[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,
    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')
        return self._search_regex(
            r'npoplayer\.token = "(.+?)"', token_page, 'token')


class NPOIE(NPOBaseIE):
    IE_NAME = 'npo.nl'
    _VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?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:d6476bceb17a8c103c76c3b708f05dd1',
                '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?',
            },
        },
    ]

    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/aflevering/%s' % video_id,
            video_id,
            # We have to remove the javascript callback
            transform_source=strip_jsonp,
        )

        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)

        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,
        }


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()
                if stream_type == 'ss':
                    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,
                    transform_source=strip_jsonp)
                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,
                    })

        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 TegenlichtVproIE(NPOIE):
    IE_NAME = 'tegenlicht.vpro.nl'
    _VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'

    _TESTS = [
        {
            'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
            'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
            'info_dict': {
                'id': 'VPWON_1169289',
                'ext': 'm4v',
                'title': 'Tegenlicht',
                'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
                'upload_date': '20130225',
            },
        },
    ]

    def _real_extract(self, url):
        name = url_basename(url)
        webpage = self._download_webpage(url, name)
        urn = self._html_search_meta('mediaurn', webpage)
        info_page = self._download_json(
            'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name)
        return self._get_info(info_page['mid'])
Commit	Line	Data
331ae266 JMF	1	from __future__ import unicode_literals
	2
	3	import re
	4
	5	from .common import InfoExtractor
	6	from ..utils import (
5c4a81d9	7	fix_xml_ampersands,
944a3de2	8	parse_duration,
c9ea760e	9	qualities,
609a61e3	10	strip_jsonp,
5c4a81d9	11	unified_strdate,
d0df9292	12	url_basename,
331ae266 JMF	13	)
	14
	15
04e0bac2 S	16	class NPOBaseIE(InfoExtractor):
	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')
	21	return self._search_regex(
	22	r'npoplayer\.token = "(.+?)"', token_page, 'token')
	23
	24
	25	class NPOIE(NPOBaseIE):
331ae266 JMF	26	IE_NAME = 'npo.nl'
	27	_VALID_URL = r'https?://www\.npo\.nl/[^/]+/[^/]+/(?P<id>[^/?]+)'
	28
944a3de2 S	29	_TESTS = [
	30	{
	31	'url': 'http://www.npo.nl/nieuwsuur/22-06-2014/VPWON_1220719',
	32	'md5': '4b3f9c429157ec4775f2c9cb7b911016',
	33	'info_dict': {
	34	'id': 'VPWON_1220719',
	35	'ext': 'm4v',
	36	'title': 'Nieuwsuur',
	37	'description': 'Dagelijks tussen tien en elf: nieuws, sport en achtergronden.',
	38	'upload_date': '20140622',
	39	},
331ae266	40	},
944a3de2 S	41	{
	42	'url': 'http://www.npo.nl/de-mega-mike-mega-thomas-show/27-02-2009/VARA_101191800',
	43	'md5': 'da50a5787dbfc1603c4ad80f31c5120b',
	44	'info_dict': {
	45	'id': 'VARA_101191800',
	46	'ext': 'm4v',
	47	'title': 'De Mega Mike & Mega Thomas show',
	48	'description': 'md5:3b74c97fc9d6901d5a665aac0e5400f4',
	49	'upload_date': '20090227',
