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

# coding: utf-8
from __future__ import unicode_literals

import random
import re

from .common import InfoExtractor
from ..compat import compat_urllib_parse_unquote
from ..utils import (
    ExtractorError,
    int_or_none,
    parse_age_limit,
    parse_duration,
)


class NRKBaseIE(InfoExtractor):
    _faked_ip = None

    def _download_webpage_handle(self, *args, **kwargs):
        # NRK checks X-Forwarded-For HTTP header in order to figure out the
        # origin of the client behind proxy. This allows to bypass geo
        # restriction by faking this header's value to some Norway IP.
        # We will do so once we encounter any geo restriction error.
        if self._faked_ip:
            # NB: str is intentional
            kwargs.setdefault(str('headers'), {})['X-Forwarded-For'] = self._faked_ip
        return super(NRKBaseIE, self)._download_webpage_handle(*args, **kwargs)

    def _fake_ip(self):
        # Use fake IP from 37.191.128.0/17 in order to workaround geo
        # restriction
        def octet(lb=0, ub=255):
            return random.randint(lb, ub)
        self._faked_ip = '37.191.%d.%d' % (octet(128), octet())

    def _real_extract(self, url):
        video_id = self._match_id(url)

        data = self._download_json(
            'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
            video_id, 'Downloading mediaelement JSON')

        title = data.get('fullTitle') or data.get('mainTitle') or data['title']
        video_id = data.get('id') or video_id

        http_headers = {'X-Forwarded-For': self._faked_ip} if self._faked_ip else {}

        entries = []

        conviva = data.get('convivaStatistics') or {}
        live = (data.get('mediaElementType') == 'Live' or
                data.get('isLive') is True or conviva.get('isLive'))

        def make_title(t):
            return self._live_title(t) if live else t

        media_assets = data.get('mediaAssets')
        if media_assets and isinstance(media_assets, list):
            def video_id_and_title(idx):
                return ((video_id, title) if len(media_assets) == 1
                        else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
            for num, asset in enumerate(media_assets, 1):
                asset_url = asset.get('url')
                if not asset_url:
                    continue
                formats = self._extract_akamai_formats(asset_url, video_id)
                if not formats:
                    continue
                self._sort_formats(formats)

                # Some f4m streams may not work with hdcore in fragments' URLs
                for f in formats:
                    extra_param = f.get('extra_param_to_segment_url')
                    if extra_param and 'hdcore' in extra_param:
                        del f['extra_param_to_segment_url']

                entry_id, entry_title = video_id_and_title(num)
                duration = parse_duration(asset.get('duration'))
                subtitles = {}
                for subtitle in ('webVtt', 'timedText'):
                    subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
                    if subtitle_url:
                        subtitles.setdefault('no', []).append({
                            'url': compat_urllib_parse_unquote(subtitle_url)
                        })
                entries.append({
                    'id': asset.get('carrierId') or entry_id,
                    'title': make_title(entry_title),
                    'duration': duration,
                    'subtitles': subtitles,
                    'formats': formats,
                    'http_headers': http_headers,
                })

        if not entries:
            media_url = data.get('mediaUrl')
            if media_url:
                formats = self._extract_akamai_formats(media_url, video_id)
                self._sort_formats(formats)
                duration = parse_duration(data.get('duration'))
                entries = [{
                    'id': video_id,
                    'title': make_title(title),
                    'duration': duration,
                    'formats': formats,
                }]

        if not entries:
            message_type = data.get('messageType', '')
            # Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
            if 'IsGeoBlocked' in message_type and not self._faked_ip:
                self.report_warning(
                    'Video is geo restricted, trying to fake IP')
                self._fake_ip()
                return self._real_extract(url)

            MESSAGES = {
                'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
                'ProgramRightsHasExpired': 'Programmet har gått ut',
                'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
            }
            raise ExtractorError(
                '%s said: %s' % (self.IE_NAME, MESSAGES.get(
                    message_type, message_type)),
                expected=True)

        series = conviva.get('seriesName') or data.get('seriesTitle')
        episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')

        thumbnails = None
        images = data.get('images')
        if images and isinstance(images, dict):
            web_images = images.get('webImages')
            if isinstance(web_images, list):
                thumbnails = [{
                    'url': image['imageUrl'],
                    'width': int_or_none(image.get('width')),
                    'height': int_or_none(image.get('height')),
                } for image in web_images if image.get('imageUrl')]

        description = data.get('description')

