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

# coding: utf-8
from __future__ import unicode_literals

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):
    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

        entries = []

        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)
                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': entry_title,
                    'duration': duration,
                    'subtitles': subtitles,
                    'formats': formats,
                })

        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': title,
                    'duration': duration,
                    'formats': formats,
                }]

        if not entries:
            if data.get('usageRights', {}).get('isGeoBlocked'):
                raise ExtractorError(
                    'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
                    expected=True)

        conviva = data.get('convivaStatistics') or {}
        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'(?:nrk:|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
    _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,
        }
    }]


class NRKTVIE(NRKBaseIE):
    IE_DESC = 'NRK TV and NRK Radio'
    _VALID_URL = r'https?://(?:tv|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
    _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 NRKPlaylistIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video|skole)(?:[^/]+/)+(?P<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 _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, 'NRK')
            for video_id in re.findall(
                r'class="[^"]*\brich\b[^"]*"[^>]+data-video-id="([^"]+)"',
                webpage)
        ]

        playlist_title = self._og_search_title(webpage)
        playlist_description = self._og_search_description(webpage)

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


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
	3
	4	import re
	5
	6	from .common import InfoExtractor
d8d540cf	7	from ..compat import compat_urllib_parse_unquote
dfb2e1a3 S	8	from ..utils import (
dfb2e1a3 S	9	ExtractorError,
d8d540cf S	10	int_or_none,
d8d540cf S	11	parse_age_limit,
76bfaf6d	12	parse_duration,
dfb2e1a3	13	)
d2176c80 S	14
d2176c80 S	15
d8d540cf	16	class NRKBaseIE(InfoExtractor):
d2176c80	17	def _real_extract(self, url):
4e6a2286	18	video_id = self._match_id(url)
d2176c80 S	19
d2176c80 S	20	data = self._download_json(
d8d540cf S	21	'http://%s/mediaelement/%s' % (self._API_HOST, video_id),
	22	video_id, 'Downloading mediaelement JSON')
	23
	24	title = data.get('fullTitle') or data.get('mainTitle') or data['title']
	25	video_id = data.get('id') or video_id
	26
	27	entries = []
	28
	29	media_assets = data.get('mediaAssets')
	30	if media_assets and isinstance(media_assets, list):
	31	def video_id_and_title(idx):
	32	return ((video_id, title) if len(media_assets) == 1
	33	else ('%s-%d' % (video_id, idx), '%s (Part %d)' % (title, idx)))
	34	for num, asset in enumerate(media_assets, 1):
	35	asset_url = asset.get('url')
	36	if not asset_url:
	37	continue
ad316425	38	formats = self._extract_akamai_formats(asset_url, video_id)
d8d540cf S	39	if not formats:
	40	continue
	41	self._sort_formats(formats)
	42	entry_id, entry_title = video_id_and_title(num)
	43	duration = parse_duration(asset.get('duration'))
	44	subtitles = {}
	45	for subtitle in ('webVtt', 'timedText'):
	46	subtitle_url = asset.get('%sSubtitlesUrl' % subtitle)
	47	if subtitle_url:
c8602b2f S	48	subtitles.setdefault('no', []).append({
	49	'url': compat_urllib_parse_unquote(subtitle_url)
	50	})
d8d540cf S	51	entries.append({
	52	'id': asset.get('carrierId') or entry_id,
	53	'title': entry_title,
	54	'duration': duration,
	55	'subtitles': subtitles,
	56	'formats': formats,
	57	})
d2176c80	58
d8d540cf S	59	if not entries:
	60	media_url = data.get('mediaUrl')
	61	if media_url:
ad316425	62	formats = self._extract_akamai_formats(media_url, video_id)
d8d540cf S	63	self._sort_formats(formats)
	64	duration = parse_duration(data.get('duration'))
	65	entries = [{
	66	'id': video_id,
	67	'title': title,
	68	'duration': duration,
	69	'formats': formats,
	70	}]
d2176c80	71
d8d540cf S	72	if not entries:
d8d540cf S	73	if data.get('usageRights', {}).get('isGeoBlocked'):
874ae035 S	74	raise ExtractorError(
	75	'NRK har ikke rettigheter til å vise dette programmet utenfor Norge',
	76	expected=True)
	77
d8d540cf S	78	conviva = data.get('convivaStatistics') or {}
	79	series = conviva.get('seriesName') or data.get('seriesTitle')
	80	episode = conviva.get('episodeName') or data.get('episodeNumberOrDate')
393d9fc6	81
d8d540cf	82	thumbnails = None
d2176c80	83	images = data.get('images')
d8d540cf S	84	if images and isinstance(images, dict):
	85	web_images = images.get('webImages')
	86	if isinstance(web_images, list):
	87	thumbnails = [{
	88	'url': image['imageUrl'],
	89	'width': int_or_none(image.get('width')),
	90	'height': int_or_none(image.get('height')),
	91	} for image in web_images if image.get('imageUrl')]
	92
	93	description = data.get('description')
	94
	95	common_info = {
	96	'description': description,
	97	'series': series,
	98	'episode': episode,
	99	'age_limit': parse_age_limit(data.get('legalAge')),
	100	'thumbnails': thumbnails,
dfb2e1a3 S	101	}
dfb2e1a3 S	102
d8d540cf S	103	vcodec = 'none' if data.get('mediaType') == 'Audio' else None
	104
	105	# TODO: extract chapters when https://github.com/rg3/youtube-dl/pull/9409 is merged
	106
	107	for entry in entries:
	108	entry.update(common_info)
	109	for f in entry['formats']:
	110	f['vcodec'] = vcodec
	111
	112	return self.playlist_result(entries, video_id, title, description)
	113
	114
	115	class NRKIE(NRKBaseIE):
	116	_VALID_URL = r'(?:nrk:\|https?://(?:www\.)?nrk\.no/video/PS\*)(?P<id>\d+)'
	117	_API_HOST = 'v8.psapi.nrk.no'
	118	_TESTS = [{
	119	# video
	120	'url': 'http://www.nrk.no/video/PS*150533',
18cf6381	121	'md5': '2f7f6eeb2aacdd99885f355428715cfa',
d8d540cf S	122	'info_dict': {
d8d540cf S	123	'id': '150533',
18cf6381	124	'ext': 'mp4',
d8d540cf S	125	'title': 'Dompap og andre fugler i Piip-Show',
	126	'description': 'md5:d9261ba34c43b61c812cb6b0269a5c8f',
	127	'duration': 263,
	128	}
	129	}, {
	130	# audio
	131	'url': 'http://www.nrk.no/video/PS*154915',
	132	# MD5 is unstable
	133	'info_dict': {
	134	'id': '154915',
	135	'ext': 'flv',
	136	'title': 'Slik høres internett ut når du er blind',
	137	'description': 'md5:a621f5cc1bd75c8d5104cb048c6b8568',
	138	'duration': 20,
	139	}
	140	}]
	141
	142
	143	class NRKTVIE(NRKBaseIE):
	144	IE_DESC = 'NRK TV and NRK Radio'
	145	_VALID_URL = r'https?://(?:tv\|radio)\.nrk(?:super)?\.no/(?:serie/[^/]+\|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
	146	_API_HOST = 'psapi-we.nrk.no'
	147
	148	_TESTS = [{
	149	'url': 'https://tv.nrk.no/serie/20-spoersmaal-tv/MUHH48000314/23-05-2014',
18cf6381	150	'md5': '4e9ca6629f09e588ed240fb11619922a',
d8d540cf	151	'info_dict': {
18cf6381	152	'id': 'MUHH48000314AA',
d8d540cf	153	'ext': 'mp4',
18cf6381	154	'title': '20 spørsmål 23.05.2014',
d8d540cf	155	'description': 'md5:bdea103bc35494c143c6a9acdd84887a',
4e790117	156	'duration': 1741,
d8d540cf	157	},
d8d540cf S	158	}, {
d8d540cf S	159	'url': 'https://tv.nrk.no/program/mdfp15000514',
18cf6381	160	'md5': '43d0be26663d380603a9cf0c24366531',
d8d540cf	161	'info_dict': {
18cf6381	162	'id': 'MDFP15000514CA',
d8d540cf	163	'ext': 'mp4',
18cf6381	164	'title': 'Grunnlovsjubiléet - Stor ståhei for ingenting 24.05.2014',
18cf6381	165	'description': 'md5:89290c5ccde1b3a24bb8050ab67fe1db',
4e790117	166	'duration': 4605,
d8d540cf	167	},
d8d540cf S	168	}, {
	169	# single playlist video
	170	'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015#del=2',
	171	'md5': 'adbd1dbd813edaf532b0a253780719c2',
	172	'info_dict': {
	173	'id': 'MSPO40010515-part2',
	174	'ext': 'flv',
