>]+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']'

    _TESTS = [{
        'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem',
        'info_dict': {
            'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem',
            'title': 'Duitse soldaten over de Slag bij Arnhem',
        },
        'playlist_count': 3,
    }]
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
331ae266	6	from ..utils import (
5c4a81d9	7	fix_xml_ampersands,
5742c18b	8	orderedSet,
944a3de2	9	parse_duration,
c9ea760e	10	qualities,
609a61e3	11	strip_jsonp,
5c4a81d9	12	unified_strdate,
331ae266 JMF	13	)
	14
	15
b9b42f2e	16	class NPOBaseIE(InfoExtractor):
04e0bac2 S	17	def _get_token(self, video_id):
	18	token_page = self._download_webpage(
	19	'http://ida.omroep.nl/npoplayer/i.js',
	20	video_id, note='Downloading token')
8d06a624	21	token = self._search_regex(
04e0bac2	22	r'npoplayer\.token = "(.+?)"', token_page, 'token')
f5ca97e3	23	# Decryption algorithm extracted from http://npoplayer.omroep.nl/csjs/npoplayer-min.js
8d06a624 YCH	24	token_l = list(token)
	25	first = second = None
	26	for i in range(5, len(token_l) - 4):
	27	if token_l[i].isdigit():
	28	if first is None:
	29	first = i
	30	elif second is None:
	31	second = i
	32	if first is None or second is None:
	33	first = 12
	34	second = 13
	35
	36	token_l[first], token_l[second] = token_l[second], token_l[first]
	37
	38	return ''.join(token_l)
04e0bac2 S	39
	40
	41	class NPOIE(NPOBaseIE):
e118031e S	42	IE_NAME = 'npo'
e118031e S	43	IE_DESC = 'npo.nl and ntr.nl'
525daedd	44	_VALID_URL = r'''(?x)
03f32a7e S	45	(?:
	46	npo:\|
	47	https?://
	48	(?:www\.)?
	49	(?:
	50	npo\.nl/(?!live\|radio)(?:[^/]+/){2}\|
	51	ntr\.nl/(?:[^/]+/){2,}\|
	52	omroepwnl\.nl/video/fragment/[^/]+__
	53	)
525daedd S	54	)
	55	(?P<id>[^/?#]+)
	56	'''
331ae266	57
944a3de2 S	58	_TESTS = [
	59	{
	60	'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
	61	'md5': '4b3f9c429157ec4775f2c9cb7b911016',
	62	'info_dict': {
	63	'id': 'VPWON_1220719',
	64	'ext': 'm4v',
	65	'title': 'Nieuwsuur',
	66	'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
	67	'upload_date': '20140622',
	68	},
331ae266	69	},
944a3de2 S	70	{
	71	'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
	72	'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
	73	'info_dict': {
	74	'id': 'VARA_101191800',
	75	'ext': 'm4v',
23fc384f	76	'title': 'De Mega Mike & Mega Thomas show: The best of.',
944a3de2 S	77	'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
	78	'upload_date': '20090227',
	79	'duration': 2400,
	80	},
	81	},
	82	{
	83	'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
	84	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	85	'info_dict': {
	86	'id': 'VPWON_1169289',
	87	'ext': 'm4v',
23fc384f	88	'title': 'Tegenlicht: De toekomst komt uit Afrika',
50ea2bb2	89	'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
944a3de2 S	90	'upload_date': '20130225',
	91	'duration': 3000,
	92	},
1ff30d7b S	93	},
	94	{
	95	'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
	96	'info_dict': {
	97	'id': 'WO_VPRO_043706',
	98	'ext': 'wmv',
	99	'title': 'De nieuwe mens - Deel 1',
	100	'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
	101	'duration': 4680,
	102	},
	103	'params': {
	104	# mplayer mms download
	105	'skip_download': True,
	106	}
	107	},
c85f3683 S	108	# non asf in streams
	109	{
	110	'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
