[yt-dlp.git] / yt_dlp / extractor / prosiebensat1.py

import hashlib
import re

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    determine_ext,
    float_or_none,
    int_or_none,
    merge_dicts,
    unified_strdate,
)


class ProSiebenSat1BaseIE(InfoExtractor):
    _GEO_BYPASS = False
    _ACCESS_ID = None
    _SUPPORTED_PROTOCOLS = 'dash:clear,hls:clear,progressive:clear'
    _V4_BASE_URL = 'https://vas-v4.p7s1video.net/4.0/get'

    def _extract_video_info(self, url, clip_id):
        client_location = url

        video = self._download_json(
            'http://vas.sim-technik.de/vas/live/v2/videos',
            clip_id, 'Downloading videos JSON', query={
                'access_token': self._TOKEN,
                'client_location': client_location,
                'client_name': self._CLIENT_NAME,
                'ids': clip_id,
            })[0]

        if not self.get_param('allow_unplayable_formats') and video.get('is_protected') is True:
            self.report_drm(clip_id)

        formats = []
        if self._ACCESS_ID:
            raw_ct = self._ENCRYPTION_KEY + clip_id + self._IV + self._ACCESS_ID
            protocols = self._download_json(
                self._V4_BASE_URL + 'protocols', clip_id,
                'Downloading protocols JSON',
                headers=self.geo_verification_headers(), query={
                    'access_id': self._ACCESS_ID,
                    'client_token': hashlib.sha1((raw_ct).encode()).hexdigest(),
                    'video_id': clip_id,
                }, fatal=False, expected_status=(403,)) or {}
            error = protocols.get('error') or {}
            if error.get('title') == 'Geo check failed':
                self.raise_geo_restricted(countries=['AT', 'CH', 'DE'])
            server_token = protocols.get('server_token')
            if server_token:
                urls = (self._download_json(
                    self._V4_BASE_URL + 'urls', clip_id, 'Downloading urls JSON', query={
                        'access_id': self._ACCESS_ID,
                        'client_token': hashlib.sha1((raw_ct + server_token + self._SUPPORTED_PROTOCOLS).encode()).hexdigest(),
                        'protocols': self._SUPPORTED_PROTOCOLS,
                        'server_token': server_token,
                        'video_id': clip_id,
                    }, fatal=False) or {}).get('urls') or {}
                for protocol, variant in urls.items():
                    source_url = variant.get('clear', {}).get('url')
                    if not source_url:
                        continue
                    if protocol == 'dash':
                        formats.extend(self._extract_mpd_formats(
                            source_url, clip_id, mpd_id=protocol, fatal=False))
                    elif protocol == 'hls':
                        formats.extend(self._extract_m3u8_formats(
                            source_url, clip_id, 'mp4', 'm3u8_native',
                            m3u8_id=protocol, fatal=False))
                    else:
                        formats.append({
                            'url': source_url,
                            'format_id': protocol,
                        })
        if not formats:
            source_ids = [str(source['id']) for source in video['sources']]

            client_id = self._SALT[:2] + hashlib.sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode()).hexdigest()

            sources = self._download_json(
                f'http://vas.sim-technik.de/vas/live/v2/videos/{clip_id}/sources',
                clip_id, 'Downloading sources JSON', query={
                    'access_token': self._TOKEN,
                    'client_id': client_id,
                    'client_location': client_location,
                    'client_name': self._CLIENT_NAME,
                })
            server_id = sources['server_id']

            def fix_bitrate(bitrate):
                bitrate = int_or_none(bitrate)
                if not bitrate:
                    return None
                return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate

            for source_id in source_ids:
                client_id = self._SALT[:2] + hashlib.sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._CLIENT_NAME]).encode()).hexdigest()
                urls = self._download_json(
                    f'http://vas.sim-technik.de/vas/live/v2/videos/{clip_id}/sources/url',
                    clip_id, 'Downloading urls JSON', fatal=False, query={
                        'access_token': self._TOKEN,
                        'client_id': client_id,
                        'client_location': client_location,
                        'client_name': self._CLIENT_NAME,
                        'server_id': server_id,
                        'source_ids': source_id,
                    })
                if not urls:
                    continue
                if urls.get('status_code') != 0:
                    raise ExtractorError('This video is unavailable', expected=True)
                urls_sources = urls['sources']
                if isinstance(urls_sources, dict):
                    urls_sources = urls_sources.values()
                for source in urls_sources:
                    source_url = source.get('url')
                    if not source_url:
                        continue
                    protocol = source.get('protocol')
                    mimetype = source.get('mimetype')
                    if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
                        formats.extend(self._extract_f4m_formats(
                            source_url, clip_id, f4m_id='hds', fatal=False))
                    elif mimetype == 'application/x-mpegURL':
                        formats.extend(self._extract_m3u8_formats(
                            source_url, clip_id, 'mp4', 'm3u8_native',
                            m3u8_id='hls', fatal=False))
                    elif mimetype == 'application/dash+xml':
                        formats.extend(self._extract_mpd_formats(
                            source_url, clip_id, mpd_id='dash', fatal=False))
                    else:
                        tbr = fix_bitrate(source['bitrate'])
                        if protocol in ('rtmp', 'rtmpe'):
                            mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
                            if not mobj:
                                continue
                            path = mobj.group('path')
                            mp4colon_index = path.rfind('mp4:')
                            app = path[:mp4colon_index]
                            play_path = path[mp4colon_index:]
                            formats.append({
                                'url': '{}/{}'.format(mobj.group('url'), app),
                                'app': app,
                                'play_path': play_path,
                                'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
                                'page_url': 'http://www.prosieben.de',
                                'tbr': tbr,
                                'ext': 'flv',
                                'format_id': 'rtmp{}'.format(f'-{tbr}' if tbr else ''),
                            })
                        else:
                            formats.append({
                                'url': source_url,
                                'tbr': tbr,
                                'format_id': 'http{}'.format(f'-{tbr}' if tbr else ''),
                            })