	175	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
	176	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	177	},
	178	'skip': 'Only works from Norway',
	179	}, {
	180	'url': 'https://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
	181	'playlist': [{
	182	'md5': '9480285eff92d64f06e02a5367970a7a',
	183	'info_dict': {
	184	'id': 'MSPO40010515-part1',
	185	'ext': 'flv',
	186	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 1:2)',
	187	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	188	},
	189	}, {
	190	'md5': 'adbd1dbd813edaf532b0a253780719c2',
	191	'info_dict': {
	192	'id': 'MSPO40010515-part2',
	193	'ext': 'flv',
	194	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn 06.01.2015 (del 2:2)',
	195	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	196	},
	197	}],
	198	'info_dict': {
	199	'id': 'MSPO40010515',
	200	'title': 'Tour de Ski: Sprint fri teknikk, kvinner og menn',
	201	'description': 'md5:238b67b97a4ac7d7b4bf0edf8cc57d26',
d8d540cf S	202	'duration': 6947.52,
	203	},
	204	'skip': 'Only works from Norway',
	205	}, {
	206	'url': 'https://radio.nrk.no/serie/dagsnytt/NPUB21019315/12-07-2015#',
	207	'only_matching': True,
	208	}]
	209
dfb2e1a3	210
faa1b5c2	211	class NRKPlaylistIE(InfoExtractor):
3099b312	212	_VALID_URL = r'https?://(?:www\.)?nrk\.no/(?!video\|skole)(?:[^/]+/)+(?P<id>[^/]+)'
faa1b5c2	213
a0914154	214	_TESTS = [{
faa1b5c2 S	215	'url': 'http://www.nrk.no/troms/gjenopplev-den-historiske-solformorkelsen-1.12270763',
	216	'info_dict': {
	217	'id': 'gjenopplev-den-historiske-solformorkelsen-1.12270763',
	218	'title': 'Gjenopplev den historiske solformørkelsen',
	219	'description': 'md5:c2df8ea3bac5654a26fc2834a542feed',
	220	},
a0914154 S	221	'playlist_count': 2,
	222	}, {
	223	'url': 'http://www.nrk.no/kultur/bok/rivertonprisen-til-karin-fossum-1.12266449',
	224	'info_dict': {
	225	'id': 'rivertonprisen-til-karin-fossum-1.12266449',
	226	'title': 'Rivertonprisen til Karin Fossum',
	227	'description': 'Første kvinne på 15 år til å vinne krimlitteraturprisen.',
	228	},
	229	'playlist_count': 5,
	230	}]
faa1b5c2 S	231
	232	def _real_extract(self, url):
	233	playlist_id = self._match_id(url)
	234
	235	webpage = self._download_webpage(url, playlist_id)
	236
	237	entries = [
	238	self.url_result('nrk:%s' % video_id, 'NRK')
	239	for video_id in re.findall(
a0914154 S	240	r'class="[^"]\brich\b[^"]"[^>]+data-video-id="([^"]+)"',
a0914154 S	241	webpage)
faa1b5c2 S	242	]
	243
	244	playlist_title = self._og_search_title(webpage)
	245	playlist_description = self._og_search_description(webpage)
	246
	247	return self.playlist_result(
	248	entries, playlist_id, playlist_title, playlist_description)
	249
	250
3099b312 S	251	class NRKSkoleIE(InfoExtractor):
3099b312 S	252	IE_DESC = 'NRK Skole'
971e3b75	253	_VALID_URL = r'https?://(?:www\.)?nrk\.no/skole/?\?.*\bmediaId=(?P<id>\d+)'
3099b312 S	254
3099b312 S	255	_TESTS = [{
971e3b75 S	256	'url': 'https://www.nrk.no/skole/?page=search&q=&mediaId=14099',
971e3b75 S	257	'md5': '6bc936b01f9dd8ed45bc58b252b2d9b6',
3099b312 S	258	'info_dict': {
3099b312 S	259	'id': '6021',
971e3b75	260	'ext': 'mp4',
3099b312 S	261	'title': 'Genetikk og eneggede tvillinger',
	262	'description': 'md5:3aca25dcf38ec30f0363428d2b265f8d',
	263	'duration': 399,
	264	},
	265	}, {
971e3b75	266	'url': 'https://www.nrk.no/skole/?page=objectives&subject=naturfag&objective=K15114&mediaId=19355',
61140904	267	'only_matching': True,
3099b312 S	268	}]
	269
	270	def _real_extract(self, url):
971e3b75 S	271	video_id = self._match_id(url)
	272
	273	webpage = self._download_webpage(
	274	'https://mimir.nrk.no/plugin/1.0/static?mediaId=%s' % video_id,
	275	video_id)
3099b312	276
971e3b75 S	277	nrk_id = self._parse_json(
	278	self._search_regex(
	279	r'<script[^>]+type=["\']application/json["\'][^>]*>({.+?})</script>',
	280	webpage, 'application json'),
	281	video_id)['activeMedia']['psId']
3099b312	282
3099b312	283	return self.url_result('nrk:%s' % nrk_id)