	111	'md5': 'b3da13de374cbe2d5332a7e910bef97f',
	112	'info_dict': {
	113	'id': 'WO_NOS_762771',
	114	'ext': 'mp4',
	115	'title': 'Hoe gaat Europa verder na Parijs?',
	116	},
	117	},
e118031e S	118	{
	119	'url': 'http://www.ntr.nl/Aap-Poot-Pies/27/detail/Aap-poot-pies/VPWON_1233944#content',
	120	'md5': '01c6a2841675995da1f0cf776f03a9c3',
	121	'info_dict': {
	122	'id': 'VPWON_1233944',
	123	'ext': 'm4v',
	124	'title': 'Aap, poot, pies',
	125	'description': 'md5:c9c8005d1869ae65b858e82c01a91fde',
	126	'upload_date': '20150508',
	127	'duration': 599,
	128	},
525daedd S	129	},
	130	{
	131	'url': 'http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698',
	132	'md5': 'd30cd8417b8b9bca1fdff27428860d08',
	133	'info_dict': {
	134	'id': 'POW_00996502',
	135	'ext': 'm4v',
	136	'title': '''"Dit is wel een 'landslide'..."''',
	137	'description': 'md5:f8d66d537dfb641380226e31ca57b8e8',
	138	'upload_date': '20150508',
	139	'duration': 462,
	140	},
e118031e	141	}
944a3de2	142	]
331ae266 JMF	143
331ae266 JMF	144	def _real_extract(self, url):
04e0bac2	145	video_id = self._match_id(url)
d0df9292	146	return self._get_info(video_id)
331ae266	147
d0df9292	148	def _get_info(self, video_id):
331ae266	149	metadata = self._download_json(
8d06a624	150	'http://e.omroep.nl/metadata/%s' % video_id,
331ae266 JMF	151	video_id,
331ae266 JMF	152	# We have to remove the javascript callback
609a61e3	153	transform_source=strip_jsonp,
331ae266	154	)
04e0bac2	155
525daedd S	156	# For some videos actual video id (prid) is different (e.g. for
	157	# http://www.omroepwnl.nl/video/fragment/vandaag-de-dag-verkiezingen__POMS_WNL_853698
	158	# video id is POMS_WNL_853698 but prid is POW_00996502)
	159	video_id = metadata.get('prid') or video_id
	160
23fc384f S	161	# titel is too generic in some cases so utilize aflevering_titel as well
	162	# when available (e.g. http://tegenlicht.vpro.nl/afleveringen/2014-2015/access-to-africa.html)
	163	title = metadata['titel']
	164	sub_title = metadata.get('aflevering_titel')
	165	if sub_title and sub_title != title:
	166	title += ': %s' % sub_title
	167
04e0bac2	168	token = self._get_token(video_id)
331ae266	169
c9ea760e	170	formats = []
1ff30d7b S	171
	172	pubopties = metadata.get('pubopties')
	173	if pubopties:
	174	quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
	175	for format_id in pubopties:
	176	format_info = self._download_json(
	177	'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
	178	% (video_id, format_id, token),
	179	video_id, 'Downloading %s JSON' % format_id)
	180	if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
	181	continue
	182	streams = format_info.get('streams')
	183	if streams:
	184	video_info = self._download_json(
	185	streams[0] + '&type=json',
	186	video_id, 'Downloading %s stream JSON' % format_id)
	187	else:
	188	video_info = format_info
	189	video_url = video_info.get('url')
	190	if not video_url:
	191	continue
	192	if format_id == 'adaptive':
a7685b3a	193	formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4'))
1ff30d7b S	194	else:
	195	formats.append({
	196	'url': video_url,
	197	'format_id': format_id,
	198	'quality': quality(format_id),
	199	})
	200
	201	streams = metadata.get('streams')
	202	if streams:
	203	for i, stream in enumerate(streams):
	204	stream_url = stream.get('url')
	205	if not stream_url:
	206	continue
5c4a81d9	207	if '.asf' not in stream_url:
a0977064 S	208	formats.append({
	209	'url': stream_url,
