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

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..compat import compat_str
from ..utils import (
    determine_ext,
    float_or_none,
    int_or_none,
    join_nonempty,
    merge_dicts,
    NO_DEFAULT,
    orderedSet,
    parse_codecs,
    qualities,
    try_get,
    unified_timestamp,
    update_url_query,
    url_or_none,
    urljoin,
)


class ZDFBaseIE(InfoExtractor):
    _GEO_COUNTRIES = ['DE']
    _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')

    def _call_api(self, url, video_id, item, api_token=None, referrer=None):
        headers = {}
        if api_token:
            headers['Api-Auth'] = 'Bearer %s' % api_token
        if referrer:
            headers['Referer'] = referrer
        return self._download_json(
            url, video_id, 'Downloading JSON %s' % item, headers=headers)

    @staticmethod
    def _extract_subtitles(src):
        subtitles = {}
        for caption in try_get(src, lambda x: x['captions'], list) or []:
            subtitle_url = url_or_none(caption.get('uri'))
            if subtitle_url:
                lang = caption.get('language', 'deu')
                subtitles.setdefault(lang, []).append({
                    'url': subtitle_url,
                })
        return subtitles

    def _extract_format(self, video_id, formats, format_urls, meta):
        format_url = url_or_none(meta.get('url'))
        if not format_url or format_url in format_urls:
            return
        format_urls.add(format_url)

        mime_type, ext = meta.get('mimeType'), determine_ext(format_url)
        if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
            new_formats = self._extract_m3u8_formats(
                format_url, video_id, 'mp4', m3u8_id='hls',
                entry_protocol='m3u8_native', fatal=False)
        elif mime_type == 'application/f4m+xml' or ext == 'f4m':
            new_formats = self._extract_f4m_formats(
                update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False)
        else:
            f = parse_codecs(meta.get('mimeCodec'))
            if not f and meta.get('type'):
                data = meta['type'].split('_')
                if try_get(data, lambda x: x[2]) == ext:
                    f = {'vcodec': data[0], 'acodec': data[1]}
            f.update({
                'url': format_url,
                'format_id': join_nonempty('http', meta.get('type'), meta.get('quality')),
            })
            new_formats = [f]
        formats.extend(merge_dicts(f, {
            'format_note': join_nonempty('quality', 'class', from_dict=meta, delim=', '),
            'language': meta.get('language'),
            'language_preference': 10 if meta.get('class') == 'main' else -10 if meta.get('class') == 'ad' else -1,
            'quality': qualities(self._QUALITIES)(meta.get('quality')),
        }) for f in new_formats)

    def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer):
        ptmd = self._call_api(
            ptmd_url, video_id, 'metadata', api_token, referrer)

        content_id = ptmd.get('basename') or ptmd_url.split('/')[-1]

        formats = []
        track_uris = set()
        for p in ptmd['priorityList']:
            formitaeten = p.get('formitaeten')
            if not isinstance(formitaeten, list):
                continue
            for f in formitaeten:
                f_qualities = f.get('qualities')
                if not isinstance(f_qualities, list):
                    continue
                for quality in f_qualities:
                    tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
                    if not tracks:
                        continue
                    for track in tracks:
                        self._extract_format(
                            content_id, formats, track_uris, {
                                'url': track.get('uri'),
                                'type': f.get('type'),
                                'mimeType': f.get('mimeType'),
                                'quality': quality.get('quality'),
                                'class': track.get('class'),
                                'language': track.get('language'),
                            })
        self._sort_formats(formats, ('hasaud', 'res', 'quality', 'language_preference'))

        duration = float_or_none(try_get(
            ptmd, lambda x: x['attributes']['duration']['value']), scale=1000)

        return {
            'extractor_key': ZDFIE.ie_key(),
            'id': content_id,
            'duration': duration,
            'formats': formats,
            'subtitles': self._extract_subtitles(ptmd),
        }

    def _extract_player(self, webpage, video_id, fatal=True):
        return self._parse_json(
            self._search_regex(
                r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
                'player JSON', default='{}' if not fatal else NO_DEFAULT,
                group='json'),
            video_id)