	50	'duration': 2400,
	51	},
	52	},
	53	{
	54	'url': 'http://www.npo.nl/tegenlicht/25-02-2013/VPWON_1169289',
	55	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	56	'info_dict': {
	57	'id': 'VPWON_1169289',
	58	'ext': 'm4v',
	59	'title': 'Tegenlicht',
	60	'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
	61	'upload_date': '20130225',
	62	'duration': 3000,
	63	},
1ff30d7b S	64	},
	65	{
	66	'url': 'http://www.npo.nl/de-nieuwe-mens-deel-1/21-07-2010/WO_VPRO_043706',
	67	'info_dict': {
	68	'id': 'WO_VPRO_043706',
	69	'ext': 'wmv',
	70	'title': 'De nieuwe mens - Deel 1',
	71	'description': 'md5:518ae51ba1293ffb80d8d8ce90b74e4b',
	72	'duration': 4680,
	73	},
	74	'params': {
	75	# mplayer mms download
	76	'skip_download': True,
	77	}
	78	},
c85f3683 S	79	# non asf in streams
	80	{
	81	'url': 'http://www.npo.nl/hoe-gaat-europa-verder-na-parijs/10-01-2015/WO_NOS_762771',
	82	'md5': 'b3da13de374cbe2d5332a7e910bef97f',
	83	'info_dict': {
	84	'id': 'WO_NOS_762771',
	85	'ext': 'mp4',
	86	'title': 'Hoe gaat Europa verder na Parijs?',
	87	},
	88	},
944a3de2	89	]
331ae266 JMF	90
331ae266 JMF	91	def _real_extract(self, url):
04e0bac2	92	video_id = self._match_id(url)
d0df9292	93	return self._get_info(video_id)
331ae266	94
d0df9292	95	def _get_info(self, video_id):
331ae266 JMF	96	metadata = self._download_json(
	97	'http://e.omroep.nl/metadata/aflevering/%s' % video_id,
	98	video_id,
	99	# We have to remove the javascript callback
609a61e3	100	transform_source=strip_jsonp,
331ae266	101	)
04e0bac2 S	102
04e0bac2 S	103	token = self._get_token(video_id)
331ae266	104
c9ea760e	105	formats = []
1ff30d7b S	106
	107	pubopties = metadata.get('pubopties')
	108	if pubopties:
	109	quality = qualities(['adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std'])
	110	for format_id in pubopties:
	111	format_info = self._download_json(
	112	'http://ida.omroep.nl/odi/?prid=%s&puboptions=%s&adaptive=yes&token=%s'
	113	% (video_id, format_id, token),
	114	video_id, 'Downloading %s JSON' % format_id)
	115	if format_info.get('error_code', 0) or format_info.get('errorcode', 0):
	116	continue
	117	streams = format_info.get('streams')
	118	if streams:
	119	video_info = self._download_json(
	120	streams[0] + '&type=json',
	121	video_id, 'Downloading %s stream JSON' % format_id)
	122	else:
	123	video_info = format_info
	124	video_url = video_info.get('url')
	125	if not video_url:
	126	continue
	127	if format_id == 'adaptive':
	128	formats.extend(self._extract_m3u8_formats(video_url, video_id))
	129	else:
	130	formats.append({
	131	'url': video_url,
	132	'format_id': format_id,
	133	'quality': quality(format_id),
	134	})
	135
	136	streams = metadata.get('streams')
	137	if streams:
	138	for i, stream in enumerate(streams):
	139	stream_url = stream.get('url')
	140	if not stream_url:
	141	continue
5c4a81d9	142	if '.asf' not in stream_url:
a0977064 S	143	formats.append({
	144	'url': stream_url,
	145	'quality': stream.get('kwaliteit'),
	146	})
	147	continue
1ff30d7b S	148	asx = self._download_xml(
	149	stream_url, video_id,
	150	'Downloading stream %d ASX playlist' % i,
	151	transform_source=fix_xml_ampersands)
	152	ref = asx.find('./ENTRY/Ref')
	153	if ref is None:
	154	continue
	155	video_url = ref.get('href')
	156	if not video_url:
	157	continue
c9ea760e	158	formats.append({
944a3de2	159	'url': video_url,
1ff30d7b S	160	'ext': stream.get('formaat', 'asf'),
1ff30d7b S	161	'quality': stream.get('kwaliteit'),
c9ea760e	162	})
1ff30d7b	163
c9ea760e	164	self._sort_formats(formats)
331ae266 JMF	165