	210	'quality': stream.get('kwaliteit'),
	211	})
	212	continue
1ff30d7b S	213	asx = self._download_xml(
	214	stream_url, video_id,
	215	'Downloading stream %d ASX playlist' % i,
	216	transform_source=fix_xml_ampersands)
	217	ref = asx.find('./ENTRY/Ref')
	218	if ref is None:
	219	continue
	220	video_url = ref.get('href')
	221	if not video_url:
	222	continue
c9ea760e	223	formats.append({
944a3de2	224	'url': video_url,
1ff30d7b S	225	'ext': stream.get('formaat', 'asf'),
1ff30d7b S	226	'quality': stream.get('kwaliteit'),
c9ea760e	227	})
1ff30d7b	228
9e91449c	229	self._sort_formats(formats)
25e5ebf3	230
9e91449c S	231	subtitles = {}
9e91449c S	232	if metadata.get('tt888') == 'ja':
b9b42f2e JMF	233	subtitles['nl'] = [{
	234	'ext': 'vtt',
	235	'url': 'http://e.omroep.nl/tt888/%s' % video_id,
	236	}]
331ae266 JMF	237
	238	return {
	239	'id': video_id,
23fc384f	240	'title': title,
536b0700	241	'description': metadata.get('info'),
944a3de2 S	242	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	243	'upload_date': unified_strdate(metadata.get('gidsdatum')),
	244	'duration': parse_duration(metadata.get('tijdsduur')),
c9ea760e	245	'formats': formats,
25e5ebf3	246	'subtitles': subtitles,
331ae266	247	}
d0df9292 JMF	248
d0df9292 JMF	249
04e0bac2 S	250	class NPOLiveIE(NPOBaseIE):
04e0bac2 S	251	IE_NAME = 'npo.nl:live'
171ca612	252	_VALID_URL = r'https?://(?:www\.)?npo\.nl/live/(?P<id>.+)'
04e0bac2 S	253
	254	_TEST = {
	255	'url': 'http://www.npo.nl/live/npo-1',
	256	'info_dict': {
	257	'id': 'LI_NEDERLAND1_136692',
	258	'display_id': 'npo-1',
	259	'ext': 'mp4',
	260	'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	261	'description': 'Livestream',
	262	'is_live': True,
	263	},
	264	'params': {
	265	'skip_download': True,
	266	}
	267	}
	268
	269	def _real_extract(self, url):
	270	display_id = self._match_id(url)
	271
	272	webpage = self._download_webpage(url, display_id)
	273
	274	live_id = self._search_regex(
	275	r'data-prid="([^"]+)"', webpage, 'live id')
	276
	277	metadata = self._download_json(
	278	'http://e.omroep.nl/metadata/%s' % live_id,
	279	display_id, transform_source=strip_jsonp)
	280
	281	token = self._get_token(display_id)
	282
	283	formats = []
	284
	285	streams = metadata.get('streams')
	286	if streams:
	287	for stream in streams:
	288	stream_type = stream.get('type').lower()
32aaeca7 S	289	# smooth streaming is not supported
32aaeca7 S	290	if stream_type in ['ss', 'ms']:
04e0bac2 S	291	continue
	292	stream_info = self._download_json(
	293	'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
	294	% (stream.get('url'), token),
	295	display_id, 'Downloading %s JSON' % stream_type)
	296	if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
	297	continue
	298	stream_url = self._download_json(
	299	stream_info['stream'], display_id,
	300	'Downloading %s URL' % stream_type,
b0872c19 S	301	'Unable to download %s URL' % stream_type,
	302	transform_source=strip_jsonp, fatal=False)
	303	if not stream_url:
	304	continue
04e0bac2 S	305	if stream_type == 'hds':
	306	f4m_formats = self._extract_f4m_formats(stream_url, display_id)
	307	# f4m downloader downloads only piece of live stream
	308	for f4m_format in f4m_formats:
	309	f4m_format['preference'] = -1
	310	formats.extend(f4m_formats)
	311	elif stream_type == 'hls':
	312	formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
	313	else:
	314	formats.append({
	315	'url': stream_url,
32aaeca7	316	'preference': -10,
04e0bac2 S	317	})
	318