        common_info = {
            'description': description,
            'series': series,
            'episode': episode,
            'age_limit': parse_age_limit(data.get('legalAge')),
            'thumbnails': thumbnails,
        }

        vcodec = 'none' if data.get('mediaType') == 'Audio' else None

        # TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged

        for entry in entries:
            entry.update(common_info)
            for f in entry['formats']:
                f['vcodec'] = vcodec

        return self.playlist_result(entries, video_id, title, description)


class NRKIE(NRKBaseIE):
    _VALID_URL = r'''(?x)
                        (?:
                            nrk:|
                            https?://
                                (?:
                                    (?:www\.)?nrk\.no/video/PS\*|
                                    v8-psapi\.nrk\.no/mediaelement/
                                )
                            )
                            (?P<id>[^/?#&]+)
                        '''
    _API_HOST = 'v8.psapi.nrk.no'
    _TESTS = [{
        # video
        'url': 'http://www.nrk.no/video/PS*150533',
        'md5': '2f7f6eeb2aacdd99885f355428715cfa',
        'info_dict': {
            'id': '150533',
            'ext': 'mp4',
            'title': 'Dompap og andre fugler i Piip-Show',
            'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
            'duration': 263,
        }
    }, {
        # audio
        'url': 'http://www.nrk.no/video/PS*154915',
        # MD5 is unstable
        'info_dict': {
            'id': '154915',
            'ext': 'flv',
            'title': 'Slik høres internett ut når du er blind',
            'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
            'duration': 20,
        }
    }, {
        'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
        'only_matching': True,
    }, {
        'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
        'only_matching': True,
    }]


class NRKTVIE(NRKBaseIE):
    IE_DESC = 'NRK TV and NRK Radio'
    _EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
    _VALID_URL = r'''(?x)
                        https?://
                            (?:tv|radio)\.nrk(?:super)?\.no/
                            (?:serie/[^/]+|program)/
                            (?![Ee]pisodes)%s
                            (?:/\d{2}-\d{2}-\d{4})?
                            (?:\#del=(?P<part_id>\d+))?
                    ''' % _EPISODE_RE
    _API_HOST = 'psapi-we.nrk.no'

    _TESTS = [{
        'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
        'md5': '4e9ca6629f09e588ed240fb11619922a',
        'info_dict': {
            'id': 'MUHH48000314AA',
            'ext': 'mp4',
            'title': '20 spørsmål 23.05.2014',
            'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
            'duration': 1741,
        },
    }, {
        'url': 'https://tv.nrk.no/program/mdfp15000514',
        'md5': '43d0be26663d380603a9cf0c24366531',
        'info_dict': {
            'id': 'MDFP15000514CA',
            'ext': 'mp4',
            'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
            'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
            'duration': 4605,
        },
    }, {
        # single playlist video
        'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
        'md5': 'adbd1dbd813edaf532b0a253780719c2',
        'info_dict': {
            'id': 'MSPO40010515-part2',
            'ext': 'flv',
            'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
            'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
        },
        'skip': 'Only works from Norway',
    }, {
        'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
        'playlist': [{
            'md5': '9480285eff92d64f06e02a5367970a7a',
            'info_dict': {
                'id': 'MSPO40010515-part1',
                'ext': 'flv',
                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
            },
        }, {
            'md5': 'adbd1dbd813edaf532b0a253780719c2',
            'info_dict': {
                'id': 'MSPO40010515-part2',
                'ext': 'flv',
                'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
                'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
            },
        }],
        'info_dict': {
            'id': 'MSPO40010515',
            'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
            'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
            'duration': 6947.52,
        },
        'skip': 'Only works from Norway',
    }, {
        'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
        'only_matching': True,
    }]


class NRKTVDirekteIE(NRKTVIE):
    IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
    _VALID_URL = r'https?://(?:tv|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'

    _TESTS = [{
        'url': 'https://tv.nrk.no/direkte/nrk1',
        'only_matching': True,
    }, {
        'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
        'only_matching': True,
    }]


class NRKPlaylistBaseIE(InfoExtractor):
    def _extract_description(self, webpage):
        pass

