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

# coding: utf-8
from __future__ import unicode_literals

import json
import re
import time

from .common import InfoExtractor
from ..compat import (
    compat_HTTPError,
    compat_str,
    compat_urlparse,
)
from ..utils import (
    determine_ext,
    ExtractorError,
    float_or_none,
    int_or_none,
    remove_end,
    try_get,
    unified_strdate,
    unified_timestamp,
    update_url_query,
    USER_AGENTS,
)


class DPlayIE(InfoExtractor):
    _VALID_URL = r'https?://(?P<domain>www\.(?P<host>dplay\.(?P<country>dk|se|no)))/(?:video(?:er|s)/)?(?P<id>[^/]+/[^/?#]+)'

    _TESTS = [{
        # non geo restricted, via secure api, unsigned download hls URL
        'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/',
        'info_dict': {
            'id': '3172',
            'display_id': 'nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet',
            'ext': 'mp4',
            'title': 'Svensken lär sig njuta av livet',
            'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
            'duration': 2650,
            'timestamp': 1365454320,
            'upload_date': '20130408',
            'creator': 'Kanal 5 (Home)',
            'series': 'Nugammalt - 77 händelser som format Sverige',
            'season_number': 1,
            'episode_number': 1,
            'age_limit': 0,
        },
    }, {
        # geo restricted, via secure api, unsigned download hls URL
        'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/',
        'info_dict': {
            'id': '70816',
            'display_id': 'mig-og-min-mor/season-6-episode-12',
            'ext': 'mp4',
            'title': 'Episode 12',
            'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90',
            'duration': 2563,
            'timestamp': 1429696800,
            'upload_date': '20150422',
            'creator': 'Kanal 4 (Home)',
            'series': 'Mig og min mor',
            'season_number': 6,
            'episode_number': 12,
            'age_limit': 0,
        },
    }, {
        # geo restricted, via direct unsigned hls URL
        'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
        'only_matching': True,
    }, {
        # disco-api
        'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
        'info_dict': {
            'id': '40206',
            'display_id': 'i-kongens-klr/sesong-1-episode-7',
            'ext': 'mp4',
            'title': 'Episode 7',
            'description': 'md5:e3e1411b2b9aebeea36a6ec5d50c60cf',
            'duration': 2611.16,
            'timestamp': 1516726800,
            'upload_date': '20180123',
            'series': 'I kongens klær',
            'season_number': 1,
            'episode_number': 7,
        },
        'params': {
            'format': 'bestvideo',
            'skip_download': True,
        },
    }, {

        'url': 'https://www.dplay.dk/videoer/singleliv/season-5-episode-3',
        'only_matching': True,
    }, {
        'url': 'https://www.dplay.se/videos/sofias-anglar/sofias-anglar-1001',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        display_id = mobj.group('id')
        domain = mobj.group('domain')

        self._initialize_geo_bypass({
            'countries': [mobj.group('country').upper()],
        })

        webpage = self._download_webpage(url, display_id)

        video_id = self._search_regex(
            r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)

        if not video_id:
            host = mobj.group('host')
            disco_base = 'https://disco-api.%s' % host
            self._download_json(
                '%s/token' % disco_base, display_id, 'Downloading token',
                query={
                    'realm': host.replace('.', ''),
                })
            video = self._download_json(
                '%s/content/videos/%s' % (disco_base, display_id), display_id,
                headers={
                    'Referer': url,
                    'x-disco-client': 'WEB:UNKNOWN:dplay-client:0.0.1',
                }, query={
                    'include': 'show'
                })
            video_id = video['data']['id']
            info = video['data']['attributes']
            title = info['name']
            formats = []
            for format_id, format_dict in self._download_json(
                    '%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id),
                    display_id)['data']['attributes']['streaming'].items():
                if not isinstance(format_dict, dict):
                    continue
                format_url = format_dict.get('url')
                if not format_url:
                    continue
                ext = determine_ext(format_url)
                if format_id == 'dash' or ext == 'mpd':
                    formats.extend(self._extract_mpd_formats(
                        format_url, display_id, mpd_id='dash', fatal=False))
                elif format_id == 'hls' or ext == 'm3u8':
                    formats.extend(self._extract_m3u8_formats(
                        format_url, display_id, 'mp4',
                        entry_protocol='m3u8_native', m3u8_id='hls',
                        fatal=False))
                else:
                    formats.append({
                        'url': format_url,
                        'format_id': format_id,
                    })
            self._sort_formats(formats)