	319	self._sort_formats(formats)
	320
	321	return {
	322	'id': live_id,
	323	'display_id': display_id,
	324	'title': self._live_title(metadata['titel']),
	325	'description': metadata['info'],
	326	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	327	'formats': formats,
	328	'is_live': True,
	329	}
	330
	331
171ca612 S	332	class NPORadioIE(InfoExtractor):
	333	IE_NAME = 'npo.nl:radio'
	334	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
	335
	336	_TEST = {
	337	'url': 'http://www.npo.nl/radio/radio-1',
	338	'info_dict': {
	339	'id': 'radio-1',
	340	'ext': 'mp3',
	341	'title': 're:^NPO Radio 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	342	'is_live': True,
	343	},
	344	'params': {
	345	'skip_download': True,
	346	}
	347	}
	348
	349	@staticmethod
	350	def _html_get_attribute_regex(attribute):
	351	return r'{0}\s=\s\'([^\']+)\''.format(attribute)
	352
	353	def _real_extract(self, url):
	354	video_id = self._match_id(url)
	355
	356	webpage = self._download_webpage(url, video_id)
	357
	358	title = self._html_search_regex(
	359	self._html_get_attribute_regex('data-channel'), webpage, 'title')
	360
	361	stream = self._parse_json(
	362	self._html_search_regex(self._html_get_attribute_regex('data-streams'), webpage, 'data-streams'),
	363	video_id)
	364
	365	codec = stream.get('codec')
	366
	367	return {
	368	'id': video_id,
	369	'url': stream['url'],
	370	'title': self._live_title(title),
	371	'acodec': codec,
	372	'ext': codec,
	373	'is_live': True,
	374	}
	375
	376
	377	class NPORadioFragmentIE(InfoExtractor):
	378	IE_NAME = 'npo.nl:radio:fragment'
	379	_VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/[^/]+/fragment/(?P<id>\d+)'
	380
	381	_TEST = {
	382	'url': 'http://www.npo.nl/radio/radio-5/fragment/174356',
	383	'md5': 'dd8cc470dad764d0fdc70a9a1e2d18c2',
	384	'info_dict': {
	385	'id': '174356',
	386	'ext': 'mp3',
	387	'title': 'Jubileumconcert Willeke Alberti',
	388	},
	389	}
	390
	391	def _real_extract(self, url):
	392	audio_id = self._match_id(url)
	393
	394	webpage = self._download_webpage(url, audio_id)
	395
396	title = self._html_search_regex(
397	r'href="/radio/[^/]+/fragment/%s" title="([^"]+)"' % audio_id,
398	webpage, 'title')
399
400	audio_url = self._search_regex(
401	r"data-streams='([^']+)'", webpage, 'audio url')
402
403	return {
404	'id': audio_id,
405	'url': audio_url,
406	'title': title,
407	}
408
409
89f26028	410	class SchoolTVIE(InfoExtractor):
	411	IE_NAME = 'schooltv'
	412	_VALID_URL = r'https?://(?:www\.)?schooltv\.nl/video/(?P<id>[^/?#&]+)'
	413
	414	_TEST = {
	415	'url': 'http://www.schooltv.nl/video/ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam/',
	416	'info_dict': {
	417	'id': 'WO_NTR_429477',
	418	'display_id': 'ademhaling-de-hele-dag-haal-je-adem-maar-wat-gebeurt-er-dan-eigenlijk-in-je-lichaam',
	419	'title': 'Ademhaling: De hele dag haal je adem. Maar wat gebeurt er dan eigenlijk in je lichaam?',
	420	'ext': 'mp4',
	421	'description': 'md5:abfa0ff690adb73fd0297fd033aaa631'
	422	},
	423	'params': {
	424	# Skip because of m3u8 download
	425	'skip_download': True
	426	}
	427	}
	428
	429	def _real_extract(self, url):
	430	display_id = self._match_id(url)
	431	webpage = self._download_webpage(url, display_id)
673fb82e	432	video_id = self._search_regex(
490b7557	433	r'data-mid=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video_id', group='id')
89f26028	434	return {
	435	'_type': 'url_transparent',
	436	'ie_key': 'NPO',