        return {
            'duration': float_or_none(video.get('duration')),
            'formats': formats,
        }


class ProSiebenSat1IE(ProSiebenSat1BaseIE):
    IE_NAME = 'prosiebensat1'
    IE_DESC = 'ProSiebenSat.1 Digital'
    _VALID_URL = r'''(?x)
                    https?://
                        (?:www\.)?
                        (?:
                            (?:beta\.)?
                            (?:
                                prosieben(?:maxx)?|sixx|sat1(?:gold)?|kabeleins(?:doku)?|the-voice-of-germany|advopedia
                            )\.(?:de|at|ch)|
                            ran\.de|fem\.com|advopedia\.de|galileo\.tv/video
                        )
                        /(?P<id>.+)
                    '''

    _TESTS = [
        {
            # Tests changes introduced in https://github.com/ytdl-org/youtube-dl/pull/6242
            # in response to fixing https://github.com/ytdl-org/youtube-dl/issues/6215:
            # - malformed f4m manifest support
            # - proper handling of URLs starting with `https?://` in 2.0 manifests
            # - recursive child f4m manifests extraction
            'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
            'info_dict': {
                'id': '2104602',
                'ext': 'mp4',
                'title': 'CIRCUS HALLIGALLI - Episode 18 - Staffel 2',
                'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
                'upload_date': '20131231',
                'duration': 5845.04,
                'series': 'CIRCUS HALLIGALLI',
                'season_number': 2,
                'episode': 'Episode 18 - Staffel 2',
                'episode_number': 18,
            },
        },
        {
            'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
            'info_dict': {
                'id': '2570327',
                'ext': 'mp4',
                'title': 'Lady-Umstyling für Audrina',
                'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
                'upload_date': '20131014',
                'duration': 606.76,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'Seems to be broken',
        },
        {
            'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
            'info_dict': {
                'id': '2429369',
                'ext': 'mp4',
                'title': 'Countdown für die Autowerkstatt',
                'description': 'md5:809fc051a457b5d8666013bc40698817',
                'upload_date': '20140223',
                'duration': 2595.04,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'This video is unavailable',
        },
        {
            'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
            'info_dict': {
                'id': '2904997',
                'ext': 'mp4',
                'title': 'Sexy laufen in Ugg Boots',
                'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
                'upload_date': '20140122',
                'duration': 245.32,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'This video is unavailable',
        },
        {
            'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
            'info_dict': {
                'id': '2906572',
                'ext': 'mp4',
                'title': 'Im Interview: Kai Wiesinger',
                'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
                'upload_date': '20140203',
                'duration': 522.56,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'This video is unavailable',
        },
        {
            'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
            'info_dict': {
                'id': '2992323',
                'ext': 'mp4',
                'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
                'description': 'md5:2669cde3febe9bce13904f701e774eb6',
                'upload_date': '20141014',
                'duration': 2410.44,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'This video is unavailable',
        },
        {
            'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
            'info_dict': {
                'id': '3004256',
                'ext': 'mp4',
                'title': 'Schalke: Tönnies möchte Raul zurück',
                'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
                'upload_date': '20140226',
                'duration': 228.96,
            },
            'params': {
                # rtmp download
                'skip_download': True,
            },
            'skip': 'This video is unavailable',
        },
        {
            'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
            'info_dict': {
                'id': '2572814',
                'ext': 'mp4',
                'title': 'The Voice of Germany - Andreas Kümmert: Rocket Man',
                'description': 'md5:6ddb02b0781c6adf778afea606652e38',
                'timestamp': 1382041620,
                'upload_date': '20131017',
                'duration': 469.88,
            },
            'params': {
                'skip_download': True,
            },
        },
        {
            'url': 'http://www.fem.com/videos/beauty-lifestyle/kurztrips-zum-valentinstag',
            'info_dict': {
                'id': '2156342',
                'ext': 'mp4',
                'title': 'Kurztrips zum Valentinstag',
                'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
                'duration': 307.24,
            },
            'params': {
                'skip_download': True,
            },
        },
        {
            'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
            'info_dict': {
                'id': '439664',
                'title': 'Episode 8 - Ganze Folge - Playlist',
                'description': 'md5:63b8963e71f481782aeea877658dec84',
            },
            'playlist_count': 2,
            'skip': 'This video is unavailable',
        },
        {
            # title in <h2 class="subtitle">
            'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
            'info_dict': {
                'id': '4895826',
                'ext': 'mp4',
                'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
                'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
                'upload_date': '20170302',
            },
            'params': {
                'skip_download': True,
            },
            'skip': 'geo restricted to Germany',
        },
        {
            # geo restricted to Germany
            'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
            'only_matching': True,
        },
        {
            # geo restricted to Germany
            'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
            'only_matching': True,
        },
        {
            # geo restricted to Germany
            'url': 'https://www.galileo.tv/video/diese-emojis-werden-oft-missverstanden',
            'only_matching': True,
        },
        {
            'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
            'only_matching': True,
        },
        {
            'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
            'only_matching': True,
        },
    ]

