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

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from .generic import GenericIE
from ..utils import (
    determine_ext,
    ExtractorError,
    get_element_by_attribute,
    qualities,
    int_or_none,
    parse_duration,
    unified_strdate,
    xpath_text,
)
from ..compat import compat_etree_fromstring


class ARDMediathekIE(InfoExtractor):
    IE_NAME = 'ARD:mediathek'
    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'

    _TESTS = [{
        'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
        'info_dict': {
            'id': '29582122',
            'ext': 'mp4',
            'title': 'Ich liebe das Leben trotzdem',
            'description': 'md5:45e4c225c72b27993314b31a84a5261c',
            'duration': 4557,
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }, {
        'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
        'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
        'info_dict': {
            'id': '29522730',
            'ext': 'mp4',
            'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)',
            'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
            'duration': 5252,
        },
    }, {
        # audio
        'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
        'md5': '219d94d8980b4f538c7fcb0865eb7f2c',
        'info_dict': {
            'id': '28488308',
            'ext': 'mp3',
            'title': 'Tod eines Fußballers',
            'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
            'duration': 3240,
        },
    }, {
        'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
        'only_matching': True,
    }]

    def _extract_media_info(self, media_info_url, webpage, video_id):
        media_info = self._download_json(
            media_info_url, video_id, 'Downloading media JSON')

        formats = self._extract_formats(media_info, video_id)

        if not formats:
            if '"fsk"' in webpage:
                raise ExtractorError(
                    'This video is only available after 20:00', expected=True)
            elif media_info.get('_geoblocked'):
                raise ExtractorError('This video is not available due to geo restriction', expected=True)

        self._sort_formats(formats)

        duration = int_or_none(media_info.get('_duration'))
        thumbnail = media_info.get('_previewImage')

        subtitles = {}
        subtitle_url = media_info.get('_subtitleUrl')
        if subtitle_url:
            subtitles['de'] = [{
                'ext': 'srt',
                'url': subtitle_url,
            }]

        return {
            'id': video_id,
            'duration': duration,
            'thumbnail': thumbnail,
            'formats': formats,
            'subtitles': subtitles,
        }

    def _extract_formats(self, media_info, video_id):
        type_ = media_info.get('_type')
        media_array = media_info.get('_mediaArray', [])
        formats = []
        for num, media in enumerate(media_array):
            for stream in media.get('_mediaStreamArray', []):
                stream_urls = stream.get('_stream')
                if not stream_urls:
                    continue
                if not isinstance(stream_urls, list):
                    stream_urls = [stream_urls]
                quality = stream.get('_quality')
                server = stream.get('_server')
                for stream_url in stream_urls:
                    ext = determine_ext(stream_url)
                    if quality != 'auto' and ext in ('f4m', 'm3u8'):
                        continue
                    if ext == 'f4m':
                        f4m_formats = self._extract_f4m_formats(
                            stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124',
                            video_id, preference=-1, f4m_id='hds', fatal=False)
                        if f4m_formats:
                            formats.extend(f4m_formats)
                    elif ext == 'm3u8':
                        m3u8_formats = self._extract_m3u8_formats(
                            stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False)
                        if m3u8_formats:
                            formats.extend(m3u8_formats)
                    else:
                        if server and server.startswith('rtmp'):
                            f = {
                                'url': server,
                                'play_path': stream_url,
                                'format_id': 'a%s-rtmp-%s' % (num, quality),
                            }
                        elif stream_url.startswith('http'):
                            f = {
                                'url': stream_url,
                                'format_id': 'a%s-%s-%s' % (num, ext, quality)
                            }
                        else:
                            continue
                        m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
                        if m:
                            f.update({
                                'width': int(m.group('width')),
                                'height': int(m.group('height')),
                            })
                        if type_ == 'audio':
                            f['vcodec'] = 'none'
                        formats.append(f)
        return formats

    def _real_extract(self, url):
        # determine video id from url
        m = re.match(self._VALID_URL, url)

        numid = re.search(r'documentId=([0-9]+)', url)
        if numid:
            video_id = numid.group(1)
        else:
            video_id = m.group('video_id')

        webpage = self._download_webpage(url, video_id)