            series = None
            try:
                included = video.get('included')
                if isinstance(included, list):
                    show = next(e for e in included if e.get('type') == 'show')
                    series = try_get(
                        show, lambda x: x['attributes']['name'], compat_str)
            except StopIteration:
                pass

            return {
                'id': video_id,
                'display_id': display_id,
                'title': title,
                'description': info.get('description'),
                'duration': float_or_none(
                    info.get('videoDuration'), scale=1000),
                'timestamp': unified_timestamp(info.get('publishStart')),
                'series': series,
                'season_number': int_or_none(info.get('seasonNumber')),
                'episode_number': int_or_none(info.get('episodeNumber')),
                'age_limit': int_or_none(info.get('minimum_age')),
                'formats': formats,
            }

        info = self._download_json(
            'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
            video_id)['data'][0]

        title = info['title']

        PROTOCOLS = ('hls', 'hds')
        formats = []

        def extract_formats(protocol, manifest_url):
            if protocol == 'hls':
                m3u8_formats = self._extract_m3u8_formats(
                    manifest_url, video_id, ext='mp4',
                    entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
                # Sometimes final URLs inside m3u8 are unsigned, let's fix this
                # ourselves. Also fragments' URLs are only served signed for
                # Safari user agent.
                query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
                for m3u8_format in m3u8_formats:
                    m3u8_format.update({
                        'url': update_url_query(m3u8_format['url'], query),
                        'http_headers': {
                            'User-Agent': USER_AGENTS['Safari'],
                        },
                    })
                formats.extend(m3u8_formats)
            elif protocol == 'hds':
                formats.extend(self._extract_f4m_formats(
                    manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
                    video_id, f4m_id=protocol, fatal=False))

        domain_tld = domain.split('.')[-1]
        if domain_tld in ('se', 'dk', 'no'):
            for protocol in PROTOCOLS:
                # Providing dsc-geo allows to bypass geo restriction in some cases
                self._set_cookie(
                    'secure.dplay.%s' % domain_tld, 'dsc-geo',
                    json.dumps({
                        'countryCode': domain_tld.upper(),
                        'expiry': (time.time() + 20 * 60) * 1000,
                    }))
                stream = self._download_json(
                    'https://secure.dplay.%s/secure/api/v2/user/authorization/stream/%s?stream_type=%s'
                    % (domain_tld, video_id, protocol), video_id,
                    'Downloading %s stream JSON' % protocol, fatal=False)
                if stream and stream.get(protocol):
                    extract_formats(protocol, stream[protocol])

        # The last resort is to try direct unsigned hls/hds URLs from info dictionary.
        # Sometimes this does work even when secure API with dsc-geo has failed (e.g.
        # http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
        if not formats:
            for protocol in PROTOCOLS:
                if info.get(protocol):
                    extract_formats(protocol, info[protocol])

        self._sort_formats(formats)

        subtitles = {}
        for lang in ('se', 'sv', 'da', 'nl', 'no'):
            for format_id in ('web_vtt', 'vtt', 'srt'):
                subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
                if subtitle_url:
                    subtitles.setdefault(lang, []).append({'url': subtitle_url})

        return {
            'id': video_id,
            'display_id': display_id,
            'title': title,
            'description': info.get('video_metadata_longDescription'),
            'duration': int_or_none(info.get('video_metadata_length'), scale=1000),
            'timestamp': int_or_none(info.get('video_publish_date')),
            'creator': info.get('video_metadata_homeChannel'),
            'series': info.get('video_metadata_show'),
            'season_number': int_or_none(info.get('season')),
            'episode_number': int_or_none(info.get('episode')),
            'age_limit': int_or_none(info.get('minimum_age')),
            'formats': formats,
            'subtitles': subtitles,
        }


class DPlayItIE(InfoExtractor):
    _VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
    _GEO_COUNTRIES = ['IT']
    _TEST = {
        'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
        'md5': '2b808ffb00fc47b884a172ca5d13053c',
        'info_dict': {
            'id': '6918',
            'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
            'ext': 'mp4',
            'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
            'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
            'thumbnail': r're:^https?://.*\.jpe?g',
            'upload_date': '20160524',
            'series': 'Biografie imbarazzanti',
            'season_number': 1,
            'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
            'episode_number': 1,
        },
    }