    _TOKEN = 'prosieben'
    _SALT = '01!8d8F_)r9]4s[qeuXfP%'
    _CLIENT_NAME = 'kolibri-2.0.19-splec4'

    _ACCESS_ID = 'x_prosiebenmaxx-de'
    _ENCRYPTION_KEY = 'Eeyeey9oquahthainoofashoyoikosag'
    _IV = 'Aeluchoc6aevechuipiexeeboowedaok'

    _CLIPID_REGEXES = [
        r'"clip_id"\s*:\s+"(\d+)"',
        r'clipid: "(\d+)"',
        r'clip[iI]d=(\d+)',
        r'clip[iI][dD]\s*=\s*["\'](\d+)',
        r"'itemImageUrl'\s*:\s*'/dynamic/thumbnails/full/\d+/(\d+)",
        r'proMamsId&quot;\s*:\s*&quot;(\d+)',
        r'proMamsId"\s*:\s*"(\d+)',
    ]
    _TITLE_REGEXES = [
        r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
        r'<header class="clearfix">\s*<h3>(.+?)</h3>',
        r'<!-- start video -->\s*<h1>(.+?)</h1>',
        r'<h1 class="att-name">\s*(.+?)</h1>',
        r'<header class="module_header">\s*<h2>([^<]+)</h2>\s*</header>',
        r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
        r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
        r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
    ]
    _DESCRIPTION_REGEXES = [
        r'<p itemprop="description">\s*(.+?)</p>',
        r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
        r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
        r'<p class="att-description">\s*(.+?)\s*</p>',
        r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
        r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
    ]
    _UPLOAD_DATE_REGEXES = [
        r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
        r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
        r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
        r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
    ]
    _PAGE_TYPE_REGEXES = [
        r'<meta name="page_type" content="([^"]+)">',
        r"'itemType'\s*:\s*'([^']*)'",
    ]
    _PLAYLIST_ID_REGEXES = [
        r'content[iI]d=(\d+)',
        r"'itemId'\s*:\s*'([^']*)'",
    ]
    _PLAYLIST_CLIP_REGEXES = [
        r'(?s)data-qvt=.+?<a href="([^"]+)"',
    ]

    def _extract_clip(self, url, webpage):
        clip_id = self._html_search_regex(
            self._CLIPID_REGEXES, webpage, 'clip id')
        title = self._html_search_regex(
            self._TITLE_REGEXES, webpage, 'title',
            default=None) or self._og_search_title(webpage)
        info = self._extract_video_info(url, clip_id)
        description = self._html_search_regex(
            self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
        if description is None:
            description = self._og_search_description(webpage)
        thumbnail = self._og_search_thumbnail(webpage)
        upload_date = unified_strdate(
            self._html_search_meta('og:published_time', webpage,
                                   'upload date', default=None)
            or self._html_search_regex(self._UPLOAD_DATE_REGEXES,
                                       webpage, 'upload date', default=None))

        json_ld = self._search_json_ld(webpage, clip_id, default={})

        return merge_dicts(info, {
            'id': clip_id,
            'title': title,
            'description': description,
            'thumbnail': thumbnail,
            'upload_date': upload_date,
        }, json_ld)

    def _extract_playlist(self, url, webpage):
        playlist_id = self._html_search_regex(
            self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
        playlist = self._parse_json(
            self._search_regex(
                r'var\s+contentResources\s*=\s*(\[.+?\]);\s*</script',
                webpage, 'playlist'),
            playlist_id)
        entries = []
        for item in playlist:
            clip_id = item.get('id') or item.get('upc')
            if not clip_id:
                continue
            info = self._extract_video_info(url, clip_id)
            info.update({
                'id': clip_id,
                'title': item.get('title') or item.get('teaser', {}).get('headline'),
                'description': item.get('teaser', {}).get('description'),
                'thumbnail': item.get('poster'),
                'duration': float_or_none(item.get('duration')),
                'series': item.get('tvShowTitle'),
                'uploader': item.get('broadcastPublisher'),
            })
            entries.append(info)
        return self.playlist_result(entries, playlist_id)