        if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
            raise ExtractorError('Video %s is no longer available' % video_id, expected=True)

        if 'Diese Sendung ist für Jugendliche unter 12 Jahren nicht geeignet. Der Clip ist deshalb nur von 20 bis 6 Uhr verfügbar.' in webpage:
            raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)

        if re.search(r'[\?&]rss($|[=&])', url):
            doc = compat_etree_fromstring(webpage.encode('utf-8'))
            if doc.tag == 'rss':
                return GenericIE()._extract_rss(url, video_id, doc)

        title = self._html_search_regex(
            [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
             r'<meta name="dcterms.title" content="(.*?)"/>',
             r'<h4 class="headline">(.*?)</h4>'],
            webpage, 'title')
        description = self._html_search_meta(
            'dcterms.abstract', webpage, 'description', default=None)
        if description is None:
            description = self._html_search_meta(
                'description', webpage, 'meta description')

        # Thumbnail is sometimes not present.
        # It is in the mobile version, but that seems to use a different URL
        # structure altogether.
        thumbnail = self._og_search_thumbnail(webpage, default=None)

        media_streams = re.findall(r'''(?x)
            mediaCollection\.addMediaStream\([0-9]+,\s*[0-9]+,\s*"[^"]*",\s*
            "([^"]+)"''', webpage)

        if media_streams:
            QUALITIES = qualities(['lo', 'hi', 'hq'])
            formats = []
            for furl in set(media_streams):
                if furl.endswith('.f4m'):
                    fid = 'f4m'
                else:
                    fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
                    fid = fid_m.group(1) if fid_m else None
                formats.append({
                    'quality': QUALITIES(fid),
                    'format_id': fid,
                    'url': furl,
                })
            self._sort_formats(formats)
            info = {
                'formats': formats,
            }
        else:  # request JSON file
            info = self._extract_media_info(
                'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id)

        info.update({
            'id': video_id,
            'title': title,
            'description': description,
            'thumbnail': thumbnail,
        })

        return info


class ARDIE(InfoExtractor):
    _VALID_URL = '(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
    _TEST = {
        'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
        'md5': 'd216c3a86493f9322545e045ddc3eb35',
        'info_dict': {
            'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
            'id': '100',
            'ext': 'mp4',
            'duration': 2600,
            'title': 'Die Story im Ersten: Mission unter falscher Flagge',
            'upload_date': '20140804',
            'thumbnail': 're:^https?://.*\.jpg$',
        }
    }

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

        player_url = mobj.group('mainurl') + '~playerXml.xml'
        doc = self._download_xml(player_url, display_id)
        video_node = doc.find('./video')
        upload_date = unified_strdate(xpath_text(
            video_node, './broadcastDate'))
        thumbnail = xpath_text(video_node, './/teaserImage//variant/url')

        formats = []
        for a in video_node.findall('.//asset'):
            f = {
                'format_id': a.attrib['type'],
                'width': int_or_none(a.find('./frameWidth').text),
                'height': int_or_none(a.find('./frameHeight').text),
                'vbr': int_or_none(a.find('./bitrateVideo').text),
                'abr': int_or_none(a.find('./bitrateAudio').text),
                'vcodec': a.find('./codecVideo').text,
                'tbr': int_or_none(a.find('./totalBitrate').text),
            }
            if a.find('./serverPrefix').text:
                f['url'] = a.find('./serverPrefix').text
                f['playpath'] = a.find('./fileName').text
            else:
                f['url'] = a.find('./fileName').text
            formats.append(f)
        self._sort_formats(formats)

        return {
            'id': mobj.group('id'),
            'formats': formats,
            'display_id': display_id,
            'title': video_node.find('./title').text,
            'duration': parse_duration(video_node.find('./duration').text),
            'upload_date': upload_date,
            'thumbnail': thumbnail,
        }


class SportschauIE(ARDMediathekIE):
    IE_NAME = 'Sportschau'
    _VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html'
    _TESTS = [{
        'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html',
        'info_dict': {
            'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100',
            'ext': 'mp4',
            'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"',
            'thumbnail': 're:^https?://.*\.jpg$',
            'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.',
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }]