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

        webpage = self._download_webpage(url, playlist_id)

        entries = [
            self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
            for video_id in re.findall(self._ITEM_RE, webpage)
        ]

        playlist_title = self. _extract_title(webpage)
        playlist_description = self._extract_description(webpage)

        return self.playlist_result(
            entries, playlist_id, playlist_title, playlist_description)


class NRKPlaylistIE(NRKPlaylistBaseIE):
    _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<id>[^/]+)'
    _ITEM_RE = r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"'
    _TESTS = [{
        'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
        'info_dict': {
            'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
            'title': 'Gjenopplev den historiske solformørkelsen',
            'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
        },
        'playlist_count': 2,
    }, {
        'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
        'info_dict': {
            'id': 'rivertonprisen-til-karin-fossum-1.12266449',
            'title': 'Rivertonprisen til Karin Fossum',
            'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
        },
        'playlist_count': 5,
    }]

    def _extract_title(self, webpage):
        return self._og_search_title(webpage, fatal=False)

    def _extract_description(self, webpage):
        return self._og_search_description(webpage)


class NRKTVEpisodesIE(NRKPlaylistBaseIE):
    _VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
    _ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
    _TESTS = [{
        'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
        'info_dict': {
            'id': '69031',
            'title': 'Nytt på nytt, sesong: 201210',
        },
        'playlist_count': 4,
    }]

    def _extract_title(self, webpage):
        return self._html_search_regex(
            r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)


class NRKSkoleIE(InfoExtractor):
    IE_DESC = 'NRK Skole'
    _VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'

    _TESTS = [{
        'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
        'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
        'info_dict': {
            'id': '6021',
            'ext': 'mp4',
            'title': 'Genetikk og eneggede tvillinger',
            'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
            'duration': 399,
        },
    }, {
        'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        video_id = self._match_id(url)

        webpage = self._download_webpage(
            'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
            video_id)

        nrk_id = self._parse_json(
            self._search_regex(
                r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
                webpage, 'application json'),
            video_id)['activeMedia']['psId']