    def _real_extract(self, url):
        video_id = self._match_id(url)
        webpage = self._download_webpage(url, video_id)
        page_type = self._search_regex(
            self._PAGE_TYPE_REGEXES, webpage,
            'page type', default='clip').lower()
        if page_type == 'clip':
            return self._extract_clip(url, webpage)
        elif page_type == 'playlist':
            return self._extract_playlist(url, webpage)
        else:
            raise ExtractorError(
                f'Unsupported page type {page_type}', expected=True)
Commit	Line	Data
e3a3ed8a	1	import hashlib
0c7214c4 S	2	import re
0c7214c4 S	3
0c7214c4	4	from .common import InfoExtractor
1cc79574	5	from ..utils import (
61be92e2	6	ExtractorError,
f01f7311	7	determine_ext,
993df6bc	8	float_or_none,
01534bf5	9	int_or_none,
38db9a40	10	merge_dicts,
2af0f87c	11	unified_strdate,
0c7214c4 S	12	)
	13
	14
9d54b02b	15	class ProSiebenSat1BaseIE(InfoExtractor):
0c15a56f	16	_GEO_BYPASS = False
c5eb75b3 RA	17	_ACCESS_ID = None
	18	_SUPPORTED_PROTOCOLS = 'dash:clear,hls:clear,progressive:clear'
	19	_V4_BASE_URL = 'https://vas-v4.p7s1video.net/4.0/get'
	20
9d54b02b RA	21	def _extract_video_info(self, url, clip_id):
	22	client_location = url
	23
	24	video = self._download_json(
	25	'http://vas.sim-technik.de/vas/live/v2/videos',
	26	clip_id, 'Downloading videos JSON', query={
	27	'access_token': self._TOKEN,
	28	'client_location': client_location,
	29	'client_name': self._CLIENT_NAME,
	30	'ids': clip_id,
	31	})[0]
	32
a06916d9	33	if not self.get_param('allow_unplayable_formats') and video.get('is_protected') is True:
88acdbc2	34	self.report_drm(clip_id)
9d54b02b	35
c5eb75b3 RA	36	formats = []
	37	if self._ACCESS_ID:
	38	raw_ct = self._ENCRYPTION_KEY + clip_id + self._IV + self._ACCESS_ID
0c15a56f	39	protocols = self._download_json(
c5eb75b3 RA	40	self._V4_BASE_URL + 'protocols', clip_id,
	41	'Downloading protocols JSON',
	42	headers=self.geo_verification_headers(), query={
	43	'access_id': self._ACCESS_ID,
e3a3ed8a	44	'client_token': hashlib.sha1((raw_ct).encode()).hexdigest(),
c5eb75b3	45	'video_id': clip_id,
0c15a56f RA	46	}, fatal=False, expected_status=(403,)) or {}
	47	error = protocols.get('error') or {}
	48	if error.get('title') == 'Geo check failed':
	49	self.raise_geo_restricted(countries=['AT', 'CH', 'DE'])
	50	server_token = protocols.get('server_token')
c5eb75b3 RA	51	if server_token:
	52	urls = (self._download_json(
	53	self._V4_BASE_URL + 'urls', clip_id, 'Downloading urls JSON', query={
	54	'access_id': self._ACCESS_ID,
e3a3ed8a	55	'client_token': hashlib.sha1((raw_ct + server_token + self._SUPPORTED_PROTOCOLS).encode()).hexdigest(),
c5eb75b3 RA	56	'protocols': self._SUPPORTED_PROTOCOLS,
	57	'server_token': server_token,
	58	'video_id': clip_id,
	59	}, fatal=False) or {}).get('urls') or {}
	60	for protocol, variant in urls.items():
	61	source_url = variant.get('clear', {}).get('url')
	62	if not source_url:
	63	continue
	64	if protocol == 'dash':
	65	formats.extend(self._extract_mpd_formats(
	66	source_url, clip_id, mpd_id=protocol, fatal=False))
	67	elif protocol == 'hls':
	68	formats.extend(self._extract_m3u8_formats(
	69	source_url, clip_id, 'mp4', 'm3u8_native',
	70	m3u8_id=protocol, fatal=False))
	71	else:
	72	formats.append({
	73	'url': source_url,
	74	'format_id': protocol,
	75	})
	76	if not formats:
add96eb9	77	source_ids = [str(source['id']) for source in video['sources']]
9d54b02b	78
add96eb9	79	client_id = self._SALT[:2] + hashlib.sha1(''.join([clip_id, self._SALT, self._TOKEN, client_location, self._SALT, self._CLIENT_NAME]).encode()).hexdigest()
9d54b02b	80
c5eb75b3	81	sources = self._download_json(
add96eb9	82	f'http://vas.sim-technik.de/vas/live/v2/videos/{clip_id}/sources',