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        video_id = mobj.group('id')
        base_url = mobj.group('baseurl')

        webpage = self._download_webpage(url, video_id)
        title = get_element_by_attribute('class', 'headline', webpage)
        description = self._html_search_meta('description', webpage, 'description')

        info = self._extract_media_info(
            base_url + '-mc_defaultQuality-h.json', webpage, video_id)

        info.update({
            'title': title,
            'description': description,
        })

        return info
Commit	Line	Data
f9b85496 PH	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
d5822b96 PH	4	import re
	5
	6	from .common import InfoExtractor
3741302a	7	from .generic import GenericIE
d5822b96	8	from ..utils import (
f9b85496	9	determine_ext,
d5822b96	10	ExtractorError,
3bf8c316	11	get_element_by_attribute,
29546b34	12	qualities,
6d3d3fc0 PH	13	int_or_none,
	14	parse_duration,
	15	unified_strdate,
bf0ff932	16	xpath_text,
d5822b96	17	)
f7854627	18	from ..compat import compat_etree_fromstring
d5822b96	19
f9b85496	20
6d3d3fc0 PH	21	class ARDMediathekIE(InfoExtractor):
6d3d3fc0 PH	22	IE_NAME = 'ARD:mediathek'
29546b34	23	_VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de\|mediathek\.daserste\.de)/(?:./)(?P<video_id>[0-9]+\|[^0-9][^/\?]+)[^/\?](?:\?.*)?'
f9b85496	24
29546b34	25	_TESTS = [{
769efa16	26	'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
29546b34	27	'info_dict': {
769efa16	28	'id': '29582122',
29546b34	29	'ext': 'mp4',
769efa16 S	30	'title': 'Ich liebe das Leben trotzdem',
	31	'description': 'md5:45e4c225c72b27993314b31a84a5261c',
	32	'duration': 4557,
	33	},
	34	'params': {
	35	# m3u8 download
	36	'skip_download': True,
29546b34	37	},
d719c6a5 S	38	}, {
	39	'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
	40	'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
	41	'info_dict': {
	42	'id': '29522730',
	43	'ext': 'mp4',
	44	'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)',
	45	'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
	46	'duration': 5252,
	47	},
86b4e98a S	48	}, {
	49	# audio
	50	'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
	51	'md5': '219d94d8980b4f538c7fcb0865eb7f2c',
	52	'info_dict': {
	53	'id': '28488308',
	54	'ext': 'mp3',
	55	'title': 'Tod eines Fußballers',
	56	'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
	57	'duration': 3240,
	58	},
769efa16 S	59	}, {
	60	'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
	61	'only_matching': True,
29546b34	62	}]
d5822b96	63
e37c92ec S	64	def _extract_media_info(self, media_info_url, webpage, video_id):
	65	media_info = self._download_json(
	66	media_info_url, video_id, 'Downloading media JSON')
	67
	68	formats = self._extract_formats(media_info, video_id)
	69
	70	if not formats:
	71	if '"fsk"' in webpage:
	72	raise ExtractorError(
	73	'This video is only available after 20:00', expected=True)
	74	elif media_info.get('_geoblocked'):
	75	raise ExtractorError('This video is not available due to geo restriction', expected=True)
	76
	77	self._sort_formats(formats)
	78
	79	duration = int_or_none(media_info.get('_duration'))
	80	thumbnail = media_info.get('_previewImage')
	81
	82	subtitles = {}
	83	subtitle_url = media_info.get('_subtitleUrl')
	84	if subtitle_url:
	85	subtitles['de'] = [{
	86	'ext': 'srt',
	87	'url': subtitle_url,
	88	}]
	89
	90	return {
	91	'id': video_id,
	92	'duration': duration,
	93	'thumbnail': thumbnail,
	94	'formats': formats,
	95	'subtitles': subtitles,
	96	}
	97
	98	def _extract_formats(self, media_info, video_id):
	99	type_ = media_info.get('_type')
	100	media_array = media_info.get('_mediaArray', [])
	101	formats = []