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

        webpage = self._download_webpage(url, display_id)

        title = remove_end(self._og_search_title(webpage), ' | Dplay')

        video_id = None

        info = self._search_regex(
            r'playback_json\s*:\s*JSON\.parse\s*\(\s*("(?:\\.|[^"\\])+?")',
            webpage, 'playback JSON', default=None)
        if info:
            for _ in range(2):
                info = self._parse_json(info, display_id, fatal=False)
                if not info:
                    break
            else:
                video_id = try_get(info, lambda x: x['data']['id'])

        if not info:
            info_url = self._search_regex(
                r'url\s*[:=]\s*["\']((?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)',
                webpage, 'info url')

            video_id = info_url.rpartition('/')[-1]

            try:
                info = self._download_json(
                    info_url, display_id, headers={
                        'Authorization': 'Bearer %s' % self._get_cookies(url).get(
                            'dplayit_token').value,
                        'Referer': url,
                    })
            except ExtractorError as e:
                if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
                    info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
                    error = info['errors'][0]
                    if error.get('code') == 'access.denied.geoblocked':
                        self.raise_geo_restricted(
                            msg=error.get('detail'), countries=self._GEO_COUNTRIES)
                    raise ExtractorError(info['errors'][0]['detail'], expected=True)
                raise

        hls_url = info['data']['attributes']['streaming']['hls']['url']

        formats = self._extract_m3u8_formats(
            hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
            m3u8_id='hls')

        series = self._html_search_regex(
            r'(?s)<h1[^>]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)</h1>',
            webpage, 'series', fatal=False)
        episode = self._search_regex(
            r'<p[^>]+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*<br/>\s*<b>([^<]+)',
            webpage, 'episode', fatal=False)

        mobj = re.search(
            r'(?s)<span[^>]+class=["\']dates["\'][^>]*>.+?\bS\.(?P<season_number>\d+)\s+E\.(?P<episode_number>\d+)\s*-\s*(?P<upload_date>\d{2}/\d{2}/\d{4})',
            webpage)
        if mobj:
            season_number = int(mobj.group('season_number'))
            episode_number = int(mobj.group('episode_number'))
            upload_date = unified_strdate(mobj.group('upload_date'))
        else:
            season_number = episode_number = upload_date = None