        return self.url_result('nrk:%s' % nrk_id)
Commit	Line	Data
dcdb292f	1	# coding: utf-8
d2176c80 S	2	from __future__ import unicode_literals
d2176c80 S	3
754e6c83	4	import random
d2176c80 S	5	import re
	6
	7	from .common import InfoExtractor
d8d540cf	8	from ..compat import compat_urllib_parse_unquote
dfb2e1a3 S	9	from ..utils import (
dfb2e1a3 S	10	ExtractorError,
d8d540cf S	11	int_or_none,
d8d540cf S	12	parse_age_limit,
76bfaf6d	13	parse_duration,
dfb2e1a3	14	)
d2176c80 S	15
d2176c80 S	16
d8d540cf	17	class NRKBaseIE(InfoExtractor):
754e6c83 S	18	_faked_ip = None
754e6c83 S	19
7e08e2ca	20	def _download_webpage_handle(self, args, *kwargs):
754e6c83 S	21	# NRK checks X-Forwarded-For HTTP header in order to figure out the
	22	# origin of the client behind proxy. This allows to bypass geo
	23	# restriction by faking this header's value to some Norway IP.
	24	# We will do so once we encounter any geo restriction error.
	25	if self._faked_ip:
7e08e2ca S	26	# NB: str is intentional
	27	kwargs.setdefault(str('headers'), {})['X-Forwarded-For'] = self._faked_ip
	28	return super(NRKBaseIE, self)._download_webpage_handle(args, *kwargs)
754e6c83 S	29
	30	def _fake_ip(self):
	31	# Use fake IP from 37.191.128.0/17 in order to workaround geo
	32	# restriction
	33	def octet(lb=0, ub=255):
	34	return random.randint(lb, ub)
	35	self._faked_ip = '37.191.%d.%d' % (octet(128), octet())
	36
d2176c80	37	def _real_extract(self, url):
4e6a2286	38	video_id = self._match_id(url)
d2176c80 S	39
d2176c80 S	40	data = self._download_json(
d8d540cf S	41	'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
	42	video_id, 'Downloading mediaelement JSON')
	43
	44	title = data.get('fullTitle') or data.get('mainTitle') or data['title']
	45	video_id = data.get('id') or video_id
	46
7e08e2ca S	47	http_headers = {'X-Forwarded-For': self._faked_ip} if self._faked_ip else {}
7e08e2ca S	48
d8d540cf S	49	entries = []
d8d540cf S	50
c80db5d3 S	51	conviva = data.get('convivaStatistics') or {}
	52	live = (data.get('mediaElementType') == 'Live' or
	53	data.get('isLive') is True or conviva.get('isLive'))
	54
	55	def make_title(t):
	56	return self._live_title(t) if live else t
	57
d8d540cf S	58	media_assets = data.get('mediaAssets')
	59	if media_assets and isinstance(media_assets, list):
	60	def video_id_and_title(idx):
	61	return ((video_id, title) if len(media_assets) == 1
	62	else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
	63	for num, asset in enumerate(media_assets, 1):
	64	asset_url = asset.get('url')
	65	if not asset_url:
	66	continue
ad316425	67	formats = self._extract_akamai_formats(asset_url, video_id)
d8d540cf S	68	if not formats:
	69	continue
	70	self._sort_formats(formats)
c80db5d3 S	71
	72	# Some f4m streams may not work with hdcore in fragments' URLs
	73	for f in formats:
	74	extra_param = f.get('extra_param_to_segment_url')
	75	if extra_param and 'hdcore' in extra_param:
	76	del f['extra_param_to_segment_url']
	77
d8d540cf S	78	entry_id, entry_title = video_id_and_title(num)
	79	duration = parse_duration(asset.get('duration'))
	80	subtitles = {}
	81	for subtitle in ('webVtt', 'timedText'):
	82	subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
	83	if subtitle_url:
c8602b2f S	84	subtitles.setdefault('no', []).append({
	85	'url': compat_urllib_parse_unquote(subtitle_url)
	86	})
d8d540cf S	87	entries.append({
d8d540cf S	88	'id': asset.get('carrierId') or entry_id,
c80db5d3	89	'title': make_title(entry_title),
d8d540cf S	90	'duration': duration,
	91	'subtitles': subtitles,
	92	'formats': formats,
7e08e2ca	93	'http_headers': http_headers,
d8d540cf	94	})
d2176c80	95
d8d540cf S	96	if not entries:
	97	media_url = data.get('mediaUrl')
	98	if media_url:
ad316425	99	formats = self._extract_akamai_formats(media_url, video_id)
d8d540cf S	100	self._sort_formats(formats)
	101	duration = parse_duration(data.get('duration'))
	102	entries = [{
	103	'id': video_id,
c80db5d3	104	'title': make_title(title),
d8d540cf S	105	'duration': duration,
	106	'formats': formats,
	107	}]
d2176c80	108
d8d540cf	109	if not entries:
50913b82 S	110	message_type = data.get('messageType', '')
	111	# Can be ProgramIsGeoBlocked or ChannelIsGeoBlocked*
	112	if 'IsGeoBlocked' in message_type and not self._faked_ip:
754e6c83 S	113	self.report_warning(
	114	'Video is geo restricted, trying to fake IP')
	115	self._fake_ip()
	116	return self._real_extract(url)
	117
	118	MESSAGES = {
	119	'ProgramRightsAreNotReady': 'Du kan dessverre ikke se eller høre programmet',
	120	'ProgramRightsHasExpired': 'Programmet har gått ut',