class ZDFIE(ZDFBaseIE):
    _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
    _TESTS = [{
        # Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html
        'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html',
        'md5': '34ec321e7eb34231fd88616c65c92db0',
        'info_dict': {
            'id': '210222_phx_nachgehakt_corona_protest',
            'ext': 'mp4',
            'title': 'Wohin führt der Protest in der Pandemie?',
            'description': 'md5:7d643fe7f565e53a24aac036b2122fbd',
            'duration': 1691,
            'timestamp': 1613948400,
            'upload_date': '20210221',
        },
    }, {
        # Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html
        'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html',
        'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
        'info_dict': {
            'id': '141007_ab18_10wochensommer_film',
            'ext': 'mp4',
            'title': 'Ab 18! - 10 Wochen Sommer',
            'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26',
            'duration': 2660,
            'timestamp': 1608604200,
            'upload_date': '20201222',
        },
    }, {
        'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
        'info_dict': {
            'id': '151025_magie_farben2_tex',
            'ext': 'mp4',
            'title': 'Die Magie der Farben (2/2)',
            'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
            'duration': 2615,
            'timestamp': 1465021200,
            'upload_date': '20160604',
        },
    }, {
        # Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
        'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
        'only_matching': True,
    }, {
        # Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
        'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
        'only_matching': True,
    }, {
        # Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
        'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
        'only_matching': True,
    }, {
        'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
        'only_matching': True,
    }, {
        'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
        'only_matching': True,
    }, {
        'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
        'only_matching': True,
    }]

    def _extract_entry(self, url, player, content, video_id):
        title = content.get('title') or content['teaserHeadline']

        t = content['mainVideoContent']['http://zdf.de/rels/target']

        ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')

        if not ptmd_path:
            ptmd_path = t[
                'http://zdf.de/rels/streams/ptmd-template'].replace(
                '{playerId}', 'ngplayer_2_4')

        info = self._extract_ptmd(
            urljoin(url, ptmd_path), video_id, player['apiToken'], url)

        thumbnails = []
        layouts = try_get(
            content, lambda x: x['teaserImageRef']['layouts'], dict)
        if layouts:
            for layout_key, layout_url in layouts.items():
                layout_url = url_or_none(layout_url)
                if not layout_url:
                    continue
                thumbnail = {
                    'url': layout_url,
                    'format_id': layout_key,
                }
                mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
                if mobj:
                    thumbnail.update({
                        'width': int(mobj.group('width')),
                        'height': int(mobj.group('height')),
                    })
                thumbnails.append(thumbnail)

        return merge_dicts(info, {
            'title': title,
            'description': content.get('leadParagraph') or content.get('teasertext'),
            'duration': int_or_none(t.get('duration')),
            'timestamp': unified_timestamp(content.get('editorialDate')),
            'thumbnails': thumbnails,
        })

    def _extract_regular(self, url, player, video_id):
        content = self._call_api(
            player['content'], video_id, 'content', player['apiToken'], url)
        return self._extract_entry(player['content'], player, content, video_id)

    def _extract_mobile(self, video_id):
        video = self._download_json(
            'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
            video_id)

        document = video['document']

        title = document['titel']
        content_id = document['basename']

        formats = []
        format_urls = set()
        for f in document['formitaeten']:
            self._extract_format(content_id, formats, format_urls, f)
        self._sort_formats(formats)

        thumbnails = []
        teaser_bild = document.get('teaserBild')
        if isinstance(teaser_bild, dict):
            for thumbnail_key, thumbnail in teaser_bild.items():
                thumbnail_url = try_get(
                    thumbnail, lambda x: x['url'], compat_str)
                if thumbnail_url:
                    thumbnails.append({
                        'url': thumbnail_url,
                        'id': thumbnail_key,
                        'width': int_or_none(thumbnail.get('width')),
                        'height': int_or_none(thumbnail.get('height')),
                    })

        return {
            'id': content_id,
            'title': title,
            'description': document.get('beschreibung'),
            'duration': int_or_none(document.get('length')),
            'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
                try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
            'thumbnails': thumbnails,
            'subtitles': self._extract_subtitles(document),
            'formats': formats,
        }

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

        webpage = self._download_webpage(url, video_id, fatal=False)
        if webpage:
            player = self._extract_player(webpage, url, fatal=False)
            if player:
                return self._extract_regular(url, player, video_id)

        return self._extract_mobile(video_id)


class ZDFChannelIE(ZDFBaseIE):
    _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
    _TESTS = [{
        'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
        'info_dict': {
            'id': 'das-aktuelle-sportstudio',
            'title': 'das aktuelle sportstudio | ZDF',
        },
        'playlist_mincount': 23,
    }, {
        'url': 'https://www.zdf.de/dokumentation/planet-e',
        'info_dict': {
            'id': 'planet-e',
            'title': 'planet e.',
        },
        'playlist_mincount': 50,
    }, {
        'url': 'https://www.zdf.de/filme/taunuskrimi/',
        'only_matching': True,
    }]