	102	for num, media in enumerate(media_array):
	103	for stream in media.get('_mediaStreamArray', []):
	104	stream_urls = stream.get('_stream')
	105	if not stream_urls:
	106	continue
	107	if not isinstance(stream_urls, list):
	108	stream_urls = [stream_urls]
	109	quality = stream.get('_quality')
	110	server = stream.get('_server')
	111	for stream_url in stream_urls:
	112	ext = determine_ext(stream_url)
1fc0b47f	113	if quality != 'auto' and ext in ('f4m', 'm3u8'):
1fc0b47f	114	continue
e37c92ec	115	if ext == 'f4m':
1fc0b47f	116	f4m_formats = self._extract_f4m_formats(
e37c92ec	117	stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124',
1fc0b47f	118	video_id, preference=-1, f4m_id='hds', fatal=False)
	119	if f4m_formats:
	120	formats.extend(f4m_formats)
e37c92ec	121	elif ext == 'm3u8':
1fc0b47f	122	m3u8_formats = self._extract_m3u8_formats(
	123	stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False)
	124	if m3u8_formats:
	125	formats.extend(m3u8_formats)
e37c92ec S	126	else:
	127	if server and server.startswith('rtmp'):
	128	f = {
	129	'url': server,
	130	'play_path': stream_url,
	131	'format_id': 'a%s-rtmp-%s' % (num, quality),
	132	}
	133	elif stream_url.startswith('http'):
	134	f = {
	135	'url': stream_url,
	136	'format_id': 'a%s-%s-%s' % (num, ext, quality)
	137	}
	138	else:
	139	continue
	140	m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
	141	if m:
	142	f.update({
	143	'width': int(m.group('width')),
	144	'height': int(m.group('height')),
	145	})
	146	if type_ == 'audio':
	147	f['vcodec'] = 'none'
	148	formats.append(f)
	149	return formats
	150
d5822b96 PH	151	def _real_extract(self, url):
	152	# determine video id from url
	153	m = re.match(self._VALID_URL, url)
	154
	155	numid = re.search(r'documentId=([0-9]+)', url)
	156	if numid:
	157	video_id = numid.group(1)
	158	else:
	159	video_id = m.group('video_id')
	160
5622f29a	161	webpage = self._download_webpage(url, video_id)
f9b85496	162
3a5beb0c S	163	if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
	164	raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
	165
39aa42ff S	166	if 'Diese Sendung ist für Jugendliche unter 12 Jahren nicht geeignet. Der Clip ist deshalb nur von 20 bis 6 Uhr verfügbar.' in webpage:
	167	raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)
	168
bfd91588	169	if re.search(r'[\?&]rss($\|[=&])', url):
f7854627	170	doc = compat_etree_fromstring(webpage.encode('utf-8'))
3741302a OE	171	if doc.tag == 'rss':
	172	return GenericIE()._extract_rss(url, video_id, doc)
	173
f9b85496	174	title = self._html_search_regex(
0f97c9a0	175	[r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
6a3fa81f	176	r'<meta name="dcterms.title" content="(.*?)"/>',
0f97c9a0 PH	177	r'<h4 class="headline">(.*?)</h4>'],
0f97c9a0 PH	178	webpage, 'title')
f9b85496	179	description = self._html_search_meta(
29546b34 PH	180	'dcterms.abstract', webpage, 'description', default=None)
	181	if description is None:
	182	description = self._html_search_meta(
	183	'description', webpage, 'meta description')
	184
	185	# Thumbnail is sometimes not present.
	186	# It is in the mobile version, but that seems to use a different URL
	187	# structure altogether.
	188	thumbnail = self._og_search_thumbnail(webpage, default=None)
	189
	190	media_streams = re.findall(r'''(?x)
	191	mediaCollection\.addMediaStream\([0-9]+,\s[0-9]+,\s"[^"]",\s
	192	"([^"]+)"''', webpage)
	193
	194	if media_streams:
	195	QUALITIES = qualities(['lo', 'hi', 'hq'])
	196	formats = []
	197	for furl in set(media_streams):
	198	if furl.endswith('.f4m'):
	199	fid = 'f4m'
	200	else:
	201	fid_m = re.match(r'.*\.([^.]+)\.[^.]+$', furl)
	202	fid = fid_m.group(1) if fid_m else None
	203	formats.append({
	204	'quality': QUALITIES(fid),