	121	'ProgramIsGeoBlocked': 'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
	122	}
	123	raise ExtractorError(
	124	'%s said: %s' % (self.IE_NAME, MESSAGES.get(
	125	message_type, message_type)),
	126	expected=True)
874ae035	127
d8d540cf S	128	series = conviva.get('seriesName') or data.get('seriesTitle')
d8d540cf S	129	episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
393d9fc6	130
d8d540cf	131	thumbnails = None
d2176c80	132	images = data.get('images')
d8d540cf S	133	if images and isinstance(images, dict):
	134	web_images = images.get('webImages')
	135	if isinstance(web_images, list):
	136	thumbnails = [{
	137	'url': image['imageUrl'],
	138	'width': int_or_none(image.get('width')),
	139	'height': int_or_none(image.get('height')),
	140	} for image in web_images if image.get('imageUrl')]
	141
	142	description = data.get('description')
	143
	144	common_info = {
	145	'description': description,
	146	'series': series,
	147	'episode': episode,
	148	'age_limit': parse_age_limit(data.get('legalAge')),
	149	'thumbnails': thumbnails,
dfb2e1a3 S	150	}
dfb2e1a3 S	151
d8d540cf S	152	vcodec = 'none' if data.get('mediaType') == 'Audio' else None
	153
	154	# TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
	155
	156	for entry in entries:
	157	entry.update(common_info)
	158	for f in entry['formats']:
	159	f['vcodec'] = vcodec
	160
	161	return self.playlist_result(entries, video_id, title, description)
	162
	163
	164	class NRKIE(NRKBaseIE):
853a71b6 S	165	_VALID_URL = r'''(?x)
	166	(?:
	167	nrk:\|
	168	https?://
	169	(?:
	170	(?:www\.)?nrk\.no/video/PS\*\|
	171	v8-psapi\.nrk\.no/mediaelement/
	172	)
	173	)
	174	(?P<id>[^/?#&]+)
	175	'''
d8d540cf S	176	_API_HOST = 'v8.psapi.nrk.no'
	177	_TESTS = [{
	178	# video
	179	'url': 'http://www.nrk.no/video/PS*150533',
18cf6381	180	'md5': '2f7f6eeb2aacdd99885f355428715cfa',
d8d540cf S	181	'info_dict': {
d8d540cf S	182	'id': '150533',
18cf6381	183	'ext': 'mp4',
d8d540cf S	184	'title': 'Dompap og andre fugler i Piip-Show',
	185	'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
	186	'duration': 263,
	187	}
	188	}, {
	189	# audio
	190	'url': 'http://www.nrk.no/video/PS*154915',
	191	# MD5 is unstable
	192	'info_dict': {
	193	'id': '154915',
	194	'ext': 'flv',
	195	'title': 'Slik høres internett ut når du er blind',
	196	'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
	197	'duration': 20,
	198	}
e2628fb6 S	199	}, {
	200	'url': 'nrk:ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
	201	'only_matching': True,
853a71b6 S	202	}, {
	203	'url': 'https://v8-psapi.nrk.no/mediaelement/ecc1b952-96dc-4a98-81b9-5296dc7a98d9',
	204	'only_matching': True,
d8d540cf S	205	}]
	206
	207
	208	class NRKTVIE(NRKBaseIE):
	209	IE_DESC = 'NRK TV and NRK Radio'
966815e1 S	210	_EPISODE_RE = r'(?P<id>[a-zA-Z]{4}\d{8})'
	211	_VALID_URL = r'''(?x)
	212	https?://
	213	(?:tv\|radio)\.nrk(?:super)?\.no/
	214	(?:serie/[^/]+\|program)/
	215	(?![Ee]pisodes)%s
	216	(?:/\d{2}-\d{2}-\d{4})?
	217	(?:\#del=(?P<part_id>\d+))?
	218	''' % _EPISODE_RE
d8d540cf S	219	_API_HOST = 'psapi-we.nrk.no'
	220
	221	_TESTS = [{
	222	'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
18cf6381	223	'md5': '4e9ca6629f09e588ed240fb11619922a',
d8d540cf	224	'info_dict': {
18cf6381	225	'id': 'MUHH48000314AA',
d8d540cf	226	'ext': 'mp4',
18cf6381	227	'title': '20 spørsmål 23.05.2014',
d8d540cf	228	'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
4e790117	229	'duration': 1741,
d8d540cf	230	},
d8d540cf S	231	}, {
d8d540cf S	232	'url': 'https://tv.nrk.no/program/mdfp15000514',
18cf6381	233	'md5': '43d0be26663d380603a9cf0c24366531',
d8d540cf	234	'info_dict': {
18cf6381	235	'id': 'MDFP15000514CA',
d8d540cf	236	'ext': 'mp4',
18cf6381	237	'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
18cf6381	238	'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
4e790117	239	'duration': 4605,
d8d540cf	240	},
d8d540cf S	241	}, {
	242	# single playlist video
	243	'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
	244	'md5': 'adbd1dbd813edaf532b0a253780719c2',
	245	'info_dict': {
	246	'id': 'MSPO40010515-part2',
	247	'ext': 'flv',
	248	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
	249	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	250	},
	251	'skip': 'Only works from Norway',
	252	}, {
	253	'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
	254	'playlist': [{
	255	'md5': '9480285eff92d64f06e02a5367970a7a',
	256	'info_dict': {
	257	'id': 'MSPO40010515-part1',
	258	'ext': 'flv',
	259	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
	260	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	261	},
	262	}, {