    @classmethod
    def suitable(cls, url):
        return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)

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

        webpage = self._download_webpage(url, channel_id)

        entries = [
            self.url_result(item_url, ie=ZDFIE.ie_key())
            for item_url in orderedSet(re.findall(
                r'data-plusbar-url=["\'](http.+?\.html)', webpage))]

        return self.playlist_result(
            entries, channel_id, self._og_search_title(webpage, fatal=False))

        r"""
        player = self._extract_player(webpage, channel_id)

        channel_id = self._search_regex(
            r'docId\s*:\s*(["\'])(?P<id>(?!\1).+?)\1', webpage,
            'channel id', group='id')

        channel = self._call_api(
            'https://api.zdf.de/content/documents/%s.json' % channel_id,
            player, url, channel_id)

        items = []
        for module in channel['module']:
            for teaser in try_get(module, lambda x: x['teaser'], list) or []:
                t = try_get(
                    teaser, lambda x: x['http://zdf.de/rels/target'], dict)
                if not t:
                    continue
                items.extend(try_get(
                    t,
                    lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
                    list) or [])
            items.extend(try_get(
                module,
                lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
                list) or [])

        entries = []
        entry_urls = set()
        for item in items:
            t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
            if not t:
                continue
            sharing_url = t.get('http://zdf.de/rels/sharing-url')
            if not sharing_url or not isinstance(sharing_url, compat_str):
                continue
            if sharing_url in entry_urls:
                continue
            entry_urls.add(sharing_url)
            entries.append(self.url_result(
                sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))