	205	'format_id': fid,
	206	'url': furl,
	207	})
e37c92ec S	208	self._sort_formats(formats)
	209	info = {
	210	'formats': formats,
	211	}
29546b34	212	else: # request JSON file
e37c92ec S	213	info = self._extract_media_info(
e37c92ec S	214	'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id)
f9b85496	215
e37c92ec	216	info.update({
f9b85496 PH	217	'id': video_id,
	218	'title': title,
	219	'description': description,
f9b85496	220	'thumbnail': thumbnail,
e37c92ec S	221	})
	222
	223	return info
6d3d3fc0 PH	224
	225
	226	class ARDIE(InfoExtractor):
	227	_VALID_URL = '(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
	228	_TEST = {
	229	'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
	230	'md5': 'd216c3a86493f9322545e045ddc3eb35',
	231	'info_dict': {
	232	'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
	233	'id': '100',
	234	'ext': 'mp4',
	235	'duration': 2600,
	236	'title': 'Die Story im Ersten: Mission unter falscher Flagge',
	237	'upload_date': '20140804',
	238	'thumbnail': 're:^https?://.*\.jpg$',
	239	}
	240	}
	241
	242	def _real_extract(self, url):
	243	mobj = re.match(self._VALID_URL, url)
	244	display_id = mobj.group('display_id')
	245
	246	player_url = mobj.group('mainurl') + '~playerXml.xml'
	247	doc = self._download_xml(player_url, display_id)
	248	video_node = doc.find('./video')
bf0ff932 PH	249	upload_date = unified_strdate(xpath_text(
	250	video_node, './broadcastDate'))
	251	thumbnail = xpath_text(video_node, './/teaserImage//variant/url')
6d3d3fc0 PH	252
	253	formats = []
	254	for a in video_node.findall('.//asset'):
	255	f = {
	256	'format_id': a.attrib['type'],
	257	'width': int_or_none(a.find('./frameWidth').text),
	258	'height': int_or_none(a.find('./frameHeight').text),
	259	'vbr': int_or_none(a.find('./bitrateVideo').text),
	260	'abr': int_or_none(a.find('./bitrateAudio').text),
	261	'vcodec': a.find('./codecVideo').text,
	262	'tbr': int_or_none(a.find('./totalBitrate').text),
	263	}
	264	if a.find('./serverPrefix').text:
	265	f['url'] = a.find('./serverPrefix').text
	266	f['playpath'] = a.find('./fileName').text
	267	else:
	268	f['url'] = a.find('./fileName').text
	269	formats.append(f)
	270	self._sort_formats(formats)
	271
	272	return {
	273	'id': mobj.group('id'),
	274	'formats': formats,
	275	'display_id': display_id,
	276	'title': video_node.find('./title').text,
	277	'duration': parse_duration(video_node.find('./duration').text),
	278	'upload_date': upload_date,
	279	'thumbnail': thumbnail,
	280	}
3bf8c316 S	281
	282
	283	class SportschauIE(ARDMediathekIE):
	284	IE_NAME = 'Sportschau'
	285	_VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html'
86b4e98a	286	_TESTS = [{
3bf8c316 S	287	'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html',
	288	'info_dict': {
	289	'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100',
	290	'ext': 'mp4',
	291	'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"',
	292	'thumbnail': 're:^https?://.*\.jpg$',
	293	'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.',
	294	},
	295	'params': {
	296	# m3u8 download
	297	'skip_download': True,
	298	},
86b4e98a	299	}]
3bf8c316 S	300
	301	def _real_extract(self, url):
	302	mobj = re.match(self._VALID_URL, url)
	303	video_id = mobj.group('id')
	304	base_url = mobj.group('baseurl')
	305
	306	webpage = self._download_webpage(url, video_id)
	307	title = get_element_by_attribute('class', 'headline', webpage)
	308	description = self._html_search_meta('description', webpage, 'description')
	309
	310	info = self._extract_media_info(
	311	base_url + '-mc_defaultQuality-h.json', webpage, video_id)
	312
	313	info.update({
	314	'title': title,
	315	'description': description,
	316	})
	317
	318	return info