	263	'md5': 'adbd1dbd813edaf532b0a253780719c2',
	264	'info_dict': {
	265	'id': 'MSPO40010515-part2',
	266	'ext': 'flv',
	267	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
	268	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	269	},
	270	}],
	271	'info_dict': {
	272	'id': 'MSPO40010515',
	273	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
	274	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	275	'duration': 6947.52,
	276	},
	277	'skip': 'Only works from Norway',
	278	}, {
	279	'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
	280	'only_matching': True,
	281	}]
	282
dfb2e1a3	283
c80db5d3 S	284	class NRKTVDirekteIE(NRKTVIE):
	285	IE_DESC = 'NRK TV Direkte and NRK Radio Direkte'
	286	_VALID_URL = r'https?://(?:tv\|radio)\.nrk\.no/direkte/(?P<id>[^/?#&]+)'
	287
	288	_TESTS = [{
	289	'url': 'https://tv.nrk.no/direkte/nrk1',
	290	'only_matching': True,
	291	}, {
	292	'url': 'https://radio.nrk.no/direkte/p1_oslo_akershus',
	293	'only_matching': True,
	294	}]
	295
	296
966815e1 S	297	class NRKPlaylistBaseIE(InfoExtractor):
	298	def _extract_description(self, webpage):
	299	pass
	300
	301	def _real_extract(self, url):
	302	playlist_id = self._match_id(url)
	303
	304	webpage = self._download_webpage(url, playlist_id)
	305
	306	entries = [
	307	self.url_result('nrk:%s' % video_id, NRKIE.ie_key())
	308	for video_id in re.findall(self._ITEM_RE, webpage)
	309	]
	310
	311	playlist_title = self. _extract_title(webpage)
	312	playlist_description = self._extract_description(webpage)
	313
	314	return self.playlist_result(
	315	entries, playlist_id, playlist_title, playlist_description)
faa1b5c2	316
966815e1 S	317
	318	class NRKPlaylistIE(NRKPlaylistBaseIE):
	319	_VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video\|skole)(?:[^/]+/)+(?P<id>[^/]+)'
	320	_ITEM_RE = r'class="[^"]\brich\b[^"]"[^>]+data-video-id="([^"]+)"'
a0914154	321	_TESTS = [{
faa1b5c2 S	322	'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
	323	'info_dict': {
	324	'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
	325	'title': 'Gjenopplev den historiske solformørkelsen',
	326	'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
	327	},
a0914154 S	328	'playlist_count': 2,
	329	}, {
	330	'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
	331	'info_dict': {
	332	'id': 'rivertonprisen-til-karin-fossum-1.12266449',
	333	'title': 'Rivertonprisen til Karin Fossum',
	334	'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
	335	},
	336	'playlist_count': 5,
	337	}]
faa1b5c2	338
966815e1 S	339	def _extract_title(self, webpage):
966815e1 S	340	return self._og_search_title(webpage, fatal=False)
faa1b5c2	341
966815e1 S	342	def _extract_description(self, webpage):
966815e1 S	343	return self._og_search_description(webpage)
faa1b5c2	344
faa1b5c2	345
966815e1 S	346	class NRKTVEpisodesIE(NRKPlaylistBaseIE):
	347	_VALID_URL = r'https?://tv\.nrk\.no/program/[Ee]pisodes/[^/]+/(?P<id>\d+)'
	348	_ITEM_RE = r'data-episode=["\']%s' % NRKTVIE._EPISODE_RE
	349	_TESTS = [{
	350	'url': 'https://tv.nrk.no/program/episodes/nytt-paa-nytt/69031',
	351	'info_dict': {
	352	'id': '69031',
	353	'title': 'Nytt på nytt, sesong: 201210',
	354	},
	355	'playlist_count': 4,
	356	}]
faa1b5c2	357
966815e1 S	358	def _extract_title(self, webpage):
	359	return self._html_search_regex(
	360	r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
faa1b5c2 S	361
faa1b5c2 S	362
3099b312 S	363	class NRKSkoleIE(InfoExtractor):
3099b312 S	364	IE_DESC = 'NRK Skole'
971e3b75	365	_VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
3099b312 S	366
3099b312 S	367	_TESTS = [{
971e3b75 S	368	'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
971e3b75 S	369	'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
3099b312 S	370	'info_dict': {
3099b312 S	371	'id': '6021',
971e3b75	372	'ext': 'mp4',
3099b312 S	373	'title': 'Genetikk og eneggede tvillinger',
	374	'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
	375	'duration': 399,
	376	},
	377	}, {
971e3b75	378	'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
61140904	379	'only_matching': True,
3099b312 S	380	}]
	381
	382	def _real_extract(self, url):
971e3b75 S	383	video_id = self._match_id(url)
	384
	385	webpage = self._download_webpage(
	386	'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
	387	video_id)
3099b312	388
971e3b75 S	389	nrk_id = self._parse_json(
	390	self._search_regex(
	391	r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
	392	webpage, 'application json'),
	393	video_id)['activeMedia']['psId']
3099b312	394
3099b312	395	return self.url_result('nrk:%s' % nrk_id)