        return self.playlist_result(entries, channel_id, channel.get('title'))
        """
Commit	Line	Data
a2e6db36	1	# coding: utf-8
919052d0	2	from __future__ import unicode_literals
a2e6db36	3
d5822b96 PH	4	import re
	5
	6	from .common import InfoExtractor
fb47cb5b	7	from ..compat import compat_str
d5822b96	8	from ..utils import (
a5c1d955	9	determine_ext,
ec5e77c5	10	float_or_none,
fb47cb5b	11	int_or_none,
34921b43	12	join_nonempty,
ec5e77c5	13	merge_dicts,
fb47cb5b S	14	NO_DEFAULT,
	15	orderedSet,
	16	parse_codecs,
	17	qualities,
	18	try_get,
	19	unified_timestamp,
	20	update_url_query,
3052a30d	21	url_or_none,
fb47cb5b	22	urljoin,
d5822b96	23	)
0b7c2485	24
fb47cb5b S	25
fb47cb5b S	26	class ZDFBaseIE(InfoExtractor):
9cf26b6e	27	_GEO_COUNTRIES = ['DE']
ec5e77c5	28	_QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd')
a2e6db36	29
ec5e77c5	30	def _call_api(self, url, video_id, item, api_token=None, referrer=None):
	31	headers = {}
	32	if api_token:
	33	headers['Api-Auth'] = 'Bearer %s' % api_token
	34	if referrer:
	35	headers['Referer'] = referrer
	36	return self._download_json(
	37	url, video_id, 'Downloading JSON %s' % item, headers=headers)
a5c1d955	38
fb47cb5b S	39	@staticmethod
	40	def _extract_subtitles(src):
	41	subtitles = {}
	42	for caption in try_get(src, lambda x: x['captions'], list) or []:
3052a30d S	43	subtitle_url = url_or_none(caption.get('uri'))
3052a30d S	44	if subtitle_url:
fb47cb5b S	45	lang = caption.get('language', 'deu')
	46	subtitles.setdefault(lang, []).append({
	47	'url': subtitle_url,
	48	})
	49	return subtitles
	50
	51	def _extract_format(self, video_id, formats, format_urls, meta):
3052a30d	52	format_url = url_or_none(meta.get('url'))
600e9003	53	if not format_url or format_url in format_urls:
fb47cb5b S	54	return
fb47cb5b S	55	format_urls.add(format_url)
600e9003	56
600e9003	57	mime_type, ext = meta.get('mimeType'), determine_ext(format_url)
fb47cb5b	58	if mime_type == 'application/x-mpegURL' or ext == 'm3u8':
600e9003	59	new_formats = self._extract_m3u8_formats(
fb47cb5b	60	format_url, video_id, 'mp4', m3u8_id='hls',
600e9003	61	entry_protocol='m3u8_native', fatal=False)
fb47cb5b	62	elif mime_type == 'application/f4m+xml' or ext == 'f4m':
600e9003	63	new_formats = self._extract_f4m_formats(
600e9003	64	update_url_query(format_url, {'hdcore': '3.7.0'}), video_id, f4m_id='hds', fatal=False)
fb47cb5b S	65	else:
fb47cb5b S	66	f = parse_codecs(meta.get('mimeCodec'))
600e9003	67	if not f and meta.get('type'):
	68	data = meta['type'].split('_')
	69	if try_get(data, lambda x: x[2]) == ext:
	70	f = {'vcodec': data[0], 'acodec': data[1]}
fb47cb5b S	71	f.update({
fb47cb5b S	72	'url': format_url,
34921b43	73	'format_id': join_nonempty('http', meta.get('type'), meta.get('quality')),
fb47cb5b	74	})
600e9003	75	new_formats = [f]
600e9003	76	formats.extend(merge_dicts(f, {
34921b43	77	'format_note': join_nonempty('quality', 'class', from_dict=meta, delim=', '),
600e9003	78	'language': meta.get('language'),
	79	'language_preference': 10 if meta.get('class') == 'main' else -10 if meta.get('class') == 'ad' else -1,
	80	'quality': qualities(self._QUALITIES)(meta.get('quality')),
	81	}) for f in new_formats)
fb47cb5b	82
ec5e77c5	83	def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer):
50de3dba	84	ptmd = self._call_api(
ec5e77c5	85	ptmd_url, video_id, 'metadata', api_token, referrer)
	86
	87	content_id = ptmd.get('basename') or ptmd_url.split('/')[-1]
fb47cb5b	88
a5c1d955	89	formats = []
fb47cb5b S	90	track_uris = set()
	91	for p in ptmd['priorityList']:
	92	formitaeten = p.get('formitaeten')
	93	if not isinstance(formitaeten, list):
a5c1d955	94	continue
fb47cb5b S	95	for f in formitaeten:
	96	f_qualities = f.get('qualities')
	97	if not isinstance(f_qualities, list):
	98	continue
	99	for quality in f_qualities:
	100	tracks = try_get(quality, lambda x: x['audio']['tracks'], list)
	101	if not tracks:
	102	continue
	103	for track in tracks:
	104	self._extract_format(
ec5e77c5	105	content_id, formats, track_uris, {
fb47cb5b S	106	'url': track.get('uri'),
	107	'type': f.get('type'),
	108	'mimeType': f.get('mimeType'),
	109	'quality': quality.get('quality'),
600e9003	110	'class': track.get('class'),
fb47cb5b S	111	'language': track.get('language'),
fb47cb5b S	112	})
600e9003	113	self._sort_formats(formats, ('hasaud', 'res', 'quality', 'language_preference'))
fb47cb5b	114
ec5e77c5	115	duration = float_or_none(try_get(
	116	ptmd, lambda x: x['attributes']['duration']['value']), scale=1000)
	117
	118	return {
	119	'extractor_key': ZDFIE.ie_key(),