c5eb75b3	83	clip_id, 'Downloading sources JSON', query={
9d54b02b RA	84	'access_token': self._TOKEN,
	85	'client_id': client_id,
	86	'client_location': client_location,
	87	'client_name': self._CLIENT_NAME,
9d54b02b	88	})
c5eb75b3 RA	89	server_id = sources['server_id']
	90
	91	def fix_bitrate(bitrate):
	92	bitrate = int_or_none(bitrate)
	93	if not bitrate:
	94	return None
	95	return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
	96
	97	for source_id in source_ids:
add96eb9	98	client_id = self._SALT[:2] + hashlib.sha1(''.join([self._SALT, clip_id, self._TOKEN, server_id, client_location, source_id, self._SALT, self._CLIENT_NAME]).encode()).hexdigest()
c5eb75b3	99	urls = self._download_json(
add96eb9	100	f'http://vas.sim-technik.de/vas/live/v2/videos/{clip_id}/sources/url',
c5eb75b3 RA	101	clip_id, 'Downloading urls JSON', fatal=False, query={
	102	'access_token': self._TOKEN,
	103	'client_id': client_id,
	104	'client_location': client_location,
	105	'client_name': self._CLIENT_NAME,
	106	'server_id': server_id,
	107	'source_ids': source_id,
	108	})
	109	if not urls:
9d54b02b	110	continue
c5eb75b3 RA	111	if urls.get('status_code') != 0:
	112	raise ExtractorError('This video is unavailable', expected=True)
	113	urls_sources = urls['sources']
	114	if isinstance(urls_sources, dict):
	115	urls_sources = urls_sources.values()
	116	for source in urls_sources:
	117	source_url = source.get('url')
	118	if not source_url:
	119	continue
	120	protocol = source.get('protocol')
	121	mimetype = source.get('mimetype')
	122	if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
	123	formats.extend(self._extract_f4m_formats(
	124	source_url, clip_id, f4m_id='hds', fatal=False))
	125	elif mimetype == 'application/x-mpegURL':
	126	formats.extend(self._extract_m3u8_formats(
	127	source_url, clip_id, 'mp4', 'm3u8_native',
	128	m3u8_id='hls', fatal=False))
	129	elif mimetype == 'application/dash+xml':
	130	formats.extend(self._extract_mpd_formats(
	131	source_url, clip_id, mpd_id='dash', fatal=False))
9d54b02b	132	else:
c5eb75b3 RA	133	tbr = fix_bitrate(source['bitrate'])
	134	if protocol in ('rtmp', 'rtmpe'):
	135	mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
	136	if not mobj:
	137	continue
	138	path = mobj.group('path')
	139	mp4colon_index = path.rfind('mp4:')
	140	app = path[:mp4colon_index]
	141	play_path = path[mp4colon_index:]
	142	formats.append({
add96eb9	143	'url': '{}/{}'.format(mobj.group('url'), app),
c5eb75b3 RA	144	'app': app,
	145	'play_path': play_path,
	146	'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
	147	'page_url': 'http://www.prosieben.de',
	148	'tbr': tbr,
	149	'ext': 'flv',
add96eb9	150	'format_id': 'rtmp{}'.format(f'-{tbr}' if tbr else ''),
c5eb75b3 RA	151	})
	152	else:
	153	formats.append({
	154	'url': source_url,
	155	'tbr': tbr,
add96eb9	156	'format_id': 'http{}'.format(f'-{tbr}' if tbr else ''),
c5eb75b3	157	})
9d54b02b RA	158
9d54b02b RA	159	return {
c5eb75b3	160	'duration': float_or_none(video.get('duration')),
9d54b02b RA	161	'formats': formats,
	162	}
	163
	164
	165	class ProSiebenSat1IE(ProSiebenSat1BaseIE):
0c7214c4 S	166	IE_NAME = 'prosiebensat1'
0c7214c4 S	167	IE_DESC = 'ProSiebenSat.1 Digital'
63c583eb S	168	_VALID_URL = r'''(?x)
	169	https?://
	170	(?:www\.)?
	171	(?:
5a586082	172	(?:beta\.)?
63c583eb	173	(?:
2cdfe977	174	prosieben(?:maxx)?\|sixx\|sat1(?:gold)?\|kabeleins(?:doku)?\|the-voice-of-germany\|advopedia
63c583eb	175	)\.(?:de\|at\|ch)\|
6e3f23d9	176	ran\.de\|fem\.com\|advopedia\.de\|galileo\.tv/video
63c583eb S	177	)
	178	/(?P<id>.+)
	179	'''
0c7214c4 S	180
	181	_TESTS = [
	182	{
067aa17e S	183	# Tests changes introduced in https://github.com/ytdl-org/youtube-dl/pull/6242
067aa17e S	184	# in response to fixing https://github.com/ytdl-org/youtube-dl/issues/6215:
ab9b890b S	185	# - malformed f4m manifest support
	186	# - proper handling of URLs starting with `https?://` in 2.0 manifests
	187	# - recursive child f4m manifests extraction
0c7214c4 S	188	'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
	189	'info_dict': {
	190	'id': '2104602',
fe5aa197	191	'ext': 'mp4',
2cdfe977	192	'title': 'CIRCUS HALLIGALLI - Episode 18 - Staffel 2',
0c7214c4 S	193	'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
	194	'upload_date': '20131231',
	195	'duration': 5845.04,
38db9a40 S	196	'series': 'CIRCUS HALLIGALLI',
	197	'season_number': 2,
	198	'episode': 'Episode 18 - Staffel 2',
	199	'episode_number': 18,
0c7214c4	200	},
0c7214c4 S	201	},
	202	{
	203	'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
	204	'info_dict': {
	205	'id': '2570327',
	206	'ext': 'mp4',
	207	'title': 'Lady-Umstyling für Audrina',
	208	'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
	209	'upload_date': '20131014',
	210	'duration': 606.76,
	211	},
	212	'params': {
	213	# rtmp download
	214	'skip_download': True,
	215	},
	216	'skip': 'Seems to be broken',
	217	},
	218	{
6dadaa99	219	'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
0c7214c4	220	'info_dict': {
6dadaa99	221	'id': '2429369',
0c7214c4	222	'ext': 'mp4',
6dadaa99 S	223	'title': 'Countdown für die Autowerkstatt',
	224	'description': 'md5:809fc051a457b5d8666013bc40698817',
	225	'upload_date': '20140223',
	226	'duration': 2595.04,
0c7214c4 S	227	},
	228	'params': {
	229	# rtmp download
	230	'skip_download': True,
	231	},
84f214d8	232	'skip': 'This video is unavailable',
0c7214c4 S	233	},
	234	{
	235	'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
	236	'info_dict': {
	237	'id': '2904997',
	238	'ext': 'mp4',
	239	'title': 'Sexy laufen in Ugg Boots',
	240	'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
	241	'upload_date': '20140122',
	242	'duration': 245.32,
	243	},
	244	'params': {
	245	# rtmp download
	246	'skip_download': True,
	247	},
84f214d8	248	'skip': 'This video is unavailable',
0c7214c4 S	249	},
	250	{
	251	'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
	252	'info_dict': {
	253	'id': '2906572',
	254	'ext': 'mp4',
	255	'title': 'Im Interview: Kai Wiesinger',
	256	'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
8d1c8cae	257	'upload_date': '20140203',
0c7214c4 S	258	'duration': 522.56,
	259	},
	260	'params': {
	261	# rtmp download
	262	'skip_download': True,
	263	},
84f214d8	264	'skip': 'This video is unavailable',
0c7214c4 S	265	},
	266	{
	267	'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
	268	'info_dict': {
	269	'id': '2992323',
	270	'ext': 'mp4',
	271	'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
	272	'description': 'md5:2669cde3febe9bce13904f701e774eb6',
8d1c8cae	273	'upload_date': '20141014',
0c7214c4 S	274	'duration': 2410.44,
	275	},
	276	'params': {
	277	# rtmp download
	278	'skip_download': True,
	279	},
84f214d8	280	'skip': 'This video is unavailable',
0c7214c4 S	281	},
	282	{
	283	'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
	284	'info_dict': {
	285	'id': '3004256',
	286	'ext': 'mp4',
	287	'title': 'Schalke: Tönnies möchte Raul zurück',
	288	'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
	289	'upload_date': '20140226',
	290	'duration': 228.96,
	291	},
	292	'params': {
	293	# rtmp download
	294	'skip_download': True,
	295	},
84f214d8	296	'skip': 'This video is unavailable',
0c7214c4 S	297	},
	298	{
	299	'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
	300	'info_dict': {
	301	'id': '2572814',
fe5aa197	302	'ext': 'mp4',
2cdfe977	303	'title': 'The Voice of Germany - Andreas Kümmert: Rocket Man',
0c7214c4	304	'description': 'md5:6ddb02b0781c6adf778afea606652e38',
38db9a40	305	'timestamp': 1382041620,
0c7214c4 S	306	'upload_date': '20131017',