        return {
            'id': compat_str(video_id or display_id),
            'display_id': display_id,
            'title': title,
            'description': self._og_search_description(webpage),
            'thumbnail': self._og_search_thumbnail(webpage),
            'series': series,
            'season_number': season_number,
            'episode': episode,
            'episode_number': episode_number,
            'upload_date': upload_date,
            'formats': formats,
        }
Commit	Line	Data
940b606a	1	# coding: utf-8
4cd759f7 JMF	2	from __future__ import unicode_literals
4cd759f7 JMF	3
940b606a S	4	import json
940b606a S	5	import re
4cd759f7 JMF	6	import time
	7
	8	from .common import InfoExtractor
0cf2352e	9	from ..compat import (
0cf2352e	10	compat_HTTPError,
15d1e8a2 S	11	compat_str,
15d1e8a2 S	12	compat_urlparse,
0cf2352e	13	)
5448b781	14	from ..utils import (
864a4576	15	determine_ext,
0cf2352e	16	ExtractorError,
864a4576	17	float_or_none,
5448b781	18	int_or_none,
0cf2352e	19	remove_end,
15d1e8a2 S	20	try_get,
15d1e8a2 S	21	unified_strdate,
864a4576	22	unified_timestamp,
5448b781	23	update_url_query,
15d1e8a2	24	USER_AGENTS,
5448b781	25	)
940b606a	26
4cd759f7 JMF	27
4cd759f7 JMF	28	class DPlayIE(InfoExtractor):
d6b15291	29	_VALID_URL = r'https?://(?P<domain>www\.(?P<host>dplay\.(?P<country>dk\|se\|no)))/(?:video(?:er\|s)/)?(?P<id>[^/]+/[^/?#]+)'
95050537	30
940b606a	31	_TESTS = [{
5448b781	32	# non geo restricted, via secure api, unsigned download hls URL
940b606a S	33	'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/',
	34	'info_dict': {
	35	'id': '3172',
864a4576	36	'display_id': 'nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet',
5448b781	37	'ext': 'mp4',
940b606a S	38	'title': 'Svensken lär sig njuta av livet',
	39	'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
	40	'duration': 2650,
	41	'timestamp': 1365454320,
	42	'upload_date': '20130408',
	43	'creator': 'Kanal 5 (Home)',
	44	'series': 'Nugammalt - 77 händelser som format Sverige',
	45	'season_number': 1,
	46	'episode_number': 1,
	47	'age_limit': 0,
95050537	48	},
940b606a	49	}, {
5448b781	50	# geo restricted, via secure api, unsigned download hls URL
940b606a S	51	'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/',
	52	'info_dict': {
	53	'id': '70816',
864a4576	54	'display_id': 'mig-og-min-mor/season-6-episode-12',
5448b781	55	'ext': 'mp4',
940b606a S	56	'title': 'Episode 12',
	57	'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90',
	58	'duration': 2563,
	59	'timestamp': 1429696800,
	60	'upload_date': '20150422',
5448b781	61	'creator': 'Kanal 4 (Home)',
940b606a S	62	'series': 'Mig og min mor',
	63	'season_number': 6,
	64	'episode_number': 12,
	65	'age_limit': 0,
	66	},
5add979d	67	}, {
fd0ff8ba	68	# geo restricted, via direct unsigned hls URL
5add979d S	69	'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
5add979d S	70	'only_matching': True,
864a4576 S	71	}, {
	72	# disco-api
	73	'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
	74	'info_dict': {
	75	'id': '40206',
	76	'display_id': 'i-kongens-klr/sesong-1-episode-7',
	77	'ext': 'mp4',
	78	'title': 'Episode 7',
	79	'description': 'md5:e3e1411b2b9aebeea36a6ec5d50c60cf',
	80	'duration': 2611.16,
	81	'timestamp': 1516726800,
	82	'upload_date': '20180123',
	83	'series': 'I kongens klær',
	84	'season_number': 1,
	85	'episode_number': 7,
	86	},
	87	'params': {
	88	'format': 'bestvideo',
	89	'skip_download': True,
	90	},
a0ee342b	91	}, {
d6b15291	92
a0ee342b S	93	'url': 'https://www.dplay.dk/videoer/singleliv/season-5-episode-3',
a0ee342b S	94	'only_matching': True,
d6b15291 S	95	}, {
	96	'url': 'https://www.dplay.se/videos/sofias-anglar/sofias-anglar-1001',
	97	'only_matching': True,
940b606a	98	}]
4cd759f7 JMF	99
4cd759f7 JMF	100	def _real_extract(self, url):
940b606a S	101	mobj = re.match(self._VALID_URL, url)
	102	display_id = mobj.group('id')
	103	domain = mobj.group('domain')
	104
5f95927a S	105	self._initialize_geo_bypass({
	106	'countries': [mobj.group('country').upper()],
	107	})
a0ee342b	108
4cd759f7	109	webpage = self._download_webpage(url, display_id)
4cd759f7	110
940b606a	111	video_id = self._search_regex(
864a4576 S	112	r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
	113
	114	if not video_id:
	115	host = mobj.group('host')
	116	disco_base = 'https://disco-api.%s' % host
	117	self._download_json(
	118	'%s/token' % disco_base, display_id, 'Downloading token',
	119	query={
	120	'realm': host.replace('.', ''),
	121	})
	122	video = self._download_json(