	120	'id': content_id,
	121	'duration': duration,
	122	'formats': formats,
	123	'subtitles': self._extract_subtitles(ptmd),
	124	}
	125
	126	def _extract_player(self, webpage, video_id, fatal=True):
	127	return self._parse_json(
	128	self._search_regex(
	129	r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage,
	130	'player JSON', default='{}' if not fatal else NO_DEFAULT,
	131	group='json'),
	132	video_id)
	133
	134
	135	class ZDFIE(ZDFBaseIE):
	136	_VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html'
	137	_TESTS = [{
	138	# Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html
	139	'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html',
	140	'md5': '34ec321e7eb34231fd88616c65c92db0',
	141	'info_dict': {
	142	'id': '210222_phx_nachgehakt_corona_protest',
	143	'ext': 'mp4',
	144	'title': 'Wohin führt der Protest in der Pandemie?',
	145	'description': 'md5:7d643fe7f565e53a24aac036b2122fbd',
	146	'duration': 1691,
	147	'timestamp': 1613948400,
	148	'upload_date': '20210221',
	149	},
	150	}, {
	151	# Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html
	152	'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html',
	153	'md5': '0aff3e7bc72c8813f5e0fae333316a1d',
	154	'info_dict': {
	155	'id': '141007_ab18_10wochensommer_film',
	156	'ext': 'mp4',
	157	'title': 'Ab 18! - 10 Wochen Sommer',
	158	'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26',
	159	'duration': 2660,
	160	'timestamp': 1608604200,
	161	'upload_date': '20201222',
	162	},
	163	}, {
	164	'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html',
	165	'info_dict': {
	166	'id': '151025_magie_farben2_tex',
	167	'ext': 'mp4',
	168	'title': 'Die Magie der Farben (2/2)',
	169	'description': 'md5:a89da10c928c6235401066b60a6d5c1a',
	170	'duration': 2615,
	171	'timestamp': 1465021200,
	172	'upload_date': '20160604',
	173	},
	174	}, {
	175	# Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche
	176	'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html',
	177	'only_matching': True,
	178	}, {
179	# Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html
180	'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html',
181	'only_matching': True,
182	}, {
183	# Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids
184	'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html',
185	'only_matching': True,
186	}, {
187	'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html',
188	'only_matching': True,
189	}, {
190	'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html',
191	'only_matching': True,
192	}, {
193	'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html',
194	'only_matching': True,
195	}]
196
197	def _extract_entry(self, url, player, content, video_id):
198	title = content.get('title') or content['teaserHeadline']
199
200	t = content['mainVideoContent']['http://zdf.de/rels/target']
201
202	ptmd_path = t.get('http://zdf.de/rels/streams/ptmd')
203
204	if not ptmd_path:
205	ptmd_path = t[
206	'http://zdf.de/rels/streams/ptmd-template'].replace(
207	'{playerId}', 'ngplayer_2_4')
208
209	info = self._extract_ptmd(
210	urljoin(url, ptmd_path), video_id, player['apiToken'], url)
211
fb47cb5b S	212	thumbnails = []
	213	layouts = try_get(
	214	content, lambda x: x['teaserImageRef']['layouts'], dict)
	215	if layouts:
	216	for layout_key, layout_url in layouts.items():
3052a30d S	217	layout_url = url_or_none(layout_url)
3052a30d S	218	if not layout_url:
fb47cb5b S	219	continue
	220	thumbnail = {
	221	'url': layout_url,
	222	'format_id': layout_key,
	223	}
	224	mobj = re.search(r'(?P<width>\d+)x(?P<height>\d+)', layout_key)
	225	if mobj:
	226	thumbnail.update({
	227	'width': int(mobj.group('width')),
	228	'height': int(mobj.group('height')),
	229	})
	230	thumbnails.append(thumbnail)
a5c1d955	231
ec5e77c5	232	return merge_dicts(info, {
fb47cb5b S	233	'title': title,
	234	'description': content.get('leadParagraph') or content.get('teasertext'),
	235	'duration': int_or_none(t.get('duration')),
	236	'timestamp': unified_timestamp(content.get('editorialDate')),
	237	'thumbnails': thumbnails,
ec5e77c5	238	})
d5822b96	239
fb47cb5b	240	def _extract_regular(self, url, player, video_id):
50de3dba	241	content = self._call_api(
ec5e77c5	242	player['content'], video_id, 'content', player['apiToken'], url)
50de3dba	243	return self._extract_entry(player['content'], player, content, video_id)
b6de53ea	244
fb47cb5b	245	def _extract_mobile(self, video_id):
ec5e77c5	246	video = self._download_json(
fb47cb5b	247	'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id,