	307	'duration': 469.88,
	308	},
	309	'params': {
0c7214c4 S	310	'skip_download': True,
	311	},
	312	},
	313	{
2cdfe977	314	'url': 'http://www.fem.com/videos/beauty-lifestyle/kurztrips-zum-valentinstag',
0c7214c4 S	315	'info_dict': {
0c7214c4 S	316	'id': '2156342',
fe5aa197	317	'ext': 'mp4',
0c7214c4	318	'title': 'Kurztrips zum Valentinstag',
81549898	319	'description': 'Romantischer Kurztrip zum Valentinstag? Nina Heinemann verrät, was sich hier wirklich lohnt.',
0c7214c4 S	320	'duration': 307.24,
	321	},
	322	'params': {
0c7214c4 S	323	'skip_download': True,
	324	},
	325	},
c84890f7 AK	326	{
	327	'url': 'http://www.prosieben.de/tv/joko-gegen-klaas/videos/playlists/episode-8-ganze-folge-playlist',
	328	'info_dict': {
	329	'id': '439664',
	330	'title': 'Episode 8 - Ganze Folge - Playlist',
6a52eed8	331	'description': 'md5:63b8963e71f481782aeea877658dec84',
c84890f7 AK	332	},
c84890f7 AK	333	'playlist_count': 2,
fe5aa197	334	'skip': 'This video is unavailable',
c84890f7	335	},
76bee08f S	336	{
	337	# title in <h2 class="subtitle">
	338	'url': 'http://www.prosieben.de/stars/oscar-award/videos/jetzt-erst-enthuellt-das-geheimnis-von-emma-stones-oscar-robe-clip',
	339	'info_dict': {
	340	'id': '4895826',
	341	'ext': 'mp4',
	342	'title': 'Jetzt erst enthüllt: Das Geheimnis von Emma Stones Oscar-Robe',
	343	'description': 'md5:e5ace2bc43fadf7b63adc6187e9450b9',
	344	'upload_date': '20170302',
	345	},
	346	'params': {
	347	'skip_download': True,
	348	},
	349	'skip': 'geo restricted to Germany',
	350	},
71ad00c0 S	351	{
	352	# geo restricted to Germany
	353	'url': 'http://www.kabeleinsdoku.de/tv/mayday-alarm-im-cockpit/video/102-notlandung-im-hudson-river-ganze-folge',
	354	'only_matching': True,
	355	},
63c583eb S	356	{
	357	# geo restricted to Germany
	358	'url': 'http://www.sat1gold.de/tv/edel-starck/video/11-staffel-1-episode-1-partner-wider-willen-ganze-folge',
	359	'only_matching': True,
	360	},
6e3f23d9	361	{
	362	# geo restricted to Germany
	363	'url': 'https://www.galileo.tv/video/diese-emojis-werden-oft-missverstanden',
	364	'only_matching': True,
	365	},
ddde9195 S	366	{
	367	'url': 'http://www.sat1gold.de/tv/edel-starck/playlist/die-gesamte-1-staffel',
	368	'only_matching': True,
	369	},
493353c7 S	370	{
	371	'url': 'http://www.advopedia.de/videos/lenssen-klaert-auf/lenssen-klaert-auf-folge-8-staffel-3-feiertage-und-freie-tage',
	372	'only_matching': True,
	373	},
0c7214c4 S	374	]
0c7214c4 S	375
9d54b02b RA	376	_TOKEN = 'prosieben'
	377	_SALT = '01!8d8F_)r9]4s[qeuXfP%'
	378	_CLIENT_NAME = 'kolibri-2.0.19-splec4'
c5eb75b3 RA	379
	380	_ACCESS_ID = 'x_prosiebenmaxx-de'
	381	_ENCRYPTION_KEY = 'Eeyeey9oquahthainoofashoyoikosag'
	382	_IV = 'Aeluchoc6aevechuipiexeeboowedaok'
	383
0c7214c4 S	384	_CLIPID_REGEXES = [
	385	r'"clip_id"\s*:\s+"(\d+)"',
	386	r'clipid: "(\d+)"',
38a9339b	387	r'clip[iI]d=(\d+)',
6e3f23d9	388	r'clip[iI][dD]\s=\s["\'](\d+)',
8d1c8cae	389	r"'itemImageUrl'\s:\s'/dynamic/thumbnails/full/\d+/(\d+)",
721a0c3c	390	r'proMamsId"\s:\s"(\d+)',
967ebbdb	391	r'proMamsId"\s:\s"(\d+)',
0c7214c4 S	392	]
	393	_TITLE_REGEXES = [
	394	r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
	395	r'<header class="clearfix">\s*<h3>(.+?)</h3>',
	396	r'<!-- start video -->\s*<h1>(.+?)</h1>',
38a9339b	397	r'<h1 class="att-name">\s*(.+?)</h1>',
8b6c896c	398	r'<header class="module_header">\s<h2>([^<]+)</h2>\s</header>',
0baedd18	399	r'<h2 class="video-title" itemprop="name">\s*(.+?)</h2>',
81549898	400	r'<div[^>]+id="veeseoTitle"[^>]*>(.+?)</div>',
76bee08f	401	r'<h2[^>]+class="subtitle"[^>]*>([^<]+)</h2>',
0c7214c4 S	402	]
	403	_DESCRIPTION_REGEXES = [
	404	r'<p itemprop="description">\s*(.+?)</p>',
	405	r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