	437	'url': 'npo:%s' % video_id,
	438	'display_id': display_id
	439	}
	440
	441
a3d8b381 S	442	class NPOPlaylistBaseIE(NPOIE):
	443	def _real_extract(self, url):
	444	playlist_id = self._match_id(url)
	445
	446	webpage = self._download_webpage(url, playlist_id)
	447
	448	entries = [
	449	self.url_result('npo:%s' % video_id if not video_id.startswith('http') else video_id)
5742c18b	450	for video_id in orderedSet(re.findall(self._PLAYLIST_ENTRY_RE, webpage))
a3d8b381 S	451	]
	452
	453	playlist_title = self._html_search_regex(
	454	self._PLAYLIST_TITLE_RE, webpage, 'playlist title',
	455	default=None) or self._og_search_title(webpage)
	456
	457	return self.playlist_result(entries, playlist_id, playlist_title)
	458
	459
	460	class VPROIE(NPOPlaylistBaseIE):
464e7924	461	IE_NAME = 'vpro'
611ac379	462	_VALID_URL = r'https?://(?:www\.)?(?:tegenlicht\.)?vpro\.nl/(?:[^/]+/){2,}(?P<id>[^/]+)\.html'
ddb19772	463	_PLAYLIST_TITLE_RE = r'<h1[^>]+class=["\'].?\bmedia-platform-title\b.?["\'][^>]*>([^<]+)'
a3d8b381	464	_PLAYLIST_ENTRY_RE = r'data-media-id="([^"]+)"'
d0df9292 JMF	465
	466	_TESTS = [
	467	{
	468	'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
	469	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	470	'info_dict': {
	471	'id': 'VPWON_1169289',
	472	'ext': 'm4v',
5ba761eb	473	'title': 'De toekomst komt uit Afrika',
611ac379	474	'description': 'md5:52cf4eefbc96fffcbdc06d024147abea',
d0df9292 JMF	475	'upload_date': '20130225',
d0df9292 JMF	476	},
ddb19772	477	'skip': 'Video gone',
d0df9292	478	},
611ac379 S	479	{
	480	'url': 'http://www.vpro.nl/programmas/2doc/2015/sergio-herman.html',
	481	'info_dict': {
	482	'id': 'sergio-herman',
ddb19772	483	'title': 'sergio herman: fucking perfect',
611ac379 S	484	},
611ac379 S	485	'playlist_count': 2,
574f42d7 S	486	},
	487	{
	488	# playlist with youtube embed
	489	'url': 'http://www.vpro.nl/programmas/2doc/2015/education-education.html',
	490	'info_dict': {
	491	'id': 'education-education',
ddb19772	492	'title': 'education education',
574f42d7 S	493	},
574f42d7 S	494	'playlist_count': 2,
611ac379	495	}
d0df9292 JMF	496	]
d0df9292 JMF	497
03f32a7e	498
a3d8b381	499	class WNLIE(NPOPlaylistBaseIE):
03f32a7e	500	_VALID_URL = r'https?://(?:www\.)?omroepwnl\.nl/video/detail/(?P<id>[^/]+)__\d+'
a3d8b381 S	501	_PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class="subject"[^>]*>(.+?)</h1>'
a3d8b381 S	502	_PLAYLIST_ENTRY_RE = r'<a[^>]+href="([^"]+)"[^>]+class="js-mid"[^>]*>Deel \d+'
03f32a7e	503
a3d8b381	504	_TESTS = [{
03f32a7e S	505	'url': 'http://www.omroepwnl.nl/video/detail/vandaag-de-dag-6-mei__060515',
	506	'info_dict': {
	507	'id': 'vandaag-de-dag-6-mei',
	508	'title': 'Vandaag de Dag 6 mei',
	509	},
	510	'playlist_count': 4,
a3d8b381	511	}]
5742c18b S	512
	513
	514	class AndereTijdenIE(NPOPlaylistBaseIE):
	515	_VALID_URL = r'https?://(?:www\.)?anderetijden\.nl/programma/(?:[^/]+/)+(?P<id>[^/?#&]+)'
	516	_PLAYLIST_TITLE_RE = r'(?s)<h1[^>]+class=["\'].?\bpage-title\b.?["\'][^>]*>(.+?)</h1>'
	517	_PLAYLIST_ENTRY_RE = r'<figure[^>]+class=["\']episode-container episode-page["\'][^>]+data-prid=["\'](.+?)["\']'
	518
	519	_TESTS = [{
	520	'url': 'http://anderetijden.nl/programma/1/Andere-Tijden/aflevering/676/Duitse-soldaten-over-de-Slag-bij-Arnhem',
	521	'info_dict': {
	522	'id': 'Duitse-soldaten-over-de-Slag-bij-Arnhem',
	523	'title': 'Duitse soldaten over de Slag bij Arnhem',
	524	},
	525	'playlist_count': 3,
	526	}]