	166	return {
	167	'id': video_id,
	168	'title': metadata['titel'],
331ae266	169	'description': metadata['info'],
944a3de2 S	170	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
	171	'upload_date': unified_strdate(metadata.get('gidsdatum')),
	172	'duration': parse_duration(metadata.get('tijdsduur')),
c9ea760e	173	'formats': formats,
331ae266	174	}
d0df9292 JMF	175
d0df9292 JMF	176
04e0bac2 S	177	class NPOLiveIE(NPOBaseIE):
	178	IE_NAME = 'npo.nl:live'
	179	_VALID_URL = r'https?://www\.npo\.nl/live/(?P<id>.+)'
	180
	181	_TEST = {
	182	'url': 'http://www.npo.nl/live/npo-1',
	183	'info_dict': {
	184	'id': 'LI_NEDERLAND1_136692',
	185	'display_id': 'npo-1',
	186	'ext': 'mp4',
	187	'title': 're:^Nederland 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
	188	'description': 'Livestream',
	189	'is_live': True,
	190	},
	191	'params': {
	192	'skip_download': True,
	193	}
	194	}
	195
	196	def _real_extract(self, url):
	197	display_id = self._match_id(url)
	198
	199	webpage = self._download_webpage(url, display_id)
	200
	201	live_id = self._search_regex(
	202	r'data-prid="([^"]+)"', webpage, 'live id')
	203
	204	metadata = self._download_json(
	205	'http://e.omroep.nl/metadata/%s' % live_id,
	206	display_id, transform_source=strip_jsonp)
	207
	208	token = self._get_token(display_id)
	209
	210	formats = []
	211
	212	streams = metadata.get('streams')
	213	if streams:
	214	for stream in streams:
	215	stream_type = stream.get('type').lower()
	216	if stream_type == 'ss':
	217	continue
	218	stream_info = self._download_json(
	219	'http://ida.omroep.nl/aapi/?stream=%s&token=%s&type=jsonp'
	220	% (stream.get('url'), token),
	221	display_id, 'Downloading %s JSON' % stream_type)
	222	if stream_info.get('error_code', 0) or stream_info.get('errorcode', 0):
	223	continue
	224	stream_url = self._download_json(
	225	stream_info['stream'], display_id,
	226	'Downloading %s URL' % stream_type,
	227	transform_source=strip_jsonp)
	228	if stream_type == 'hds':
	229	f4m_formats = self._extract_f4m_formats(stream_url, display_id)
	230	# f4m downloader downloads only piece of live stream
	231	for f4m_format in f4m_formats:
	232	f4m_format['preference'] = -1
	233	formats.extend(f4m_formats)
	234	elif stream_type == 'hls':
	235	formats.extend(self._extract_m3u8_formats(stream_url, display_id, 'mp4'))
	236	else:
	237	formats.append({
	238	'url': stream_url,
	239	})
	240
241	self._sort_formats(formats)
242
243	return {
244	'id': live_id,
245	'display_id': display_id,
246	'title': self._live_title(metadata['titel']),
247	'description': metadata['info'],
248	'thumbnail': metadata.get('images', [{'url': None}])[-1]['url'],
249	'formats': formats,
250	'is_live': True,
251	}
252
253
d0df9292 JMF	254	class TegenlichtVproIE(NPOIE):
	255	IE_NAME = 'tegenlicht.vpro.nl'
	256	_VALID_URL = r'https?://tegenlicht\.vpro\.nl/afleveringen/.*?'
	257
	258	_TESTS = [
	259	{
	260	'url': 'http://tegenlicht.vpro.nl/afleveringen/2012-2013/de-toekomst-komt-uit-afrika.html',
	261	'md5': 'f8065e4e5a7824068ed3c7e783178f2c',
	262	'info_dict': {
	263	'id': 'VPWON_1169289',
	264	'ext': 'm4v',
	265	'title': 'Tegenlicht',
	266	'description': 'md5:d6476bceb17a8c103c76c3b708f05dd1',
	267	'upload_date': '20130225',
	268	},
	269	},
	270	]
	271
	272	def _real_extract(self, url):
	273	name = url_basename(url)
	274	webpage = self._download_webpage(url, name)
	275	urn = self._html_search_meta('mediaurn', webpage)
	276	info_page = self._download_json(
	277	'http://rs.vpro.nl/v2/api/media/%s.json' % urn, name)
	278	return self._get_info(info_page['mid'])