	406	r'<div class="g-plusone" data-size="medium"></div>\s</div>\s</header>\s(.+?)\s<footer>',
38a9339b	407	r'<p class="att-description">\s(.+?)\s</p>',
0baedd18	408	r'<p class="video-description" itemprop="description">\s*(.+?)</p>',
81549898	409	r'<div[^>]+id="veeseoDescription"[^>]*>(.+?)</div>',
0c7214c4 S	410	]
0c7214c4 S	411	_UPLOAD_DATE_REGEXES = [
0c7214c4 S	412	r'<span>\s(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \\|\s<span itemprop="duration"',
	413	r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
	414	r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
	415	r'(\d{2}\.\d{2}\.\d{4}) \\| \d{2}:\d{2} Min<br/>',
	416	]
6a52eed8 S	417	_PAGE_TYPE_REGEXES = [
6a52eed8 S	418	r'<meta name="page_type" content="([^"]+)">',
c84890f7 AK	419	r"'itemType'\s:\s'([^']*)'",
c84890f7 AK	420	]
6a52eed8 S	421	_PLAYLIST_ID_REGEXES = [
6a52eed8 S	422	r'content[iI]d=(\d+)',
c84890f7 AK	423	r"'itemId'\s:\s'([^']*)'",
c84890f7 AK	424	]
6a52eed8 S	425	_PLAYLIST_CLIP_REGEXES = [
6a52eed8 S	426	r'(?s)data-qvt=.+?<a href="([^"]+)"',
c84890f7	427	]
0c7214c4	428
6a52eed8	429	def _extract_clip(self, url, webpage):
84f214d8 RA	430	clip_id = self._html_search_regex(
84f214d8 RA	431	self._CLIPID_REGEXES, webpage, 'clip id')
76bee08f S	432	title = self._html_search_regex(
	433	self._TITLE_REGEXES, webpage, 'title',
	434	default=None) or self._og_search_title(webpage)
9d54b02b	435	info = self._extract_video_info(url, clip_id)
84f214d8	436	description = self._html_search_regex(
7882f111	437	self._DESCRIPTION_REGEXES, webpage, 'description', default=None)
fe5aa197	438	if description is None:
7882f111	439	description = self._og_search_description(webpage)
84f214d8	440	thumbnail = self._og_search_thumbnail(webpage)
2cdfe977 PS	441	upload_date = unified_strdate(
	442	self._html_search_meta('og:published_time', webpage,
	443	'upload date', default=None)
	444	or self._html_search_regex(self._UPLOAD_DATE_REGEXES,
	445	webpage, 'upload date', default=None))
84f214d8	446
38db9a40 S	447	json_ld = self._search_json_ld(webpage, clip_id, default={})
	448
	449	return merge_dicts(info, {
0c7214c4 S	450	'id': clip_id,
	451	'title': title,
	452	'description': description,
	453	'thumbnail': thumbnail,
	454	'upload_date': upload_date,
38db9a40	455	}, json_ld)
6a52eed8 S	456
	457	def _extract_playlist(self, url, webpage):
	458	playlist_id = self._html_search_regex(
	459	self._PLAYLIST_ID_REGEXES, webpage, 'playlist id')
ddde9195 S	460	playlist = self._parse_json(
ddde9195 S	461	self._search_regex(
ec85ded8	462	r'var\s+contentResources\s=\s(\[.+?\]);\s*</script',
ddde9195 S	463	webpage, 'playlist'),
	464	playlist_id)
	465	entries = []
	466	for item in playlist:
	467	clip_id = item.get('id') or item.get('upc')
	468	if not clip_id:
	469	continue
	470	info = self._extract_video_info(url, clip_id)
	471	info.update({
	472	'id': clip_id,
	473	'title': item.get('title') or item.get('teaser', {}).get('headline'),
	474	'description': item.get('teaser', {}).get('description'),
	475	'thumbnail': item.get('poster'),
	476	'duration': float_or_none(item.get('duration')),
	477	'series': item.get('tvShowTitle'),
	478	'uploader': item.get('broadcastPublisher'),
	479	})
	480	entries.append(info)
	481	return self.playlist_result(entries, playlist_id)
6a52eed8 S	482
	483	def _real_extract(self, url):
	484	video_id = self._match_id(url)
	485	webpage = self._download_webpage(url, video_id)
	486	page_type = self._search_regex(
	487	self._PAGE_TYPE_REGEXES, webpage,
	488	'page type', default='clip').lower()
	489	if page_type == 'clip':
	490	return self._extract_clip(url, webpage)
	491	elif page_type == 'playlist':
	492	return self._extract_playlist(url, webpage)
8ffb8e63 S	493	else:
8ffb8e63 S	494	raise ExtractorError(
add96eb9	495	f'Unsupported page type {page_type}', expected=True)