	123	'%s/content/videos/%s' % (disco_base, display_id), display_id,
	124	headers={
	125	'Referer': url,
	126	'x-disco-client': 'WEB:UNKNOWN:dplay-client:0.0.1',
	127	}, query={
	128	'include': 'show'
	129	})
	130	video_id = video['data']['id']
	131	info = video['data']['attributes']
	132	title = info['name']
	133	formats = []
	134	for format_id, format_dict in self._download_json(
	135	'%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id),
	136	display_id)['data']['attributes']['streaming'].items():
	137	if not isinstance(format_dict, dict):
	138	continue
	139	format_url = format_dict.get('url')
	140	if not format_url:
	141	continue
	142	ext = determine_ext(format_url)
	143	if format_id == 'dash' or ext == 'mpd':
	144	formats.extend(self._extract_mpd_formats(
	145	format_url, display_id, mpd_id='dash', fatal=False))
	146	elif format_id == 'hls' or ext == 'm3u8':
	147	formats.extend(self._extract_m3u8_formats(
	148	format_url, display_id, 'mp4',
	149	entry_protocol='m3u8_native', m3u8_id='hls',
	150	fatal=False))
	151	else:
	152	formats.append({
	153	'url': format_url,
	154	'format_id': format_id,
	155	})
	156	self._sort_formats(formats)
	157
	158	series = None
	159	try:
	160	included = video.get('included')
	161	if isinstance(included, list):
	162	show = next(e for e in included if e.get('type') == 'show')
	163	series = try_get(
	164	show, lambda x: x['attributes']['name'], compat_str)
	165	except StopIteration:
	166	pass
	167
	168	return {
	169	'id': video_id,
	170	'display_id': display_id,
	171	'title': title,
	172	'description': info.get('description'),
	173	'duration': float_or_none(
	174	info.get('videoDuration'), scale=1000),
	175	'timestamp': unified_timestamp(info.get('publishStart')),
176	'series': series,
177	'season_number': int_or_none(info.get('seasonNumber')),
178	'episode_number': int_or_none(info.get('episodeNumber')),
179	'age_limit': int_or_none(info.get('minimum_age')),
180	'formats': formats,
181	}
95050537	182
940b606a S	183	info = self._download_json(
940b606a S	184	'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
4cd759f7 JMF	185	video_id)['data'][0]
4cd759f7 JMF	186
940b606a	187	title = info['title']
95050537	188
940b606a S	189	PROTOCOLS = ('hls', 'hds')
940b606a S	190	formats = []
95050537	191
940b606a S	192	def extract_formats(protocol, manifest_url):
940b606a S	193	if protocol == 'hls':
5448b781	194	m3u8_formats = self._extract_m3u8_formats(
940b606a	195	manifest_url, video_id, ext='mp4',
5448b781 S	196	entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
5448b781 S	197	# Sometimes final URLs inside m3u8 are unsigned, let's fix this
3530e0d3 S	198	# ourselves. Also fragments' URLs are only served signed for
3530e0d3 S	199	# Safari user agent.
5448b781 S	200	query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
5448b781 S	201	for m3u8_format in m3u8_formats:
3530e0d3 S	202	m3u8_format.update({
	203	'url': update_url_query(m3u8_format['url'], query),
	204	'http_headers': {
	205	'User-Agent': USER_AGENTS['Safari'],
	206	},
	207	})
5448b781	208	formats.extend(m3u8_formats)
940b606a S	209	elif protocol == 'hds':
	210	formats.extend(self._extract_f4m_formats(
	211	manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
	212	video_id, f4m_id=protocol, fatal=False))
	213
	214	domain_tld = domain.split('.')[-1]
397ec446	215	if domain_tld in ('se', 'dk', 'no'):
940b606a	216	for protocol in PROTOCOLS:
fd0ff8ba	217	# Providing dsc-geo allows to bypass geo restriction in some cases
940b606a S	218	self._set_cookie(
	219	'secure.dplay.%s' % domain_tld, 'dsc-geo',
	220	json.dumps({
	221	'countryCode': domain_tld.upper(),
	222	'expiry': (time.time() + 20 * 60) * 1000,
	223	}))
	224	stream = self._download_json(
	225	'https://secure.dplay.%s/secure/api/v2/user/authorization/stream/%s?stream_type=%s'
	226	% (domain_tld, video_id, protocol), video_id,
	227	'Downloading %s stream JSON' % protocol, fatal=False)
	228	if stream and stream.get(protocol):
	229	extract_formats(protocol, stream[protocol])
fd0ff8ba S	230
	231	# The last resort is to try direct unsigned hls/hds URLs from info dictionary.
	232	# Sometimes this does work even when secure API with dsc-geo has failed (e.g.
	233	# http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
	234	if not formats:
940b606a S	235	for protocol in PROTOCOLS:
	236	if info.get(protocol):
	237	extract_formats(protocol, info[protocol])
4cd759f7	238
19dbaeec S	239	self._sort_formats(formats)
19dbaeec S	240
e239413f S	241	subtitles = {}
	242	for lang in ('se', 'sv', 'da', 'nl', 'no'):