ec5e77c5	248	video_id)
	249
	250	document = video['document']
b6de53ea	251
fb47cb5b	252	title = document['titel']
ec5e77c5	253	content_id = document['basename']
b6de53ea	254
fb47cb5b S	255	formats = []
	256	format_urls = set()
	257	for f in document['formitaeten']:
ec5e77c5	258	self._extract_format(content_id, formats, format_urls, f)
fb47cb5b S	259	self._sort_formats(formats)
	260
	261	thumbnails = []
	262	teaser_bild = document.get('teaserBild')
	263	if isinstance(teaser_bild, dict):
	264	for thumbnail_key, thumbnail in teaser_bild.items():
	265	thumbnail_url = try_get(
	266	thumbnail, lambda x: x['url'], compat_str)
	267	if thumbnail_url:
	268	thumbnails.append({
	269	'url': thumbnail_url,
	270	'id': thumbnail_key,
	271	'width': int_or_none(thumbnail.get('width')),
	272	'height': int_or_none(thumbnail.get('height')),
	273	})
b6de53ea	274
fb47cb5b	275	return {
ec5e77c5	276	'id': content_id,
fb47cb5b S	277	'title': title,
	278	'description': document.get('beschreibung'),
	279	'duration': int_or_none(document.get('length')),
ec5e77c5	280	'timestamp': unified_timestamp(document.get('date')) or unified_timestamp(
ec5e77c5	281	try_get(video, lambda x: x['meta']['editorialDate'], compat_str)),
fb47cb5b S	282	'thumbnails': thumbnails,
	283	'subtitles': self._extract_subtitles(document),
	284	'formats': formats,
	285	}
b6de53ea	286
fb47cb5b S	287	def _real_extract(self, url):
fb47cb5b S	288	video_id = self._match_id(url)
b6de53ea	289
fb47cb5b S	290	webpage = self._download_webpage(url, video_id, fatal=False)
	291	if webpage:
	292	player = self._extract_player(webpage, url, fatal=False)
	293	if player:
	294	return self._extract_regular(url, player, video_id)
b6de53ea	295
fb47cb5b	296	return self._extract_mobile(video_id)
b6de53ea	297
b6de53ea	298
fb47cb5b S	299	class ZDFChannelIE(ZDFBaseIE):
fb47cb5b S	300	_VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)'
c2404463	301	_TESTS = [{
fb47cb5b	302	'url': 'https://www.zdf.de/sport/das-aktuelle-sportstudio',
8560c618	303	'info_dict': {
fb47cb5b S	304	'id': 'das-aktuelle-sportstudio',
fb47cb5b S	305	'title': 'das aktuelle sportstudio \| ZDF',
8560c618	306	},
b4cbdbd4	307	'playlist_mincount': 23,
c2404463	308	}, {
fb47cb5b S	309	'url': 'https://www.zdf.de/dokumentation/planet-e',
	310	'info_dict': {
	311	'id': 'planet-e',
	312	'title': 'planet e.',
	313	},
b4cbdbd4	314	'playlist_mincount': 50,
c2404463	315	}, {
fb47cb5b	316	'url': 'https://www.zdf.de/filme/taunuskrimi/',
c2404463 S	317	'only_matching': True,
c2404463 S	318	}]
fb47cb5b S	319
	320	@classmethod
	321	def suitable(cls, url):
	322	return False if ZDFIE.suitable(url) else super(ZDFChannelIE, cls).suitable(url)
9abd500a PH	323
	324	def _real_extract(self, url):
	325	channel_id = self._match_id(url)
8560c618	326
fb47cb5b S	327	webpage = self._download_webpage(url, channel_id)
	328
	329	entries = [
	330	self.url_result(item_url, ie=ZDFIE.ie_key())
	331	for item_url in orderedSet(re.findall(
	332	r'data-plusbar-url=["\'](http.+?\.html)', webpage))]
	333
	334	return self.playlist_result(
	335	entries, channel_id, self._og_search_title(webpage, fatal=False))
	336
ec85ded8	337	r"""
fb47cb5b S	338	player = self._extract_player(webpage, channel_id)
	339
	340	channel_id = self._search_regex(
	341	r'docId\s:\s(["\'])(?P<id>(?!\1).+?)\1', webpage,
	342	'channel id', group='id')
	343
	344	channel = self._call_api(
	345	'https://api.zdf.de/content/documents/%s.json' % channel_id,
	346	player, url, channel_id)
	347
	348	items = []
	349	for module in channel['module']:
	350	for teaser in try_get(module, lambda x: x['teaser'], list) or []:
	351	t = try_get(
	352	teaser, lambda x: x['http://zdf.de/rels/target'], dict)
	353	if not t:
	354	continue
	355	items.extend(try_get(
	356	t,
	357	lambda x: x['resultsWithVideo']['http://zdf.de/rels/search/results'],
	358	list) or [])
	359	items.extend(try_get(
	360	module,
	361	lambda x: x['filterRef']['resultsWithVideo']['http://zdf.de/rels/search/results'],
	362	list) or [])
	363
	364	entries = []
	365	entry_urls = set()
	366	for item in items:
	367	t = try_get(item, lambda x: x['http://zdf.de/rels/target'], dict)
	368	if not t:
	369	continue
	370	sharing_url = t.get('http://zdf.de/rels/sharing-url')
	371	if not sharing_url or not isinstance(sharing_url, compat_str):
	372	continue
	373	if sharing_url in entry_urls:
	374	continue
	375	entry_urls.add(sharing_url)
	376	entries.append(self.url_result(
	377	sharing_url, ie=ZDFIE.ie_key(), video_id=t.get('id')))
	378
	379	return self.playlist_result(entries, channel_id, channel.get('title'))
	380	"""