	243	for format_id in ('web_vtt', 'vtt', 'srt'):
	244	subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
	245	if subtitle_url:
	246	subtitles.setdefault(lang, []).append({'url': subtitle_url})
	247
4cd759f7 JMF	248	return {
	249	'id': video_id,
	250	'display_id': display_id,
940b606a	251	'title': title,
95050537	252	'description': info.get('video_metadata_longDescription'),
940b606a S	253	'duration': int_or_none(info.get('video_metadata_length'), scale=1000),
	254	'timestamp': int_or_none(info.get('video_publish_date')),
	255	'creator': info.get('video_metadata_homeChannel'),
	256	'series': info.get('video_metadata_show'),
95050537 AR	257	'season_number': int_or_none(info.get('season')),
95050537 AR	258	'episode_number': int_or_none(info.get('episode')),
940b606a S	259	'age_limit': int_or_none(info.get('minimum_age')),
940b606a S	260	'formats': formats,
e239413f	261	'subtitles': subtitles,
4cd759f7	262	}
0cf2352e S	263
	264
	265	class DPlayItIE(InfoExtractor):
	266	_VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
	267	_GEO_COUNTRIES = ['IT']
	268	_TEST = {
	269	'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
	270	'md5': '2b808ffb00fc47b884a172ca5d13053c',
	271	'info_dict': {
	272	'id': '6918',
	273	'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
	274	'ext': 'mp4',
	275	'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
	276	'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
	277	'thumbnail': r're:^https?://.*\.jpe?g',
	278	'upload_date': '20160524',
	279	'series': 'Biografie imbarazzanti',
	280	'season_number': 1,
	281	'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
	282	'episode_number': 1,
	283	},
	284	}
	285
	286	def _real_extract(self, url):
	287	display_id = self._match_id(url)
	288
	289	webpage = self._download_webpage(url, display_id)
	290
0cf2352e S	291	title = remove_end(self._og_search_title(webpage), ' \| Dplay')
0cf2352e S	292
15d1e8a2 S	293	video_id = None
	294
	295	info = self._search_regex(
	296	r'playback_json\s:\sJSON\.parse\s\(\s("(?:\\.\|[^"\\])+?")',
	297	webpage, 'playback JSON', default=None)
	298	if info:
	299	for _ in range(2):
	300	info = self._parse_json(info, display_id, fatal=False)
	301	if not info:
	302	break
	303	else:
	304	video_id = try_get(info, lambda x: x['data']['id'])
	305
	306	if not info:
	307	info_url = self._search_regex(
	308	r'url\s[:=]\s["\']((?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)',
	309	webpage, 'info url')
	310
	311	video_id = info_url.rpartition('/')[-1]
	312
	313	try:
	314	info = self._download_json(
	315	info_url, display_id, headers={
	316	'Authorization': 'Bearer %s' % self._get_cookies(url).get(
	317	'dplayit_token').value,
	318	'Referer': url,
	319	})
	320	except ExtractorError as e:
	321	if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
	322	info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
	323	error = info['errors'][0]
	324	if error.get('code') == 'access.denied.geoblocked':
	325	self.raise_geo_restricted(
	326	msg=error.get('detail'), countries=self._GEO_COUNTRIES)
	327	raise ExtractorError(info['errors'][0]['detail'], expected=True)
	328	raise
0cf2352e S	329
	330	hls_url = info['data']['attributes']['streaming']['hls']['url']
	331
	332	formats = self._extract_m3u8_formats(
	333	hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
	334	m3u8_id='hls')
	335
	336	series = self._html_search_regex(
	337	r'(?s)<h1[^>]+class=["\'].?\bshow_title\b.?["\'][^>]*>(.+?)</h1>',
	338	webpage, 'series', fatal=False)
	339	episode = self._search_regex(
	340	r'<p[^>]+class=["\'].?\bdesc_ep\b.?["\'][^>]>\s<br/>\s*<b>([^<]+)',
	341	webpage, 'episode', fatal=False)
	342
	343	mobj = re.search(
	344	r'(?s)<span[^>]+class=["\']dates["\'][^>]>.+?\bS\.(?P<season_number>\d+)\s+E\.(?P<episode_number>\d+)\s-\s*(?P<upload_date>\d{2}/\d{2}/\d{4})',
	345	webpage)
	346	if mobj:
	347	season_number = int(mobj.group('season_number'))
	348	episode_number = int(mobj.group('episode_number'))
	349	upload_date = unified_strdate(mobj.group('upload_date'))
	350	else:
	351	season_number = episode_number = upload_date = None
	352
	353	return {
15d1e8a2	354	'id': compat_str(video_id or display_id),
0cf2352e S	355	'display_id': display_id,
	356	'title': title,
	357	'description': self._og_search_description(webpage),
	358	'thumbnail': self._og_search_thumbnail(webpage),
	359	'series': series,
	360	'season_number': season_number,
	361	'episode': episode,
	362	'episode_number': episode_number,
	363	'upload_date': upload_date,
	364	'formats': formats,
	365	}